Soccer predictions: actual data vs predicted data

Slate 20261017 · 9 match(es)

How to read this report. Each match shows the model's headline prediction, the stage-by-stage lambda (goal-rate) build, both teams' standing/form inputs, and every starter's projected line next to their real season baseline. A moderate / large deviation badge flags a projection that strays far from the player's season form — a quick way to spot when the inputs look right but a projection does not. When the real result is known an actual-vs-predicted banner appears.

All matches overview

Match	Score	Winner	H / D / A	O2.5	BTTS	Conf	λ H/A	Actual	Result
Valencia vs Athletic Club LaLiga	0-0	Draw	33.9% / 31.3% / 34.8%	32.0%	39.8%	0.69	0.99 / 1.00	-	pending
Espanyol vs Atlético Madrid LaLiga	1-2	Atlético Madrid	21.1% / 19.5% / 59.5%	70.8%	66.6%	0.76	1.34 / 2.33	-	pending
Newcastle vs Aston Villa Premier League	1-1	Draw	43.9% / 25.8% / 30.3%	50.1%	54.0%	0.71	1.49 / 1.19	-	pending
Real Betis vs Barcelona LaLiga	1-3	Barcelona	10.2% / 11.1% / 78.6%	89.8%	75.8%	0.83	1.49 / 3.81	-	pending
Everton vs Chelsea Premier League	1-2	Chelsea	35.0% / 21.4% / 43.7%	72.7%	71.8%	0.72	1.78 / 2.00	-	pending
Villarreal vs Elche LaLiga	1-0	Villarreal	61.9% / 21.5% / 16.6%	54.1%	50.9%	0.77	1.95 / 0.90	-	pending
Fulham vs Hull Premier League	1-1	Draw	42.4% / 28.0% / 29.6%	41.5%	47.2%	0.7	1.30 / 1.04	-	pending
Man City vs Ipswich Premier League	3-1	Man City	75.2% / 13.6% / 11.2%	79.7%	66.0%	0.79	3.08 / 1.18	-	pending
Brentford vs Liverpool Premier League	1-1	Draw	35.2% / 24.5% / 40.2%	57.2%	60.0%	0.7	1.43 / 1.55	-	pending

Match detail

Valencia vs Athletic Club

LaLiga · 20261017 · 4-2-3-1 vs 4-2-3-1

Predicted 0-0Winner: DrawWinning H 1-0 (13.3%)Winning A 0-1 (13.6%)Confidence 0.69λ home/away 0.99 / 1.00Most-likely P 13.9%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Valencia vs Athletic Club. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 4-2-3-1 (duel lambdas: Home 1.008, Away 0.981). Final calibrated Poisson lambdas: Home 0.986, Away 1.002 (most likely score 0-0).

Home Starters 11/11 XGBoost (10 FBref) Away Starters 11/11 XGBoost (10 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.01 / 0.98 Final λ (H/A) 0.99 / 1.00

FBref Team Priors: Valencia (Gls/90: 0.3333333333333333, GA/90: 0.33 → E[λ]: 0.247) vs Athletic Club (Gls/90: 1.0, GA/90: 1.0 → E[λ]: 0.244) | Blend Weight: 0.228

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Valencia	Athletic Club
League position	19.00	10.00
Points per game	0.94	1.35
Attack index	0.79	0.92
Defense (concede) index	0.88	1.08
Form multiplier	0.98	1.02
Standing λ multiplier	0.87	1.00

Lambda decomposition (how the goal rates were built)

Stage	Valencia λ	Athletic Club λ
FBref Squad Prior E[λ] (weight 0.228)	0.247	0.244
Pre-competitive-floor	0.986	1.002
Pre-calibration (Micro-duels)	1.008	0.981
Final λ	0.986	1.002

competition calibration scale 1.0competition observed total goals 0.0quality gap applied Falsemismatch score pressure applied Falsecompetitive underdog floor applied Falsestage discount 1.0

Scoreline distribution & markets

Top 5 scorelines	P
0-0	13.9%
1-1	13.7%
0-1	13.6%
1-0	13.3%
0-2	6.9%

Home win	33.9%
Draw	31.3%
Away win	34.8%
Over 2.5	32.0%
BTTS yes	39.8%

Pregame fair line — favorite Athletic Club (34.8%), fair total 1.99, fair spread -0.02

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Valencia — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Hugo Duro F 74 0 0.30 0.20 53% 2 0.7 6.6 56.3 31 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0008, quality 0.531)	`market_value_eur=10700000` `quality_factor=0.5306` `boost_factor=1.0008` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0044)	`blend_weight=0.0044` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=0.8213` `prior_reliability=0.0667` `competition=La Liga`	totalShots: 1.5974 -> 1.5903
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Aymeric Laporte, Yeray Álvarez', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.192, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9878, 'skill_edge': 0.0197, 'height_edge': -0.028, 'attacker_skill': 0.5306, 'defender_skill': 0.4869, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Aymeric Laporte', 'market_value': 8400000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5365}, {'player': 'Yeray Álvarez', 'market_value': 930000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}], 'xg_override': 0.3222}	totalShots: 1.5942 -> 1.5748 shotOffTarget: 0.904 -> 0.893 onTargetScoringAttempt: 0.6902 -> 0.6818
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.3222 -> 0.3012
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 18 by projected rating, fifa_overall 77.0)	`match_squad_rank=18` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.2601` `goal_rate=0.2591` `scorer_weight=0.2501`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0283` `assist_rate=0.1289` `provider_weight=0.0359`	-

Dani Raba F 63 0 0.13 0.02 462% 2 0.2 6.8 9.0 30 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9949, quality 0.437)	`market_value_eur=1000000` `quality_factor=0.4374` `boost_factor=0.9949` `reference_quality=0.505` `history_matches=15` `history_shrink=0.25`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0011)	`blend_weight=0.0011` `history_matches=15` `season_overlap_discount=1.0` `per90_to_match_factor=0.6983` `prior_reliability=0.0067` `competition=La Liga`	totalCross: 1.0816 -> 1.088 totalShots: 1.5428 -> 1.5487
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Aymeric Laporte, Yeray Álvarez', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9496, 'skill_edge': -0.0223, 'height_edge': -0.012, 'attacker_skill': 0.4374, 'defender_skill': 0.4869, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Aymeric Laporte', 'market_value': 8400000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5365}, {'player': 'Yeray Álvarez', 'market_value': 930000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}], 'xg_override': 0.1413}	totalShots: 1.5487 -> 1.4706 shotOffTarget: 0.9485 -> 0.9007 onTargetScoringAttempt: 0.5937 -> 0.5638 bigChanceMissed: 0.3122 -> 0.2965 expectedGoalsOnTarget: 0.2863 -> 0.2719 goals: 0.1807 -> 0.1716 expectedGoals: 0.1488 -> 0.1413
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1413 -> 0.1321 expected_assists: 0.0658 -> 0.0721
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 74.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.1237` `goal_rate=0.1716` `scorer_weight=0.153`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0696` `assist_rate=0.1915` `provider_weight=0.0901`	-

Javier Guerra M 82 0 0.16 0.10 53% 2 1.1 6.9 58.3 34 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0055, quality 0.686)	`market_value_eur=26000000` `quality_factor=0.6858` `boost_factor=1.0055` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0119)	`blend_weight=0.0119` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=0.9077` `prior_reliability=0.18` `competition=La Liga`	totalShots: 1.8708 -> 1.8607 totalCross: 1.615 -> 1.6238 onTargetScoringAttempt: 0.6002 -> 0.593 fouls: 0.5468 -> 0.5403
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Jesús Areso', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.581, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.1485, 'skill_edge': 0.085, 'height_edge': 0.016, 'attacker_skill': 0.6858, 'defender_skill': 0.497, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Jesús Areso', 'market_value': 5500000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.497}], 'xg_override': 0.1689}	aerialWon: 0.6784 -> 1.2 totalShots: 1.8635 -> 2.1403 shotOffTarget: 1.2705 -> 1.4592 onTargetScoringAttempt: 0.593 -> 0.6811 bigChanceMissed: 0.1913 -> 0.2197 goals: 0.1787 -> 0.2052 expectedGoals: 0.1471 -> 0.1689 expectedGoalsOnTarget: 0.0816 -> 0.0937
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1689 -> 0.158 expected_assists: 0.1191 -> 0.1305
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 82.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.1462` `goal_rate=0.2052` `scorer_weight=0.1842`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.1223` `assist_rate=0.1203` `provider_weight=0.1314`	-

Guido Rodríguez M 85 0 0.05 0.04 34% 1 0.6 7.0 58.7 27 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9972, quality 0.445)	`market_value_eur=3600000` `quality_factor=0.4449` `boost_factor=0.9972` `reference_quality=0.505` `history_matches=27` `history_shrink=0.1562`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0135)	`blend_weight=0.0135` `history_matches=27` `season_overlap_discount=1.0` `per90_to_match_factor=0.9402` `prior_reliability=0.1333` `competition=La Liga`	interceptionWon: 1.1908 -> 1.2065 wonTackle: 1.988 -> 1.9802
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Iñigo Ruiz de Galarreta', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.471, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.0748, 'skill_edge': 0.0069, 'height_edge': 0.044, 'attacker_skill': 0.4527, 'defender_skill': 0.4374, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Iñigo Ruiz de Galarreta', 'market_value': 1800000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4374}], 'xg_override': 0.0571}	aerialWon: 1.1431 -> 1.2 totalShots: 0.7055 -> 0.7583 shotOffTarget: 0.5287 -> 0.5682 onTargetScoringAttempt: 0.1768 -> 0.19 goals: 0.0692 -> 0.0744
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 78.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.052` `goal_rate=0.0744` `scorer_weight=0.062`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0385` `assist_rate=0.0549` `provider_weight=0.048`	-

Pepelu M 83 0 0.03 0.08 57% 1 0.4 6.9 50.8 39 XGBoost M_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9995, quality 0.490)	`market_value_eur=7300000` `quality_factor=0.4903` `boost_factor=0.9995` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Iñigo Ruiz de Galarreta', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.495, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.1099, 'skill_edge': 0.0307, 'height_edge': 0.044, 'attacker_skill': 0.5057, 'defender_skill': 0.4374, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Iñigo Ruiz de Galarreta', 'market_value': 1800000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4374}], 'xg_override': 0.0368}	aerialWon: 0.4863 -> 1.2 totalShots: 0.526 -> 0.5838 shotOffTarget: 0.4094 -> 0.4544 onTargetScoringAttempt: 0.1166 -> 0.1294
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 76.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0338` `goal_rate=0.0437` `scorer_weight=0.0371`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0422` `assist_rate=0.0547` `provider_weight=0.0505`	-

José Luis Gayà D 82 0 0.01 0.02 35% 1 0.3 6.7 54.2 29 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9982, quality 0.454)	`market_value_eur=4300000` `quality_factor=0.4535` `boost_factor=0.9982` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0040)	`blend_weight=0.004` `history_matches=38` `season_overlap_discount=1.0` `per90_to_match_factor=0.908` `prior_reliability=0.0533` `competition=La Liga`	wasFouled: 0.7249 -> 0.7311 totalCross: 2.4743 -> 2.4689
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Álex Berenguer', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0578 -> 0.0633
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 80.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0137` `goal_rate=0.0063` `scorer_weight=0.012`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0613` `assist_rate=0.0433` `provider_weight=0.0548`	-

Justin De Haas D 83 0 0.00 0.00 101% 0 0.1 6.8 34.4 8 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9983, quality 0.495)	`market_value_eur=7700000` `quality_factor=0.4951` `boost_factor=0.9983` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0120)	`blend_weight=0.012` `history_matches=4` `season_overlap_discount=1.0` `per90_to_match_factor=0.9223` `prior_reliability=0.0333` `competition=La Liga`	fouls: 0.685 -> 0.6989 interceptionWon: 0.9194 -> 0.9083 wonTackle: 0.6472 -> 0.6394 totalShots: 0.444 -> 0.4387
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Iñaki Williams', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 76.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0044` `goal_rate=0.0` `scorer_weight=0.004`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0173` `assist_rate=0.3095` `provider_weight=0.0219`	-

Mouctar Diakhaby D 73 0 0.03 0.01 180% 0 0.2 6.8 41.2 9 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9944, quality 0.437)	`market_value_eur=1500000` `quality_factor=0.4374` `boost_factor=0.9944` `reference_quality=0.505` `history_matches=13` `history_shrink=0.2778`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0301)	`blend_weight=0.0301` `history_matches=13` `season_overlap_discount=1.0` `per90_to_match_factor=0.8106` `prior_reliability=0.1667` `competition=La Liga`	interceptionWon: 1.0644 -> 1.0812 wonTackle: 0.6524 -> 0.6425 fouls: 0.6461 -> 0.6364 totalShots: 0.481 -> 0.486
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Iñaki Williams', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 76.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0304` `goal_rate=0.0342` `scorer_weight=0.0313`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0081` `assist_rate=0.0` `provider_weight=0.0071`	-

Pablo Maffeo D 82 0 0.01 0.01 ok 0 0.2 6.6 58.3 29 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9976, quality 0.437)	`market_value_eur=2400000` `quality_factor=0.4374` `boost_factor=0.9976` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0040)	`blend_weight=0.004` `history_matches=38` `season_overlap_discount=1.0` `per90_to_match_factor=0.9117` `prior_reliability=0.0533` `competition=La Liga`	totalCross: 3.1888 -> 3.1989
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Peio Canales', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0643 -> 0.0705
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 17 by projected rating, fifa_overall 82.0)	`match_squad_rank=17` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0121` `goal_rate=0.0012` `scorer_weight=0.0099`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0681` `assist_rate=0.0876` `provider_weight=0.0849`	-

Stole Dimitrievski G 90 0 0.00 - n/a 0.0 6.9 63.1 31 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9969, quality 0.437)	`market_value_eur=2100000` `quality_factor=0.4374` `boost_factor=0.9969` `reference_quality=0.505` `history_matches=28` `history_shrink=0.1515`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0197)	`blend_weight=0.0197` `history_matches=28` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.2` `competition=La Liga`	saves: 3.1346 -> 3.1254 wasFouled: 0.3306 -> 0.3241
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Iñaki Williams, Robert Navarro', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 0.9807}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	shots_faced: 9.3 -> 9.5 expected_saves: 2.27 -> 2.32 high_pressure_saves: 0.78 -> 0.8
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 77.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.0105`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.6588`	-

David Otorbi F 82 0 0.14 0.04 240% 1 0.4 6.9 32.8 19 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9920, quality 0.441)	`market_value_eur=3300000` `quality_factor=0.4412` `boost_factor=0.992` `reference_quality=0.505` `history_matches=7` `history_shrink=0.4167`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0072)	`blend_weight=0.0072` `history_matches=7` `season_overlap_discount=1.0` `per90_to_match_factor=0.9087` `prior_reliability=0.0267` `competition=La Liga`	totalCross: 1.9873 -> 2.0715 wasFouled: 0.3551 -> 0.369 wonTackle: 0.646 -> 0.6578 totalShots: 1.3298 -> 1.3202
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Aymeric Laporte, Yeray Álvarez', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9286, 'skill_edge': -0.0206, 'height_edge': -0.028, 'attacker_skill': 0.4412, 'defender_skill': 0.4869, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Aymeric Laporte', 'market_value': 8400000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5365}, {'player': 'Yeray Álvarez', 'market_value': 930000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}], 'xg_override': 0.1498}	totalShots: 1.3266 -> 1.2319 shotOffTarget: 0.8815 -> 0.8186 onTargetScoringAttempt: 0.4451 -> 0.4133 goals: 0.1964 -> 0.1824 expectedGoals: 0.1613 -> 0.1498 expectedGoalsOnTarget: 0.1157 -> 0.1074
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.99 (squad xG sum 1.0543, xG scale 0.9352, xA scale 1.0957, team xA 0.64 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=0.986` `squad_xg_sum=1.0543` `squad_xa_sum=0.5849` `team_expected_assists=0.6409` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1498 -> 0.1401 expected_assists: 0.0894 -> 0.098
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 64.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.1307` `goal_rate=0.1824` `scorer_weight=0.163`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0934` `assist_rate=0.0` `provider_weight=0.0784`	-

Team totals 0 0.99 12 6.8

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Kayne van Oevelen	G	4.5m	11.1%	Market-value substitute prior
Arnau Martínez	D	13.9m	22.5%	Market-value substitute prior
César Tárrega	D	13.9m	22.3%	Market-value substitute prior
Harvey Elliott	M	22.8m	53.5%	Market-value substitute prior
Filip Ugrinić	M	29.8m	43.8%	Market-value substitute prior
Diego López	F	25.6m	57.3%	Market-value substitute prior

Athletic Club — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Robert Navarro F 73 0 0.18 0.12 47% 2 0.8 7.0 50.5 38 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9996, quality 0.493)	`market_value_eur=7500000` `quality_factor=0.4927` `boost_factor=0.9996` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Justin De Haas, Mouctar Diakhaby', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9251, 'skill_edge': 0.005, 'height_edge': -0.056, 'attacker_skill': 0.4927, 'defender_skill': 0.4815, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Justin De Haas', 'market_value': 7700000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5256}, {'player': 'Mouctar Diakhaby', 'market_value': 1500000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}], 'xg_override': 0.187}	totalShots: 1.9585 -> 1.8119 shotOffTarget: 1.305 -> 1.2073 onTargetScoringAttempt: 0.6535 -> 0.6046 expectedGoalsOnTarget: 0.3527 -> 0.3263 goals: 0.2738 -> 0.2533 expectedGoals: 0.2021 -> 0.187 bigChanceMissed: 0.1668 -> 0.1543
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.187 -> 0.1801 expected_assists: 0.1583 -> 0.1332
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 76.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.1648` `goal_rate=0.2533` `scorer_weight=0.2251`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.1247` `assist_rate=0.0` `provider_weight=0.1066`	-

Iñigo Ruiz de Galarreta M 72 0 0.05 0.02 128% 1 0.6 7.1 55.5 29 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9977, quality 0.437)	`market_value_eur=1800000` `quality_factor=0.4374` `boost_factor=0.9977` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0039)	`blend_weight=0.0039` `history_matches=39` `season_overlap_discount=1.0` `per90_to_match_factor=0.795` `prior_reliability=0.0533` `competition=La Liga`	fouls: 1.7392 -> 1.748
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Guido Rodríguez', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.369, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 0.9252, 'skill_edge': -0.0069, 'height_edge': -0.044, 'attacker_skill': 0.4374, 'defender_skill': 0.4527, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Guido Rodríguez', 'market_value': 3600000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4527}], 'xg_override': 0.0474}	aerialWon: 0.984 -> 0.9 totalShots: 0.9908 -> 0.9167 shotOffTarget: 0.7397 -> 0.6844 onTargetScoringAttempt: 0.2501 -> 0.2314
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1648 -> 0.1387
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 82.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0446` `goal_rate=0.0464` `scorer_weight=0.0451`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.1295` `assist_rate=0.1` `provider_weight=0.111`	-

Beñat Gerenabarrena M 88 0 0.03 0.00 807% 0 0.1 6.8 32.9 10 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9939, quality 0.465)	`market_value_eur=5200000` `quality_factor=0.4646` `boost_factor=0.9939` `reference_quality=0.505` `history_matches=5` `history_shrink=0.5`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0455)	`blend_weight=0.0455` `history_matches=5` `season_overlap_discount=1.0` `per90_to_match_factor=0.9752` `prior_reliability=0.14` `competition=La Liga`	totalCross: 1.2566 -> 1.3262 wasFouled: 1.1854 -> 1.216 interceptionWon: 2.5892 -> 2.6193 fouls: 2.656 -> 2.6831 wonTackle: 1.8674 -> 1.8881 onTargetScoringAttempt: 0.1349 -> 0.1496
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Guido Rodríguez', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.382, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 0.9447, 'skill_edge': 0.0104, 'height_edge': -0.048, 'attacker_skill': 0.4759, 'defender_skill': 0.4527, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Guido Rodríguez', 'market_value': 3600000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4527}], 'xg_override': 0.0294}	aerialWon: 0.9541 -> 0.9 totalShots: 0.522 -> 0.4931 shotOffTarget: 0.3724 -> 0.3518 onTargetScoringAttempt: 0.1496 -> 0.1413
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.0476 -> 0.0401
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 80.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0279` `goal_rate=0.0383` `scorer_weight=0.0315`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0393` `assist_rate=0.0699` `provider_weight=0.0493`	-

Yuri Berchiche D 82 0 0.04 0.02 89% 1 1.0 7.1 55.8 26 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9976, quality 0.437)	`market_value_eur=910000` `quality_factor=0.4374` `boost_factor=0.9976` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0057)	`blend_weight=0.0057` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.9063` `prior_reliability=0.0733` `competition=La Liga`	totalShots: 1.252 -> 1.2588 wonTackle: 1.0452 -> 1.0393
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Dani Raba', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1701 -> 0.1432
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 79.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0395` `goal_rate=0.0` `scorer_weight=0.0322`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.1334` `assist_rate=0.1756` `provider_weight=0.1468`	-

Aymeric Laporte D 87 0 0.06 0.07 ok 0 0.5 7.0 54.7 25 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9999, quality 0.503)	`market_value_eur=8400000` `quality_factor=0.5035` `boost_factor=0.9999` `reference_quality=0.505` `history_matches=31` `history_shrink=0.1389`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0144)	`blend_weight=0.0144` `history_matches=31` `season_overlap_discount=1.0` `per90_to_match_factor=0.9676` `prior_reliability=0.16` `competition=La Liga`	wonTackle: 0.8362 -> 0.8416
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Hugo Duro', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 84.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0613` `goal_rate=0.0` `scorer_weight=0.0506`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.011` `assist_rate=0.0041` `provider_weight=0.0095`	-

Yeray Álvarez D 79 0 0.03 0.01 88% 1 0.2 6.8 38.4 19 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9937, quality 0.437)	`market_value_eur=930000` `quality_factor=0.4374` `boost_factor=0.9937` `reference_quality=0.505` `history_matches=11` `history_shrink=0.3125`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0271)	`blend_weight=0.0271` `history_matches=11` `season_overlap_discount=1.0` `per90_to_match_factor=0.8828` `prior_reliability=0.1333` `competition=La Liga`	totalCross: 0.6179 -> 0.649 totalShots: 0.5172 -> 0.539 wonTackle: 0.6709 -> 0.6527 wasFouled: 0.809 -> 0.799 goalAssist: 0.114 -> 0.1229 interceptionWon: 1.0644 -> 1.0714
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Hugo Duro', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 13 by projected rating, fifa_overall 77.0)	`match_squad_rank=13` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0267` `goal_rate=0.0` `scorer_weight=0.0217`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0138` `assist_rate=0.1229` `provider_weight=0.0174`	-

Jesús Areso D 79 0 0.01 0.01 77% 0 0.1 6.7 36.4 29 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9985, quality 0.468)	`market_value_eur=5500000` `quality_factor=0.4683` `boost_factor=0.9985` `reference_quality=0.505` `history_matches=31` `history_shrink=0.1389`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0096)	`blend_weight=0.0096` `history_matches=31` `season_overlap_discount=1.0` `per90_to_match_factor=0.8791` `prior_reliability=0.1067` `competition=La Liga`	wasFouled: 0.5308 -> 0.5416 interceptionWon: 1.1785 -> 1.1724 wonTackle: 1.0826 -> 1.0775
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Javier Guerra', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.0499 -> 0.042
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 17 by projected rating, fifa_overall 80.0)	`match_squad_rank=17` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0106` `goal_rate=0.0` `scorer_weight=0.0088`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0411` `assist_rate=0.0223` `provider_weight=0.0336`	-

Unai Simón G 90 0 0.00 - n/a 0.0 7.0 57.3 30 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0042, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0042` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0135)	`blend_weight=0.0135` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.2` `competition=La Liga`	saves: 2.9236 -> 2.9382
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Hugo Duro, Javier Guerra', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.0083}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	shots_faced: 9.6 -> 9.4 expected_saves: 2.35 -> 2.3 high_pressure_saves: 0.81 -> 0.79
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 85.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.0881`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.7312`	-

Peio Canales M 88 0 0.09 0.02 244% 1 0.5 6.9 33.2 8 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0023, quality 0.519)	`market_value_eur=9700000` `quality_factor=0.5189` `boost_factor=1.0023` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0554)	`blend_weight=0.0554` `history_matches=4` `season_overlap_discount=1.0` `per90_to_match_factor=0.9778` `prior_reliability=0.1533` `competition=La Liga`	wasFouled: 1.1142 -> 1.1467 totalCross: 1.1554 -> 1.1856 onTargetScoringAttempt: 0.3554 -> 0.3828 wonTackle: 0.4545 -> 0.4764 totalShots: 1.0404 -> 1.0537 fouls: 0.1931 -> 0.2059 interceptionWon: 0.2272 -> 0.2382 goals: 0.0954 -> 0.0901
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Pablo Maffeo', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.522, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.0622, 'skill_edge': 0.0343, 'height_edge': 0.008, 'attacker_skill': 0.5189, 'defender_skill': 0.4428, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Pablo Maffeo', 'market_value': 2400000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.4428}], 'xg_override': 0.0886}	aerialWon: 0.3145 -> 0.9 totalShots: 1.0678 -> 1.1342 shotOffTarget: 0.685 -> 0.7276 onTargetScoringAttempt: 0.3828 -> 0.4066 goals: 0.0901 -> 0.0957 expectedGoals: 0.0834 -> 0.0886
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1047 -> 0.0881
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 73.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0818` `goal_rate=0.0957` `scorer_weight=0.0897`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0843` `assist_rate=0.0824` `provider_weight=0.0754`	-

Álex Berenguer F 83 0 0.13 0.07 69% 1 1.0 6.8 50.7 30 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9981, quality 0.451)	`market_value_eur=4099999` `quality_factor=0.4511` `boost_factor=0.9981` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0119)	`blend_weight=0.0119` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.9273` `prior_reliability=0.1533` `competition=La Liga`	totalShots: 1.4308 -> 1.4373
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Justin De Haas, Mouctar Diakhaby', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.8917, 'skill_edge': -0.0137, 'height_edge': -0.06, 'attacker_skill': 0.4511, 'defender_skill': 0.4815, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Justin De Haas', 'market_value': 7700000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5256}, {'player': 'Mouctar Diakhaby', 'market_value': 1500000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}], 'xg_override': 0.1309}	totalShots: 1.4373 -> 1.2816 shotOffTarget: 0.9185 -> 0.819 onTargetScoringAttempt: 0.5156 -> 0.4598 expectedGoalsOnTarget: 0.2224 -> 0.1983 goals: 0.1775 -> 0.1583 bigChanceMissed: 0.1489 -> 0.1328 expectedGoals: 0.1468 -> 0.1309
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1104 -> 0.0929
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 81.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.1184` `goal_rate=0.1583` `scorer_weight=0.1451`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0887` `assist_rate=0.0223` `provider_weight=0.0743`	-

Iñaki Williams F 71 0 0.34 0.20 72% 2 1.5 6.8 52.4 34 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9995, quality 0.490)	`market_value_eur=7300000` `quality_factor=0.4903` `boost_factor=0.9995` `reference_quality=0.505` `history_matches=36` `history_shrink=0.122`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0106)	`blend_weight=0.0106` `history_matches=36` `season_overlap_discount=1.0` `per90_to_match_factor=0.7911` `prior_reliability=0.1333` `competition=La Liga`	onTargetScoringAttempt: 0.9339 -> 0.9408 totalShots: 2.2335 -> 2.2266 fouls: 1.0307 -> 1.024 wasFouled: 0.6691 -> 0.6746 totalCross: 1.2806 -> 1.2754
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Justin De Haas, Mouctar Diakhaby', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9705, 'skill_edge': 0.0039, 'height_edge': -0.024, 'attacker_skill': 0.4903, 'defender_skill': 0.4815, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Justin De Haas', 'market_value': 7700000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5256}, {'player': 'Mouctar Diakhaby', 'market_value': 1500000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}], 'xg_override': 0.3484}	totalShots: 2.2403 -> 2.1742 shotOffTarget: 1.2995 -> 1.2612 onTargetScoringAttempt: 0.9408 -> 0.913 bigChanceMissed: 0.4186 -> 0.4063 goals: 0.398 -> 0.3863 expectedGoals: 0.359 -> 0.3484 expectedGoalsOnTarget: 0.2817 -> 0.2734
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.00 (squad xG sum 1.0407, xG scale 0.9628, xA scale 0.8417, team xA 0.80 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.002` `squad_xg_sum=1.0407` `squad_xa_sum=0.9524` `team_expected_assists=0.8016` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.3484 -> 0.3355 expected_assists: 0.0628 -> 0.0529
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 80.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.285` `goal_rate=0.3863` `scorer_weight=0.3674`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0515` `assist_rate=0.3586` `provider_weight=0.0661`	-

Team totals 0 1.00 10 6.9

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Alex Padilla	G	4.4m	10.9%	Market-value substitute prior
Dani Vivian	D	14.3m	33.0%	Market-value substitute prior
Aitor Paredes	D	13.8m	22.1%	Market-value substitute prior
Mikel Jauregizar	M	24.7m	52.3%	Market-value substitute prior
Oihan Sancet	M	22.9m	32.6%	Market-value substitute prior
Nico Williams	F	25.8m	62.0%	Market-value substitute prior

Antagonist / tactical summary

Midfield is evenly contested — Javier Guerra vs Peio Canales will be the pivotal duel. Total antagonist xG — Home: 0.88, Away: 0.83. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Midfield — possession 50.7% / 49.3%, dominance ratio 0.53, key duel: Javier Guerra vs Peio Canales

Javier Guerra xG 0.17Hugo Duro xG 0.32Dani Raba xG 0.14Peio Canales xG 0.09Iñaki Williams xG 0.35Robert Navarro xG 0.19

Summary

Antagonist matchup analysis for Valencia vs Athletic Club. Most likely scoreline: 0-0. Lambdas derived from 6 home and 6 away individual player duels. Head-to-head: Valencia 3W-1D-6L vs Athletic Club over 10 meetings.

Tactical parity between 4-2-3-1 and 4-2-3-1. Match outcome driven primarily by individual player quality. Midfield is evenly contested — Javier Guerra vs Peio Canales will be the pivotal duel. Total antagonist xG — Home: 0.88, Away: 0.83. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Espanyol vs Atlético Madrid

LaLiga · 20261017 · 4-2-3-1 vs 4-2-3-1

Predicted 1-2Winner: Atlético MadridConfidence 0.76λ home/away 1.34 / 2.33Most-likely P 9.3%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Espanyol vs Atlético Madrid. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 4-2-3-1 (duel lambdas: Home 1.472, Away 2.175). Final calibrated Poisson lambdas: Home 1.337, Away 2.325 (most likely score 1-2).

Home Starters 11/11 XGBoost (7 FBref) Away Starters 11/11 XGBoost (8 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.47 / 2.17 Final λ (H/A) 1.34 / 2.33

FBref Team Priors: Espanyol (Gls/90: 1.6666666666666667, GA/90: 1.67 → E[λ]: 2.877) vs Atlético Madrid (Gls/90: 2.333333333333333, GA/90: 2.33 → E[λ]: 2.886) | Blend Weight: 0.228

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Espanyol	Atlético Madrid
League position	12.00	2.00
Points per game	1.16	1.94
Attack index	0.96	1.18
Defense (concede) index	0.86	0.96
Form multiplier	0.96	1.04
Standing λ multiplier	0.92	1.06

Lambda decomposition (how the goal rates were built)

Stage	Espanyol λ	Atlético Madrid λ
FBref Squad Prior E[λ] (weight 0.228)	2.877	2.886
Pre-mismatch	1.440	2.221
Pre-competitive-floor	1.402	2.437
Pre-calibration (Micro-duels)	1.472	2.175
Final λ	1.337	2.325

competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Truecompetitive underdog floor applied Falsestage discount 1.0

Scoreline distribution & markets

Top 5 scorelines	P
1-2	9.3%
1-1	8.0%
1-3	7.2%
0-2	6.9%
2-2	6.2%

Home win	21.1%
Draw	19.5%
Away win	59.5%
Over 2.5	70.8%
BTTS yes	66.6%

Pregame fair line — favorite Atlético Madrid (59.5%), fair total 3.66, fair spread -0.99

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Espanyol — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Roberto Fernández F 84 1 0.47 0.27 72% 2 1.4 6.8 60.9 32 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0040, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.004` `reference_quality=0.505` `history_matches=46` `history_shrink=0.098`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0123)	`blend_weight=0.0123` `history_matches=46` `season_overlap_discount=1.0` `per90_to_match_factor=0.9283` `prior_reliability=0.1933` `competition=La Liga`	totalOffside: 1.2053 -> 1.1905 fouls: 1.6304 -> 1.6221 goals: 0.4367 -> 0.4432
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Robin Le Normand, Cristian Romero', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 0.9341, 'skill_edge': -0.0448, 'height_edge': 0.0, 'attacker_skill': 0.6408, 'defender_skill': 0.7403, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Robin Le Normand', 'market_value': 17200000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.6589}, {'player': 'Cristian Romero', 'market_value': 37000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8218}], 'xg_override': 0.3744}	totalShots: 2.3003 -> 2.1487 onTargetScoringAttempt: 1.1595 -> 1.0831 shotOffTarget: 1.1408 -> 1.0656 bigChanceMissed: 0.727 -> 0.6791 expectedGoalsOnTarget: 0.4821 -> 0.4503 goals: 0.4432 -> 0.414 expectedGoals: 0.4008 -> 0.3744
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.3744 -> 0.467 expected_assists: 0.0423 -> 0.0475
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 74.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.3731` `goal_rate=0.414` `scorer_weight=0.5106`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0464` `assist_rate=0.0303` `provider_weight=0.0413`	-

Bryan Zaragoza F 76 0 0.12 0.13 ok 1 1.1 7.1 57.4 14 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9994, quality 0.493)	`market_value_eur=7500000` `quality_factor=0.4927` `boost_factor=0.9994` `reference_quality=0.505` `history_matches=24` `history_shrink=0.1724`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Robin Le Normand, Cristian Romero', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 0.748, 'skill_edge': -0.1115, 'height_edge': -0.06, 'attacker_skill': 0.4927, 'defender_skill': 0.7403, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Robin Le Normand', 'market_value': 17200000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.6589}, {'player': 'Cristian Romero', 'market_value': 37000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8218}], 'xg_override': 0.0969}	totalShots: 1.6283 -> 1.218 shotOffTarget: 1.0928 -> 0.8174 onTargetScoringAttempt: 0.5354 -> 0.4005 expectedGoalsOnTarget: 0.2317 -> 0.1733 goals: 0.1711 -> 0.128 expectedGoals: 0.1295 -> 0.0969
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0969 -> 0.1209 expected_assists: 0.1507 -> 0.1694
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 77.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1139` `goal_rate=0.128` `scorer_weight=0.1465`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1558` `assist_rate=0.0527` `provider_weight=0.1355`	-

Javier Hernández M 78 0 0.15 0.07 107% 2 1.0 6.8 49.6 12 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9970, quality 0.479)	`market_value_eur=6400000` `quality_factor=0.4793` `boost_factor=0.997` `reference_quality=0.505` `history_matches=8` `history_shrink=0.3846`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0417)	`blend_weight=0.0417` `history_matches=8` `season_overlap_discount=1.0` `per90_to_match_factor=0.8669` `prior_reliability=0.1667` `competition=La Liga`	totalCross: 2.0181 -> 1.9484 fouls: 1.8982 -> 1.9347 wasFouled: 1.6031 -> 1.6374 goalAssist: 0.2186 -> 0.2528 totalShots: 1.7569 -> 1.7274 onTargetScoringAttempt: 0.6749 -> 0.6612 goals: 0.189 -> 0.1811 interceptionWon: 0.1262 -> 0.1209
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Jan Oblak', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.26, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 0.7646, 'skill_edge': -0.1041, 'height_edge': -0.056, 'attacker_skill': 0.4793, 'defender_skill': 0.7107, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Jan Oblak', 'market_value': 16000000, 'role': 'GK', 'price_index': 0.55, 'adjusted_quality': 0.7107}], 'xg_override': 0.1203}	aerialWon: 0.2056 -> 0.9 totalShots: 1.7431 -> 1.3328 shotOffTarget: 1.0819 -> 0.8272 onTargetScoringAttempt: 0.6612 -> 0.5056 goals: 0.1811 -> 0.1385 expectedGoals: 0.1574 -> 0.1203 expectedGoalsOnTarget: 0.1304 -> 0.0997
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.1203 -> 0.15 expected_assists: 0.1375 -> 0.1544
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 86.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1393` `goal_rate=0.1385` `scorer_weight=0.166`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1431` `assist_rate=0.2528` `provider_weight=0.193`	-

Tyrhys Dolan F 79 0 0.11 0.05 123% 1 0.7 6.7 54.1 41 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0008, quality 0.532)	`market_value_eur=10800000` `quality_factor=0.5318` `boost_factor=1.0008` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0126)	`blend_weight=0.0126` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.8801` `prior_reliability=0.1867` `competition=La Liga`	interceptionWon: 0.8672 -> 0.8761
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Robin Le Normand, Cristian Romero', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 0.7739, 'skill_edge': -0.0938, 'height_edge': -0.06, 'attacker_skill': 0.5318, 'defender_skill': 0.7403, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Robin Le Normand', 'market_value': 17200000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.6589}, {'player': 'Cristian Romero', 'market_value': 37000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8218}], 'xg_override': 0.0895}	totalShots: 1.2703 -> 0.9831 shotOffTarget: 0.8314 -> 0.6434 onTargetScoringAttempt: 0.4376 -> 0.3387 goals: 0.1395 -> 0.108 expectedGoals: 0.1157 -> 0.0895 expectedGoalsOnTarget: 0.099 -> 0.0766
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0895 -> 0.1116 expected_assists: 0.0892 -> 0.1002
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 74.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1056` `goal_rate=0.108` `scorer_weight=0.1264`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0953` `assist_rate=0.083` `provider_weight=0.0957`	-

Pol Lozano M 76 0 0.04 0.02 98% 0 0.4 6.8 46.0 29 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9988, quality 0.472)	`market_value_eur=5800000` `quality_factor=0.472` `boost_factor=0.9988` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0021)	`blend_weight=0.0021` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.8396` `prior_reliability=0.0267` `competition=La Liga`	wonTackle: 0.977 -> 0.9837
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Giuliano Simeone', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.276, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.789, 'skill_edge': -0.1555, 'height_edge': 0.012, 'attacker_skill': 0.4845, 'defender_skill': 0.8301, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Giuliano Simeone', 'market_value': 44000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.8301}], 'xg_override': 0.036}	aerialWon: 0.458 -> 0.9 totalShots: 0.6119 -> 0.4828 shotOffTarget: 0.4221 -> 0.333 onTargetScoringAttempt: 0.1862 -> 0.1469 expectedGoalsOnTarget: 0.0793 -> 0.0626 goals: 0.0555 -> 0.0438 expectedGoals: 0.0456 -> 0.036
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.036 -> 0.0449 expected_assists: 0.0627 -> 0.0704
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 75.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0439` `goal_rate=0.0438` `scorer_weight=0.049`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.068` `assist_rate=0.0596` `provider_weight=0.0685`	-

Edu Expósito M 83 0 0.12 0.08 47% 1 1.4 7.2 63.1 31 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9985, quality 0.456)	`market_value_eur=4500000` `quality_factor=0.456` `boost_factor=0.9985` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0119)	`blend_weight=0.0119` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=0.923` `prior_reliability=0.18` `competition=La Liga`	totalCross: 3.9074 -> 3.873 totalShots: 1.3963 -> 1.4167
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Giuliano Simeone', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.264, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.7708, 'skill_edge': -0.1639, 'height_edge': 0.008, 'attacker_skill': 0.4658, 'defender_skill': 0.8301, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Giuliano Simeone', 'market_value': 44000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.8301}], 'xg_override': 0.0988}	totalShots: 1.4167 -> 1.092 shotOffTarget: 0.8781 -> 0.6768 onTargetScoringAttempt: 0.5161 -> 0.3978 goals: 0.1416 -> 0.1091 expectedGoals: 0.1282 -> 0.0988 aerialWon: 0.8886 -> 0.9
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0988 -> 0.1232 expected_assists: 0.1469 -> 0.165
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 77.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1159` `goal_rate=0.1091` `scorer_weight=0.1317`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1521` `assist_rate=0.1029` `provider_weight=0.132`	-

Quilindschy Hartman D 86 0 0.02 0.01 227% 0 0.1 6.7 29.5 25 FBref Prior D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0018, quality 0.541)	`market_value_eur=11600000` `quality_factor=0.5411` `boost_factor=1.0018` `reference_quality=0.505` `history_matches=25` `history_shrink=0.1667`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Jonathan David', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.0847 -> 0.0951
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 13 by projected rating, fifa_overall 75.0)	`match_squad_rank=13` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0193` `goal_rate=0.0` `scorer_weight=0.0156`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0907` `assist_rate=0.1331` `provider_weight=0.1189`	-

Leandro Cabrera D 90 0 0.05 0.06 ok 1 0.9 6.8 59.5 29 XGBoost D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9978, quality 0.437)	`market_value_eur=930000` `quality_factor=0.4374` `boost_factor=0.9978` `reference_quality=0.505` `history_matches=42` `history_shrink=0.1064`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Jonathan David', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0432 -> 0.0539
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 76.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0525` `goal_rate=0.0085` `scorer_weight=0.0431`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0372` `assist_rate=0.0` `provider_weight=0.0303`	-

Clemens Riedel D 86 0 0.01 0.00 454% 0 0.1 6.8 40.6 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9974, quality 0.449)	`market_value_eur=3900000` `quality_factor=0.4486` `boost_factor=0.9974` `reference_quality=0.505` `history_matches=27` `history_shrink=0.1562`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0115)	`blend_weight=0.0115` `history_matches=27` `season_overlap_discount=1.0` `per90_to_match_factor=0.9501` `prior_reliability=0.1133` `competition=La Liga`	fouls: 1.7557 -> 1.7741 interceptionWon: 0.9404 -> 0.9296 wonTackle: 0.6711 -> 0.6762
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Alejandro Grimaldo', 'matchup_zone': 'Right Wing vs Left Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 70.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0066` `goal_rate=0.0` `scorer_weight=0.006`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0108` `assist_rate=0.0082` `provider_weight=0.0106`	-

Omar El Hilali D 88 0 0.00 - n/a 0 0.0 6.8 58.7 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0026, quality 0.589)	`market_value_eur=15900000` `quality_factor=0.5889` `boost_factor=1.0026` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0135)	`blend_weight=0.0135` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.98` `prior_reliability=0.2` `competition=La Liga`	totalCross: 2.6746 -> 2.687 wasFouled: 0.812 -> 0.801 wonTackle: 1.2521 -> 1.244 totalShots: 0.4349 -> 0.429
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Koke', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1115 -> 0.1252
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 75.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `goal_rate=0.0` `scorer_weight=0.001`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1177` `assist_rate=0.1077` `provider_weight=0.1223`	-

Ángel Fortuño G 90 0 0.00 - n/a 0.0 7.0 4.6 32 XGBoost G_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9855, quality 0.437)	`market_value_eur=290000` `quality_factor=0.4374` `boost_factor=0.9855` `reference_quality=0.505` `history_matches=2` `history_shrink=0.7143`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Jonathan David, Ademola Lookman', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 2.1749}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.34 (squad xG sum 1.0719, xG scale 1.2473, xA scale 1.1232, team xA 1.07 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.337` `squad_xg_sum=1.0719` `squad_xa_sum=0.9523` `team_expected_assists=1.0696` `team_xa_to_xg_ratio=0.8`	shots_faced: 20.7 -> 22.1 expected_saves: 5.07 -> 5.41 high_pressure_saves: 1.74 -> 1.86
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 89.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.4295`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=1.0297`	-

Team totals 1 1.34 11 6.8

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Marko Dmitrović	G	4.4m	13.1%	Market-value substitute prior
Vanja Drkušić	D	19.9m	41.2%	Market-value substitute prior
Unai Núñez	D	13.6m	21.7%	Market-value substitute prior
Urko González	M	22.4m	31.5%	Market-value substitute prior
Gabriel Moscardo	M	34.7m	69.6%	Market-value substitute prior
Javi Puado	F	25.5m	56.8%	Market-value substitute prior

Atlético Madrid — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Jonathan David F 66 1 0.67 0.31 118% 3 1.4 7.0 37.5 8 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0342, quality 0.710)	`market_value_eur=29000000` `quality_factor=0.71` `boost_factor=1.0342` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Leandro Cabrera, Quilindschy Hartman', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.264, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.0935, 'skill_edge': 0.0896, 'height_edge': -0.026, 'attacker_skill': 0.71, 'defender_skill': 0.5108, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Leandro Cabrera', 'market_value': 930000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}, {'player': 'Quilindschy Hartman', 'market_value': 11600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5842}], 'xg_override': 0.4631}	totalShots: 2.4171 -> 2.6431 bigChanceMissed: 1.4255 -> 1.5588 shotOffTarget: 1.3173 -> 1.4405 onTargetScoringAttempt: 1.0998 -> 1.2026 expectedGoalsOnTarget: 0.7013 -> 0.7669 goals: 0.4546 -> 0.4971 expectedGoals: 0.4235 -> 0.4631
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	assists: 1.0 -> 0
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.4631 -> 0.6659 expected_assists: 0.0885 -> 0.1282
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 80.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.4862` `goal_rate=0.4971` `scorer_weight=0.71`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1203` `assist_rate=0.6925` `provider_weight=0.1603`	-

Ademola Lookman F 66 1 0.45 0.23 99% 2 1.5 6.9 57.0 27 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0177, quality 0.800)	`market_value_eur=44000000` `quality_factor=0.8002` `boost_factor=1.0177` `reference_quality=0.505` `history_matches=20` `history_shrink=0.2`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0225)	`blend_weight=0.0225` `history_matches=20` `season_overlap_discount=1.0` `per90_to_match_factor=0.7377` `prior_reliability=0.1733` `competition=La Liga`	wasFouled: 0.9645 -> 0.9747 totalShots: 2.0823 -> 2.0739 totalCross: 0.9201 -> 0.9122 interceptionWon: 0.017 -> 0.023
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Leandro Cabrera, Quilindschy Hartman', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.28, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.1179, 'skill_edge': 0.1302, 'height_edge': -0.05, 'attacker_skill': 0.8002, 'defender_skill': 0.5108, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Leandro Cabrera', 'market_value': 930000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}, {'player': 'Quilindschy Hartman', 'market_value': 11600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5842}], 'xg_override': 0.3135}	totalShots: 2.0831 -> 2.3287 shotOffTarget: 1.2601 -> 1.4087 onTargetScoringAttempt: 0.823 -> 0.92 bigChanceMissed: 0.5396 -> 0.6032 goals: 0.2837 -> 0.3171 expectedGoals: 0.2804 -> 0.3135 expectedGoalsOnTarget: 0.1582 -> 0.1769
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.3135 -> 0.4508 expected_assists: 0.1091 -> 0.158
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 13 by projected rating, fifa_overall 83.0)	`match_squad_rank=13` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.3629` `goal_rate=0.3171` `scorer_weight=0.4553`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1462` `assist_rate=0.0` `provider_weight=0.1264`	-

Pablo Barrios M 77 0 0.11 0.03 253% 1 0.7 7.2 55.5 14 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0170, quality 0.879)	`market_value_eur=77000000` `quality_factor=0.8792` `boost_factor=1.017` `reference_quality=0.505` `history_matches=28` `history_shrink=0.1515`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0151)	`blend_weight=0.0151` `history_matches=28` `season_overlap_discount=1.0` `per90_to_match_factor=0.851` `prior_reliability=0.1533` `competition=La Liga`	totalCross: 1.4483 -> 1.432 wasFouled: 0.9393 -> 0.953 totalShots: 0.9418 -> 0.9332 interceptionWon: 1.6066 -> 1.5991
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Pol Lozano, Edu Expósito', 'matchup_zone': 'Attacking Mid vs Double Pivot — Tight Space Battle', 'duel_win_rate': 0.446, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.3021, 'skill_edge': 0.1835, 'height_edge': 0.022, 'attacker_skill': 0.8829, 'defender_skill': 0.4751, 'attacker_role': 'CAM', 'attacker_mv_price_index': 0.95, 'defender_adjusted_parts': [{'player': 'Pol Lozano', 'market_value': 5800000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4845}, {'player': 'Edu Expósito', 'market_value': 4500000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4658}], 'xg_override': 0.078}	aerialWon: 0.7071 -> 1.1 totalShots: 0.9435 -> 1.2285 shotOffTarget: 0.6796 -> 0.8849 onTargetScoringAttempt: 0.2639 -> 0.3436 expectedGoalsOnTarget: 0.1087 -> 0.1415 goals: 0.061 -> 0.0794 expectedGoals: 0.0599 -> 0.078 bigChanceMissed: 0.0271 -> 0.0353
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.078 -> 0.1122 expected_assists: 0.1178 -> 0.1706
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 83.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1061` `goal_rate=0.0794` `scorer_weight=0.1134`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1568` `assist_rate=0.1155` `provider_weight=0.1683`	-

Giuliano Simeone M 76 0 0.21 0.10 102% 2 0.8 6.9 57.6 29 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0101, quality 0.800)	`market_value_eur=44000000` `quality_factor=0.8002` `boost_factor=1.0101` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0118)	`blend_weight=0.0118` `history_matches=39` `season_overlap_discount=1.0` `per90_to_match_factor=0.8459` `prior_reliability=0.16` `competition=La Liga`	interceptionWon: 0.6547 -> 0.6636
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Pol Lozano', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.564, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.211, 'skill_edge': 0.1555, 'height_edge': -0.012, 'attacker_skill': 0.8301, 'defender_skill': 0.4845, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Pol Lozano', 'market_value': 5800000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4845}], 'xg_override': 0.145}`	aerialWon: 1.217 -> 0.9 totalShots: 1.1308 -> 1.3694 shotOffTarget: 0.7054 -> 0.8542 onTargetScoringAttempt: 0.4211 -> 0.51 expectedGoals: 0.1197 -> 0.145 expectedGoalsOnTarget: 0.1109 -> 0.1343 bigChanceMissed: 0.1073 -> 0.1299 goals: 0.093 -> 0.1126
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.145 -> 0.2085 expected_assists: 0.1076 -> 0.1558
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 81.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1882` `goal_rate=0.1126` `scorer_weight=0.1738`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1443` `assist_rate=0.1065` `provider_weight=0.1547`	-

Koke M 83 0 0.06 0.02 229% 1 0.3 7.2 61.3 29 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9985, quality 0.457)	`market_value_eur=4600000` `quality_factor=0.4572` `boost_factor=0.9985` `reference_quality=0.505` `history_matches=42` `history_shrink=0.1064`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0083)	`blend_weight=0.0083` `history_matches=42` `season_overlap_discount=1.0` `per90_to_match_factor=0.9172` `prior_reliability=0.12` `competition=La Liga`	wasFouled: 1.4015 -> 1.4068
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Omar El Hilali', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.366, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 0.8327, 'skill_edge': -0.0898, 'height_edge': -0.024, 'attacker_skill': 0.4572, 'defender_skill': 0.6569, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Omar El Hilali', 'market_value': 15900000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.6569}], 'xg_override': 0.0404}	aerialWon: 0.209 -> 1.0 totalShots: 0.6744 -> 0.5616 shotOffTarget: 0.5203 -> 0.4333 onTargetScoringAttempt: 0.1541 -> 0.1283 goals: 0.0559 -> 0.0465 expectedGoals: 0.0485 -> 0.0404
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0404 -> 0.0581 expected_assists: 0.123 -> 0.1781
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 81.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0564` `goal_rate=0.0465` `scorer_weight=0.062`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1631` `assist_rate=0.123` `provider_weight=0.1781`	-

Johnny Cardoso M 79 0 0.06 0.01 624% 1 0.2 6.9 48.7 24 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0065, quality 0.618)	`market_value_eur=18700000` `quality_factor=0.6181` `boost_factor=1.0065` `reference_quality=0.505` `history_matches=21` `history_shrink=0.1923`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0008)	`blend_weight=0.0008` `history_matches=21` `season_overlap_discount=1.0` `per90_to_match_factor=0.88` `prior_reliability=0.0067` `competition=La Liga`	wasFouled: 0.7388 -> 0.7456 interceptionWon: 1.3335 -> 1.3398 wonTackle: 1.5205 -> 1.5266 fouls: 1.6242 -> 1.6302
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Pol Lozano', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.535, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.1685, 'skill_edge': 0.0746, 'height_edge': 0.04, 'attacker_skill': 0.6503, 'defender_skill': 0.4845, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Pol Lozano', 'market_value': 5800000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4845}], 'xg_override': 0.0387}`	aerialWon: 0.8353 -> 1.2 totalShots: 0.5498 -> 0.6424 shotOffTarget: 0.4173 -> 0.4876 onTargetScoringAttempt: 0.1325 -> 0.1548 goals: 0.0451 -> 0.0527 expectedGoals: 0.0331 -> 0.0387
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0387 -> 0.0557 expected_assists: 0.0568 -> 0.0823
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 80.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0542` `goal_rate=0.0527` `scorer_weight=0.0645`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.079` `assist_rate=0.051` `provider_weight=0.0778`	-

Alejandro Grimaldo M 88 0 0.30 0.21 43% 2 1.3 7.2 51.2 18 XGBoost M_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0150, quality 0.625)	`market_value_eur=19400000` `quality_factor=0.6251` `boost_factor=1.015` `reference_quality=0.505` `history_matches=7` `history_shrink=0.4167`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Clemens Riedel', 'matchup_zone': 'Right Wing vs Left Back — Flank Battle', 'duel_win_rate': 0.49, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.0149, 'skill_edge': 0.0702, 'height_edge': -0.06, 'attacker_skill': 0.6251, 'defender_skill': 0.4692, 'attacker_role': 'RW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Clemens Riedel', 'market_value': 3900000, 'role': 'LB', 'price_index': 0.7, 'adjusted_quality': 0.4692}], 'xg_override': 0.2062}	aerialWon: 0.4386 -> 0.8 totalShots: 2.1496 -> 2.1816 shotOffTarget: 1.4366 -> 1.458 onTargetScoringAttempt: 0.7129 -> 0.7235
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2062 -> 0.2965 expected_assists: 0.149 -> 0.2159
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 89.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2566` `goal_rate=0.1879` `scorer_weight=0.2753`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1942` `assist_rate=0.1317` `provider_weight=0.1971`	-

Robin Le Normand D 80 0 0.03 0.01 344% 0 0.1 6.7 51.0 28 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0039, quality 0.603)	`market_value_eur=17200000` `quality_factor=0.6027` `boost_factor=1.0039` `reference_quality=0.505` `history_matches=33` `history_shrink=0.1316`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0103)	`blend_weight=0.0103` `history_matches=33` `season_overlap_discount=1.0` `per90_to_match_factor=0.8928` `prior_reliability=0.12` `competition=La Liga`	interceptionWon: 1.0099 -> 1.0199
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Roberto Fernández', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0221 -> 0.0318
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 16 by projected rating, fifa_overall 81.0)	`match_squad_rank=16` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0313` `goal_rate=0.1469` `scorer_weight=0.0398`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0106` `assist_rate=0.0581` `provider_weight=0.0134`	-

Cristian Romero D 81 0 0.20 0.08 141% 1 0.9 7.0 56.5 18 FBref Prior D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0118, quality 0.764)	`market_value_eur=37000000` `quality_factor=0.7641` `boost_factor=1.0118` `reference_quality=0.505` `history_matches=28` `history_shrink=0.1515`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Roberto Fernández', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1379 -> 0.1983 expected_assists: 0.0454 -> 0.0657
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 82.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1799` `goal_rate=0.2601` `scorer_weight=0.2479`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0636` `assist_rate=0.0044` `provider_weight=0.0526`	-

Marcos Llorente D 83 0 0.08 0.04 119% 1 0.6 7.0 59.6 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0048, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0048` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0077)	`blend_weight=0.0077` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.921` `prior_reliability=0.1` `competition=La Liga`	fouls: 0.7436 -> 0.7664 totalCross: 1.7696 -> 1.7559 interceptionWon: 1.1007 -> 1.1112 wasFouled: 0.239 -> 0.2467
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Roberto Fernández', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0567 -> 0.0815 expected_assists: 0.0898 -> 0.13
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 85.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0783` `goal_rate=0.0` `scorer_weight=0.0652`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1219` `assist_rate=0.108` `provider_weight=0.1469`	-

Jan Oblak G 90 0 0.00 - n/a 0.0 7.0 54.9 27 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0030, quality 0.590)	`market_value_eur=16000000` `quality_factor=0.59` `boost_factor=1.003` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0151)	`blend_weight=0.0151` `history_matches=38` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.2` `competition=La Liga`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Roberto Fernández, Javier Hernández', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.4715}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.33 (squad xG sum 1.6168, xG scale 1.4380, xA scale 1.4481, team xA 1.51 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=2.325` `squad_xg_sum=1.6168` `squad_xa_sum=1.0436` `team_expected_assists=1.5113` `team_xa_to_xg_ratio=0.65`	shots_faced: 14.0 -> 12.7 expected_saves: 3.43 -> 3.11 high_pressure_saves: 1.18 -> 1.07
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 88.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=2.409`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=1.5221`	-

Team totals 2 2.33 16 7.0

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Juan Musso	G	4.4m	10.9%	Market-value substitute prior
Marc Pubill	D	15.1m	24.8%	Market-value substitute prior
Dávid Hancko	D	14.8m	24.3%	Market-value substitute prior
Morten Hjulmand	M	17.7m	46.6%	Market-value substitute prior
Rodrigo Mendoza	M	13.6m	45.2%	Market-value substitute prior
Julián Alvarez	F	25.9m	75.6%	Market-value substitute prior

Antagonist / tactical summary

Away midfield dictates play with ~53% possession expected (Javier Hernández vs Pablo Barrios). Total antagonist xG — Home: 0.82, Away: 1.28. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Midfield — possession 46.6% / 53.4%, dominance ratio 0.39, key duel: Javier Hernández vs Pablo Barrios

Javier Hernández xG 0.12Roberto Fernández xG 0.37Bryan Zaragoza xG 0.10Koke xG 0.04Alejandro Grimaldo xG 0.21Pablo Barrios xG 0.08

Summary

Antagonist matchup analysis for Espanyol vs Atlético Madrid. Most likely scoreline: 1-2. Lambdas derived from 6 home and 7 away individual player duels. Head-to-head: Espanyol 1W-5D-4L vs Atlético Madrid over 10 meetings.

Tactical parity between 4-2-3-1 and 4-2-3-1. Match outcome driven primarily by individual player quality. Away midfield dictates play with ~53% possession expected (Javier Hernández vs Pablo Barrios). Total antagonist xG — Home: 0.82, Away: 1.28. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Newcastle vs Aston Villa

Premier League · 20261017 · 4-3-3 vs 4-2-3-1

Predicted 1-1Winner: DrawWinning H 1-0 (10.1%)Winning A 0-1 (8.1%)Confidence 0.71λ home/away 1.49 / 1.19Most-likely P 12.3%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Newcastle vs Aston Villa. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-3-3 vs 4-2-3-1 (duel lambdas: Home 1.466, Away 1.210). Final calibrated Poisson lambdas: Home 1.487, Away 1.193 (most likely score 1-1).

Home Starters 11/11 XGBoost (8 FBref) Away Starters 11/11 XGBoost (9 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.47 / 1.21 Final λ (H/A) 1.49 / 1.19

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Newcastle	Aston Villa
League position	11.00	16.00
Points per game	1.34	1.09
Attack index	1.04	0.89
Defense (concede) index	1.08	1.16
Form multiplier	1.01	0.98
Standing λ multiplier	1.05	1.01

Lambda decomposition (how the goal rates were built)

Stage	Newcastle λ	Aston Villa λ
Pre-competitive-floor	1.487	1.193
Pre-calibration (Micro-duels)	1.466	1.210
Final λ	1.487	1.193

competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Falsecompetitive underdog floor applied Falsestage discount 1.0

Scoreline distribution & markets

Top 5 scorelines	P
1-1	12.3%
1-0	10.1%
2-1	9.0%
0-1	8.1%
2-0	7.6%

Home win	43.9%
Draw	25.8%
Away win	30.3%
Over 2.5	50.1%
BTTS yes	54.0%

Pregame fair line — favorite Newcastle (43.9%), fair total 2.68, fair spread 0.29

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Newcastle — projected line vs season baseline (4-3-3)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Sven Botman D 89 0 0.05 0.06 ok 1 0.5 6.9 47.0 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0097, quality 0.739)	`market_value_eur=33000000` `quality_factor=0.7389` `boost_factor=1.0097` `reference_quality=0.505` `history_matches=31` `history_shrink=0.1389`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0120)	`blend_weight=0.012` `history_matches=31` `season_overlap_discount=1.0` `per90_to_match_factor=0.9897` `prior_reliability=0.1333` `competition=Premier League`	interceptionWon: 1.2071 -> 1.1926 wonTackle: 0.8191 -> 0.8092 fouls: 0.4595 -> 0.454
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Nicolas Jackson', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0385 -> 0.0487 expected_assists: 0.0397 -> 0.0539
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 81.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0475` `goal_rate=0.1649` `scorer_weight=0.0609`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0525` `assist_rate=0.0` `provider_weight=0.0431`	-

Malick Thiaw D 90 0 0.08 0.11 ok 0 0.5 6.8 55.3 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0099, quality 0.809)	`market_value_eur=46000000` `quality_factor=0.8089` `boost_factor=1.0099` `reference_quality=0.505` `history_matches=41` `history_shrink=0.1087`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0094)	`blend_weight=0.0094` `history_matches=41` `season_overlap_discount=1.0` `per90_to_match_factor=0.998` `prior_reliability=0.1333` `competition=Premier League`	wasFouled: 0.3895 -> 0.3999 interceptionWon: 0.9389 -> 0.9301 wonTackle: 0.6124 -> 0.6066
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Nicolas Jackson', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0613 -> 0.0776 expected_assists: 0.0182 -> 0.0247
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 81.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0747` `goal_rate=0.0156` `scorer_weight=0.0621`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0244` `assist_rate=0.0` `provider_weight=0.0198`	-

Amar Dedić D 83 0 0.05 0.01 213% 1 0.4 6.7 29.2 11 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0307, quality 0.710)	`market_value_eur=29000000` `quality_factor=0.71` `boost_factor=1.0307` `reference_quality=0.505` `history_matches=5` `history_shrink=0.5`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0433)	`blend_weight=0.0433` `history_matches=5` `season_overlap_discount=1.0` `per90_to_match_factor=0.9257` `prior_reliability=0.1333` `competition=Premier League`	interceptionWon: 1.0501 -> 1.0046 wonTackle: 1.133 -> 1.104 wasFouled: 0.9059 -> 0.9268 goalAssist: 0.2337 -> 0.2436 totalShots: 0.6381 -> 0.6305
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'John McGinn', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0365 -> 0.0462 expected_assists: 0.0214 -> 0.0291
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 15 by projected rating, fifa_overall 76.0)	`match_squad_rank=15` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0451` `goal_rate=0.0` `scorer_weight=0.037`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0287` `assist_rate=0.2436` `provider_weight=0.0364`	-

Lukáš Horníček G 90 0 0.00 - n/a 0.0 7.0 35.3 12 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0235, quality 0.677)	`market_value_eur=25000000` `quality_factor=0.6773` `boost_factor=1.0235` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0394)	`blend_weight=0.0394` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	saves: 3.03 -> 3.0879 fouls: 0.2465 -> 0.2368
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Nicolas Jackson, Johan Manzambi', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.2102}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	shots_faced: 11.5 -> 11.3 expected_saves: 2.81 -> 2.76 high_pressure_saves: 0.97 -> 0.95
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 77.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.5944`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.9789`	-

Lewis Hall D 88 0 0.06 0.03 128% 1 0.6 6.9 51.3 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0119, quality 0.831)	`market_value_eur=52000000` `quality_factor=0.8309` `boost_factor=1.0119` `reference_quality=0.505` `history_matches=36` `history_shrink=0.122`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0106)	`blend_weight=0.0106` `history_matches=36` `season_overlap_discount=1.0` `per90_to_match_factor=0.9809` `prior_reliability=0.1333` `competition=Premier League`	fouls: 0.5949 -> 0.6042 wonTackle: 1.4003 -> 1.4062 onTargetScoringAttempt: 0.5166 -> 0.5111
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Johan Manzambi', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0494 -> 0.0625 expected_assists: 0.1073 -> 0.1457
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 83.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0606` `goal_rate=0.1434` `scorer_weight=0.0781`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1356` `assist_rate=0.0948` `provider_weight=0.1341`	-

Lewis Miley M 81 0 0.05 0.03 50% 1 0.2 6.9 45.2 23 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0100, quality 0.732)	`market_value_eur=32000000` `quality_factor=0.732` `boost_factor=1.01` `reference_quality=0.505` `history_matches=29` `history_shrink=0.1471`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0115)	`blend_weight=0.0115` `history_matches=29` `season_overlap_discount=1.0` `per90_to_match_factor=0.8979` `prior_reliability=0.12` `competition=Premier League`	totalCross: 1.2287 -> 1.2146 wasFouled: 0.2134 -> 0.2224 wonTackle: 1.1935 -> 1.1855
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'João Gomes', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel', 'duel_win_rate': 0.481, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 1.046, 'skill_edge': -0.0167, 'height_edge': 0.048, 'attacker_skill': 0.7554, 'defender_skill': 0.7926, 'attacker_role': 'CM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'João Gomes', 'market_value': 36000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.7926}], 'xg_override': 0.0369}	aerialWon: 1.7516 -> 1.2 totalShots: 0.4895 -> 0.512 shotOffTarget: 0.3554 -> 0.3717 onTargetScoringAttempt: 0.1326 -> 0.1387
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0369 -> 0.0467 expected_assists: 0.0657 -> 0.0892
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 77.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0456` `goal_rate=0.0446` `scorer_weight=0.0508`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0853` `assist_rate=0.0739` `provider_weight=0.0955`	-

Nico González M 72 0 0.05 0.01 589% 1 0.1 7.0 31.3 26 XGBoost M_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0146, quality 0.837)	`market_value_eur=54000000` `quality_factor=0.837` `boost_factor=1.0146` `reference_quality=0.505` `history_matches=29` `history_shrink=0.1471`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'João Gomes', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel', 'duel_win_rate': 0.517, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 1.0985, 'skill_edge': 0.027, 'height_edge': 0.04, 'attacker_skill': 0.8526, 'defender_skill': 0.7926, 'attacker_role': 'CM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'João Gomes', 'market_value': 36000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.7926}], 'xg_override': 0.0363}	aerialWon: 0.7637 -> 1.2 totalShots: 0.5997 -> 0.6588 shotOffTarget: 0.4438 -> 0.4875 onTargetScoringAttempt: 0.1559 -> 0.1713
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0363 -> 0.046 expected_assists: 0.0865 -> 0.1174
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 81.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.045` `goal_rate=0.0468` `scorer_weight=0.0516`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1108` `assist_rate=0.0642` `provider_weight=0.0982`	-

Joelinton M 85 0 0.09 0.08 ok 1 1.0 6.8 48.7 18 XGBoost M_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0062, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0062` `reference_quality=0.505` `history_matches=28` `history_shrink=0.1515`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'João Gomes', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.407, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.9808, 'skill_edge': -0.053, 'height_edge': 0.04, 'attacker_skill': 0.6747, 'defender_skill': 0.7926, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'João Gomes', 'market_value': 36000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.7926}], 'xg_override': 0.0723}	aerialWon: 2.7846 -> 1.2 totalShots: 0.9628 -> 0.9443 shotOffTarget: 0.6874 -> 0.6742 onTargetScoringAttempt: 0.2753 -> 0.27
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0723 -> 0.0915 expected_assists: 0.0583 -> 0.0791
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 80.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0874` `goal_rate=0.0758` `scorer_weight=0.0941`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0761` `assist_rate=0.0603` `provider_weight=0.0806`	-

Harvey Barnes F 86 0 0.27 0.13 106% 2 1.5 6.9 55.3 37 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0083, quality 0.770)	`market_value_eur=38000000` `quality_factor=0.7699` `boost_factor=1.0083` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0090)	`blend_weight=0.009` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.9518` `prior_reliability=0.1333` `competition=Premier League`	totalCross: 2.4973 -> 2.4876 totalShots: 1.8925 -> 1.8841 onTargetScoringAttempt: 0.6189 -> 0.6133
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Victor Lindelöf', 'matchup_zone': 'Right Wing vs Left Back — Flank Battle', 'duel_win_rate': 0.584, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.1533, 'skill_edge': 0.1283, 'height_edge': -0.024, 'attacker_skill': 0.7699, 'defender_skill': 0.4849, 'attacker_role': 'RW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Victor Lindelöf', 'market_value': 4800000, 'role': 'LB', 'price_index': 0.7, 'adjusted_quality': 0.4849}], 'xg_override': 0.2165}	totalShots: 1.8869 -> 2.1762 shotOffTarget: 1.2736 -> 1.4688 onTargetScoringAttempt: 0.6133 -> 0.7073 goals: 0.2066 -> 0.2383 expectedGoalsOnTarget: 0.2055 -> 0.237 expectedGoals: 0.1877 -> 0.2165 bigChanceMissed: 0.1815 -> 0.2093
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2165 -> 0.2741 expected_assists: 0.1188 -> 0.1613
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 80.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2397` `goal_rate=0.2383` `scorer_weight=0.2965`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.149` `assist_rate=0.3167` `provider_weight=0.2016`	-

Matías Fernández-Pardo F 77 0 0.32 0.05 489% 2 0.2 6.9 26.6 10 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0579, quality 0.853)	`market_value_eur=60000000` `quality_factor=0.8526` `boost_factor=1.0579` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Aaron Wan-Bissaka, Pau Torres', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.268, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.1005, 'skill_edge': 0.0864, 'height_edge': -0.018, 'attacker_skill': 0.8526, 'defender_skill': 0.6606, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Aaron Wan-Bissaka', 'market_value': 14100000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.6191}, {'player': 'Pau Torres', 'market_value': 21000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7022}], 'xg_override': 0.2507}	totalShots: 1.3573 -> 1.4937 bigChanceMissed: 0.8225 -> 0.9052 shotOffTarget: 0.7118 -> 0.7833 onTargetScoringAttempt: 0.6455 -> 0.7104 goals: 0.2522 -> 0.2775 expectedGoals: 0.2278 -> 0.2507
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2507 -> 0.3174 expected_assists: 0.0833 -> 0.1131
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 78.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.272` `goal_rate=0.2775` `scorer_weight=0.3457`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1069` `assist_rate=0.0` `provider_weight=0.0905`	-

Yoane Wissa F 82 1 0.32 0.19 73% 2 0.8 6.7 41.1 34 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0077, quality 0.660)	`market_value_eur=23000000` `quality_factor=0.6595` `boost_factor=1.0077` `reference_quality=0.505` `history_matches=25` `history_shrink=0.1667`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0137)	`blend_weight=0.0137` `history_matches=25` `season_overlap_discount=1.0` `per90_to_match_factor=0.9112` `prior_reliability=0.1267` `competition=Premier League`	wasFouled: 1.2941 -> 1.329 totalShots: 1.5309 -> 1.5497 totalOffside: 0.2619 -> 0.2715 goalAssist: 0.0835 -> 0.0889
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Matty Cash', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.424, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 0.9174, 'skill_edge': -0.0202, 'height_edge': -0.036, 'attacker_skill': 0.6595, 'defender_skill': 0.7044, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Matty Cash', 'market_value': 19800000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.7044}], 'xg_override': 0.2559}	totalShots: 1.5497 -> 1.4217 onTargetScoringAttempt: 0.8219 -> 0.754 shotOffTarget: 0.711 -> 0.6523 bigChanceMissed: 0.6211 -> 0.5698 goals: 0.3069 -> 0.2816 expectedGoalsOnTarget: 0.2949 -> 0.2705 expectedGoals: 0.2789 -> 0.2559
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.1746, xG scale 1.2660, xA scale 1.3576, team xA 0.97 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.487` `squad_xg_sum=1.1746` `squad_xa_sum=0.712` `team_expected_assists=0.9666` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2559 -> 0.324 expected_assists: 0.0433 -> 0.0588
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 14 by projected rating, fifa_overall 80.0)	`match_squad_rank=14` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.2767` `goal_rate=0.2816` `scorer_weight=0.3512`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0571` `assist_rate=0.0889` `provider_weight=0.0735`	-

Team totals 1 1.49 13 6.9

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Ewen Jaouen	G	5.1m	14.2%	Market-value substitute prior
Tino Livramento	D	15.0m	36.4%	Market-value substitute prior
Dan Burn	D	13.6m	30.1%	Market-value substitute prior
Aladji Bamba	M	18.5m	71.0%	Market-value substitute prior
Sean Steur	M	21.2m	51.6%	Market-value substitute prior
Nick Woltemade	F	23.1m	75.2%	Market-value substitute prior

Aston Villa — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Nicolas Jackson F 78 1 0.34 0.11 202% 2 1.1 6.9 32.7 11 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0443, quality 0.800)	`market_value_eur=44000000` `quality_factor=0.8002` `boost_factor=1.0443` `reference_quality=0.505` `history_matches=5` `history_shrink=0.5`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0173)	`blend_weight=0.0173` `history_matches=5` `season_overlap_discount=1.0` `per90_to_match_factor=0.8672` `prior_reliability=0.0533` `competition=Premier League`	totalShots: 2.5898 -> 2.5449 onTargetScoringAttempt: 1.0936 -> 1.0747 wasFouled: 0.6938 -> 0.6818 totalOffside: 0.6516 -> 0.6403 totalCross: 0.5592 -> 0.5683 fouls: 0.4428 -> 0.4351 goals: 0.4106 -> 0.4035
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Sven Botman, Malick Thiaw', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.9376, 'skill_edge': -0.0124, 'height_edge': -0.03, 'attacker_skill': 0.8002, 'defender_skill': 0.8279, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Sven Botman', 'market_value': 33000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8002}, {'player': 'Malick Thiaw', 'market_value': 46000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8555}], 'xg_override': 0.3716}	totalShots: 2.5709 -> 2.4105 shotOffTarget: 1.4962 -> 1.4028 onTargetScoringAttempt: 1.0747 -> 1.0076 goals: 0.4035 -> 0.3783 expectedGoals: 0.3963 -> 0.3716 expectedGoalsOnTarget: 0.3911 -> 0.3667 bigChanceMissed: 0.2394 -> 0.2245
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.3716 -> 0.3445 expected_assists: 0.0851 -> 0.0966
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 79.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.2914` `goal_rate=0.3783` `scorer_weight=0.35`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0921` `assist_rate=0.0` `provider_weight=0.0773`	-

Emiliano Buendía F 77 0 0.15 0.11 36% 1 1.3 6.9 49.2 41 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0022, quality 0.575)	`market_value_eur=14600000` `quality_factor=0.5748` `boost_factor=1.0022` `reference_quality=0.505` `history_matches=42` `history_shrink=0.1064`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Sven Botman, Malick Thiaw', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.7444, 'skill_edge': -0.1139, 'height_edge': -0.06, 'attacker_skill': 0.5748, 'defender_skill': 0.8279, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Sven Botman', 'market_value': 33000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8002}, {'player': 'Malick Thiaw', 'market_value': 46000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8555}], 'xg_override': 0.1578}	totalShots: 1.8582 -> 1.3833 shotOffTarget: 1.256 -> 0.935 onTargetScoringAttempt: 0.6022 -> 0.4483 bigChanceMissed: 0.2665 -> 0.1984 expectedGoals: 0.212 -> 0.1578 goals: 0.2012 -> 0.1498 expectedGoalsOnTarget: 0.0888 -> 0.0661
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1578 -> 0.1463 expected_assists: 0.0906 -> 0.1028
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 80.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1361` `goal_rate=0.1498` `scorer_weight=0.1407`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0977` `assist_rate=0.0031` `provider_weight=0.0822`	-

Johan Manzambi M 83 0 0.23 0.11 114% 3 1.2 6.8 23.8 9 XGBoost M_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0688, quality 0.872)	`market_value_eur=71000000` `quality_factor=0.8721` `boost_factor=1.0688` `reference_quality=0.505` `history_matches=3` `history_shrink=0.625`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Lewis Hall', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.49, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 1.0149, 'skill_edge': -0.0019, 'height_edge': 0.012, 'attacker_skill': 0.8721, 'defender_skill': 0.8762, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Lewis Hall', 'market_value': 52000000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.8762}], 'xg_override': 0.253}	aerialWon: 0.0 -> 1.1 totalShots: 2.8375 -> 2.8798 shotOffTarget: 2.0029 -> 2.0327 onTargetScoringAttempt: 0.8346 -> 0.847 expectedGoalsOnTarget: 0.3966 -> 0.4025
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.253 -> 0.2346 expected_assists: 0.1279 -> 0.1453
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 80.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2091` `goal_rate=0.1967` `scorer_weight=0.191`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1352` `assist_rate=0.224` `provider_weight=0.1816`	-

João Gomes M 83 0 0.04 0.02 55% 1 0.4 6.8 50.4 27 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0088, quality 0.758)	`market_value_eur=36000000` `quality_factor=0.7581` `boost_factor=1.0088` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0020)	`blend_weight=0.002` `history_matches=38` `season_overlap_discount=1.0` `per90_to_match_factor=0.924` `prior_reliability=0.0267` `competition=Premier League`	wonTackle: 2.1057 -> 2.1108
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Lewis Miley', 'matchup_zone': 'Holding Mid vs Central Mid — Midfield Anchor Duel', 'duel_win_rate': 0.369, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.954, 'skill_edge': 0.0167, 'height_edge': -0.048, 'attacker_skill': 0.7926, 'defender_skill': 0.7554, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Lewis Miley', 'market_value': 32000000, 'role': 'CM', 'price_index': 0.9, 'adjusted_quality': 0.7554}], 'xg_override': 0.0404}	aerialWon: 0.5886 -> 0.9 totalShots: 0.5646 -> 0.5386 shotOffTarget: 0.3916 -> 0.3736 onTargetScoringAttempt: 0.173 -> 0.165
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0486 -> 0.0552
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 78.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0368` `goal_rate=0.0472` `scorer_weight=0.0403`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0537` `assist_rate=0.0672` `provider_weight=0.0656`	-

Boubacar Kamara M 87 0 0.05 0.01 308% 1 0.2 7.0 39.9 14 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0163, quality 0.821)	`market_value_eur=49000000` `quality_factor=0.8206` `boost_factor=1.0163` `reference_quality=0.505` `history_matches=24` `history_shrink=0.1724`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0149)	`blend_weight=0.0149` `history_matches=24` `season_overlap_discount=1.0` `per90_to_match_factor=0.9696` `prior_reliability=0.1333` `competition=Premier League`	wasFouled: 2.0842 -> 2.0603 fouls: 1.2201 -> 1.2381 totalShots: 0.6953 -> 0.6849 interceptionWon: 0.8163 -> 0.8113
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Lewis Miley', 'matchup_zone': 'Holding Mid vs Central Mid — Midfield Anchor Duel', 'duel_win_rate': 0.425, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 1.037, 'skill_edge': 0.0412, 'height_edge': -0.016, 'attacker_skill': 0.847, 'defender_skill': 0.7554, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Lewis Miley', 'market_value': 32000000, 'role': 'CM', 'price_index': 0.9, 'adjusted_quality': 0.7554}], 'xg_override': 0.0499}	aerialWon: 1.7837 -> 1.1 totalShots: 0.6921 -> 0.7177 shotOffTarget: 0.4892 -> 0.5073 onTargetScoringAttempt: 0.2029 -> 0.2104
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0585 -> 0.0664
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 84.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0452` `goal_rate=0.0548` `scorer_weight=0.0486`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0642` `assist_rate=0.0654` `provider_weight=0.0706`	-

Aaron Wan-Bissaka D 80 0 0.00 - n/a 0 0.0 6.6 43.0 28 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0028, quality 0.569)	`market_value_eur=14100000` `quality_factor=0.5693` `boost_factor=1.0028` `reference_quality=0.505` `history_matches=30` `history_shrink=0.1429`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0006)	`blend_weight=0.0006` `history_matches=30` `season_overlap_discount=1.0` `per90_to_match_factor=0.8919` `prior_reliability=0.0067` `competition=Premier League`	fouls: 0.5661 -> 0.5713
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Matías Fernández-Pardo', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 18 by projected rating, fifa_overall 78.0)	`match_squad_rank=18` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `goal_rate=0.0` `scorer_weight=0.001`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0386` `assist_rate=0.0512` `provider_weight=0.0471`	-

Pau Torres D 78 0 0.07 0.04 59% 0 0.2 6.8 29.8 27 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0070, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.007` `reference_quality=0.505` `history_matches=24` `history_shrink=0.1724`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0134)	`blend_weight=0.0134` `history_matches=24` `season_overlap_discount=1.0` `per90_to_match_factor=0.8711` `prior_reliability=0.12` `competition=Premier League`	wonTackle: 0.6169 -> 0.6086 totalShots: 0.4948 -> 0.4881 fouls: 0.5651 -> 0.5705
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Matías Fernández-Pardo', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0706 -> 0.0655
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 81.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0634` `goal_rate=0.0` `scorer_weight=0.0524`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0125` `assist_rate=0.0` `provider_weight=0.0103`	-

Victor Lindelöf D 86 0 0.05 0.02 130% 0 0.1 6.8 33.2 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9976, quality 0.460)	`market_value_eur=4800000` `quality_factor=0.4597` `boost_factor=0.9976` `reference_quality=0.505` `history_matches=23` `history_shrink=0.1786`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0155)	`blend_weight=0.0155` `history_matches=23` `season_overlap_discount=1.0` `per90_to_match_factor=0.9599` `prior_reliability=0.1333` `competition=Premier League`	interceptionWon: 0.8342 -> 0.8213 wonTackle: 0.5665 -> 0.5577 totalShots: 0.3209 -> 0.3159
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Harvey Barnes', 'matchup_zone': 'Right Wing vs Left Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 77.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0533` `goal_rate=0.0` `scorer_weight=0.0438`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0218` `assist_rate=0.0117` `provider_weight=0.0196`	-

Matty Cash D 86 0 0.03 0.02 ok 0 0.5 6.7 52.7 28 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0041, quality 0.629)	`market_value_eur=19800000` `quality_factor=0.6291` `boost_factor=1.0041` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0092)	`blend_weight=0.0092` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.9592` `prior_reliability=0.1267` `competition=Premier League`	totalCross: 2.5781 -> 2.5545 wonTackle: 1.2518 -> 1.2634 wasFouled: 1.227 -> 1.2343 fouls: 1.3666 -> 1.3726
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Yoane Wissa', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0432 -> 0.049
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 13 by projected rating, fifa_overall 81.0)	`match_squad_rank=13` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0252` `goal_rate=0.0064` `scorer_weight=0.0204`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0478` `assist_rate=0.0663` `provider_weight=0.0611`	-

Zion Suzuki G 90 0 0.00 - n/a 0.0 7.0 28.6 12 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0330, quality 0.725)	`market_value_eur=31000000` `quality_factor=0.7249` `boost_factor=1.033` `reference_quality=0.505` `history_matches=5` `history_shrink=0.5`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0217)	`blend_weight=0.0217` `history_matches=5` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.0667` `competition=Premier League`	saves: 3.0968 -> 3.0513 wasFouled: 0.6509 -> 0.6368
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Yoane Wissa, Matías Fernández-Pardo', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.4661}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	shots_faced: 13.9 -> 14.1 expected_saves: 3.4 -> 3.45 high_pressure_saves: 1.17 -> 1.19
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 76.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.1267`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.814`	-

John McGinn F 83 0 0.16 0.12 33% 1 0.8 6.9 46.7 34 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0017, quality 0.552)	`market_value_eur=12600000` `quality_factor=0.5525` `boost_factor=1.0017` `reference_quality=0.505` `history_matches=36` `history_shrink=0.122`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0100)	`blend_weight=0.01` `history_matches=36` `season_overlap_discount=1.0` `per90_to_match_factor=0.9261` `prior_reliability=0.1267` `competition=Premier League`	fouls: 1.2055 -> 1.213 onTargetScoringAttempt: 0.649 -> 0.6425
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Sven Botman, Malick Thiaw', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.7296, 'skill_edge': -0.1239, 'height_edge': -0.06, 'attacker_skill': 0.5525, 'defender_skill': 0.8279, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Sven Botman', 'market_value': 33000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8002}, {'player': 'Malick Thiaw', 'market_value': 46000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8555}], 'xg_override': 0.1767}	totalShots: 1.5712 -> 1.1463 shotOffTarget: 0.9233 -> 0.6736 onTargetScoringAttempt: 0.6425 -> 0.4688 bigChanceMissed: 0.304 -> 0.2218 expectedGoalsOnTarget: 0.247 -> 0.1802 expectedGoals: 0.2422 -> 0.1767 goals: 0.2366 -> 0.1726
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.19 (squad xG sum 1.2865, xG scale 0.9273, xA scale 1.1352, team xA 0.78 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.193` `squad_xg_sum=1.2865` `squad_xa_sum=0.6831` `team_expected_assists=0.7755` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1767 -> 0.1639 expected_assists: 0.097 -> 0.1101
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 83.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1512` `goal_rate=0.1726` `scorer_weight=0.1609`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1043` `assist_rate=0.3316` `provider_weight=0.1376`	-

Team totals 1 1.19 10 6.8

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Marco Bizot	G	4.4m	13.1%	Market-value substitute prior
Ian Maatsen	D	14.8m	24.1%	Market-value substitute prior
Matteo Ruggeri	D	11.5m	62.3%	Market-value substitute prior
Amadou Onana	M	23.4m	56.1%	Market-value substitute prior
Douglas Luiz	M	22.7m	53.2%	Market-value substitute prior
Ibrahim Mbaye	F	26.4m	60.8%	Market-value substitute prior

Antagonist / tactical summary

Midfield is evenly contested — Nico González vs Johan Manzambi will be the pivotal duel. Total antagonist xG — Home: 0.87, Away: 1.05. Home danger zone: Left Wing vs Right Back — Flank Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Midfield — possession 48.9% / 51.1%, dominance ratio 0.47, key duel: Nico González vs Johan Manzambi

Yoane Wissa xG 0.26Harvey Barnes xG 0.22Matías Fernández-Pardo xG 0.25Johan Manzambi xG 0.25Nicolas Jackson xG 0.37Emiliano Buendía xG 0.16

Summary

Antagonist matchup analysis for Newcastle vs Aston Villa. Most likely scoreline: 1-1. Lambdas derived from 6 home and 6 away individual player duels. Head-to-head: Newcastle 5W-2D-3L vs Aston Villa over 10 meetings.

Tactical parity between 4-3-3 and 4-2-3-1. Match outcome driven primarily by individual player quality. Midfield is evenly contested — Nico González vs Johan Manzambi will be the pivotal duel. Total antagonist xG — Home: 0.87, Away: 1.05. Home danger zone: Left Wing vs Right Back — Flank Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Real Betis vs Barcelona

LaLiga · 20261017 · 4-2-3-1 vs 4-3-3

Predicted 1-3Winner: BarcelonaWinning H 2-1 (2.1%)Winning A 1-3 (6.9%)Confidence 0.83λ home/away 1.49 / 3.81Most-likely P 6.9%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Real Betis vs Barcelona. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 4-3-3 (duel lambdas: Home 1.484, Away 3.376). Final calibrated Poisson lambdas: Home 1.493, Away 3.815 (most likely score 1-3).

Home Starters 11/11 XGBoost (9 FBref) Away Starters 11/11 XGBoost (10 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.48 / 3.38 Final λ (H/A) 1.49 / 3.81

FBref Team Priors: Real Betis (Gls/90: 1.3333333333333333, GA/90: 1.33 → E[λ]: 3.5) vs FC Barcelona (Gls/90: 3.6666666666666665, GA/90: 4.0 → E[λ]: 3.5) | Blend Weight: 0.228

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Real Betis	Barcelona
League position	3.00	1.00
Points per game	1.94	2.30
Attack index	1.05	1.70
Defense (concede) index	0.83	0.89
Form multiplier	0.98	1.02
Standing λ multiplier	0.94	1.19

Lambda decomposition (how the goal rates were built)

Stage	Real Betis λ	Barcelona λ
FBref Squad Prior E[λ] (weight 0.228)	3.500	3.500
Pre-mismatch	1.692	3.616
Pre-calibration (Micro-duels)	1.484	3.376
Final λ	1.493	3.815

Scoreline distribution & markets

Top 5 scorelines	P
1-3	6.9%
1-4	6.6%
1-2	5.4%
2-3	5.1%
1-5	5.0%

Home win	10.2%
Draw	11.1%
Away win	78.6%
Over 2.5	89.8%
BTTS yes	75.8%

Pregame fair line — favorite Barcelona (78.6%), fair total 5.31, fair spread -2.32

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Real Betis — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Cucho Hernández F 80 0 0.29 0.28 ok 2 2.1 6.8 60.3 26 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0030, quality 0.595)	`market_value_eur=16500000` `quality_factor=0.5953` `boost_factor=1.003` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Gerard Martín, Pau Cubarsí', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.7678, 'skill_edge': -0.118, 'height_edge': -0.04, 'attacker_skill': 0.5953, 'defender_skill': 0.8575, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Gerard Martín', 'market_value': 36000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8168}, {'player': 'Pau Cubarsí', 'market_value': 94000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8981}], 'xg_override': 0.2398}	totalShots: 2.3122 -> 1.7753 shotOffTarget: 1.4651 -> 1.1249 onTargetScoringAttempt: 0.8471 -> 0.6504 bigChanceMissed: 0.5111 -> 0.3924 expectedGoalsOnTarget: 0.4101 -> 0.3149 goals: 0.3292 -> 0.2528 expectedGoals: 0.3123 -> 0.2398
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	expected_goals: 0.2398 -> 0.2915 expected_assists: 0.0722 -> 0.0878
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 79.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2529` `goal_rate=0.2528` `scorer_weight=0.3046`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0841` `assist_rate=0.2287` `provider_weight=0.1098`	-

Abdessamad Ezzalzouli F 82 1 0.32 0.24 ok 2 1.8 7.2 59.0 31 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0112, quality 0.796)	`market_value_eur=43000000` `quality_factor=0.7957` `boost_factor=1.0112` `reference_quality=0.505` `history_matches=34` `history_shrink=0.1282`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Gerard Martín, Pau Cubarsí', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.9121, 'skill_edge': -0.0278, 'height_edge': -0.032, 'attacker_skill': 0.7957, 'defender_skill': 0.8575, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Gerard Martín', 'market_value': 36000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8168}, {'player': 'Pau Cubarsí', 'market_value': 94000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8981}], 'xg_override': 0.2598}	totalShots: 2.233 -> 2.0367 shotOffTarget: 1.3492 -> 1.2306 onTargetScoringAttempt: 0.8837 -> 0.806 expectedGoalsOnTarget: 0.4842 -> 0.4416 goals: 0.3117 -> 0.2843 expectedGoals: 0.2848 -> 0.2598 bigChanceMissed: 0.2394 -> 0.2184
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	expected_goals: 0.2598 -> 0.3158 expected_assists: 0.2126 -> 0.2583
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 86.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.2708` `goal_rate=0.2843` `scorer_weight=0.3408`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.2276` `assist_rate=0.1321` `provider_weight=0.2066`	-

Isco M 68 0 0.17 0.21 ok 2 1.2 7.1 57.3 18 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9957, quality 0.445)	`market_value_eur=3600000` `quality_factor=0.4449` `boost_factor=0.9957` `reference_quality=0.505` `history_matches=16` `history_shrink=0.2381`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0134)	`blend_weight=0.0134` `history_matches=16` `season_overlap_discount=1.0` `per90_to_match_factor=0.7559` `prior_reliability=0.0867` `competition=La Liga`	fouls: 0.1064 -> 0.1362 totalCross: 3.141 -> 3.1535 onTargetScoringAttempt: 0.8152 -> 0.8269 totalShots: 2.2728 -> 2.2649
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Xavi Espart', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.34, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 0.7941, 'skill_edge': -0.144, 'height_edge': 0.004, 'attacker_skill': 0.4449, 'defender_skill': 0.765, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Xavi Espart', 'market_value': 26000000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.765}], 'xg_override': 0.1435}	aerialWon: 0.0 -> 0.9 totalShots: 2.2845 -> 1.8141 shotOffTarget: 1.4576 -> 1.1575 onTargetScoringAttempt: 0.8269 -> 0.6566 expectedGoalsOnTarget: 0.5649 -> 0.4486 bigChanceMissed: 0.2684 -> 0.2131 goals: 0.2027 -> 0.161 expectedGoals: 0.1807 -> 0.1435
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	expected_goals: 0.1435 -> 0.1744 expected_assists: 0.14 -> 0.1702
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 83.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.16` `goal_rate=0.161` `scorer_weight=0.1902`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1565` `assist_rate=0.1322` `provider_weight=0.1632`	-

Antony F 80 0 0.21 0.23 ok 2 2.9 7.2 61.6 38 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0086, quality 0.764)	`market_value_eur=37000000` `quality_factor=0.7641` `boost_factor=1.0086` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0144)	`blend_weight=0.0144` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.8851` `prior_reliability=0.2` `competition=La Liga`	totalShots: 2.6078 -> 2.6341 totalCross: 1.8774 -> 1.8886 onTargetScoringAttempt: 1.0219 -> 1.0285 wasFouled: 1.338 -> 1.3442
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Gerard Martín, Pau Cubarsí', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.8618, 'skill_edge': -0.042, 'height_edge': -0.052, 'attacker_skill': 0.7641, 'defender_skill': 0.8575, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Gerard Martín', 'market_value': 36000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8168}, {'player': 'Pau Cubarsí', 'market_value': 94000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8981}], 'xg_override': 0.1698}	totalShots: 2.6341 -> 2.2701 shotOffTarget: 1.5858 -> 1.3666 onTargetScoringAttempt: 1.0285 -> 0.8864 bigChanceMissed: 0.4034 -> 0.3477 expectedGoalsOnTarget: 0.2937 -> 0.2531 goals: 0.2272 -> 0.1958 expectedGoals: 0.197 -> 0.1698
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	expected_goals: 0.1698 -> 0.2064 expected_assists: 0.1224 -> 0.1488
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 81.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1865` `goal_rate=0.1958` `scorer_weight=0.2308`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1383` `assist_rate=0.0524` `provider_weight=0.119`	-

Facundo Bernal M 80 0 0.04 0.02 88% 1 0.6 6.8 47.6 10 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0017, quality 0.519)	`market_value_eur=9700000` `quality_factor=0.5189` `boost_factor=1.0017` `reference_quality=0.505` `history_matches=7` `history_shrink=0.4167`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0433)	`blend_weight=0.0433` `history_matches=7` `season_overlap_discount=1.0` `per90_to_match_factor=0.8868` `prior_reliability=0.16` `competition=La Liga`	fouls: 1.1403 -> 1.1069 totalShots: 0.7465 -> 0.7614 wonTackle: 0.6407 -> 0.6289 interceptionWon: 1.6268 -> 1.6364 onTargetScoringAttempt: 0.17 -> 0.1784 wasFouled: 0.8607 -> 0.8554
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Fermín López', 'matchup_zone': 'Holding Mid vs Central Mid — Midfield Anchor Duel', 'duel_win_rate': 0.282, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.8271, 'skill_edge': -0.1617, 'height_edge': 0.044, 'attacker_skill': 0.5389, 'defender_skill': 0.8981, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Fermín López', 'market_value': 113000000, 'role': 'CM', 'price_index': 0.9, 'adjusted_quality': 0.8981}], 'xg_override': 0.0319}	aerialWon: 0.5021 -> 1.2 totalShots: 0.7614 -> 0.6298 shotOffTarget: 0.5765 -> 0.4768 onTargetScoringAttempt: 0.1784 -> 0.1476 goals: 0.0483 -> 0.0399 expectedGoals: 0.0386 -> 0.0319
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	expected_goals: 0.0319 -> 0.0388 expected_assists: 0.0291 -> 0.0354
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 74.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0381` `goal_rate=0.0399` `scorer_weight=0.0426`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0348` `assist_rate=0.0379` `provider_weight=0.0393`	-

Pablo Fornals M 81 0 0.08 0.06 ok 1 0.9 7.0 59.2 32 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0000, quality 0.503)	`market_value_eur=8400000` `quality_factor=0.5035` `boost_factor=1.0` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0097)	`blend_weight=0.0097` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=0.9013` `prior_reliability=0.1467` `competition=La Liga`	totalShots: 1.201 -> 1.2127 totalCross: 3.075 -> 3.0849 wasFouled: 0.9458 -> 0.9366 interceptionWon: 0.8006 -> 0.8088 wonTackle: 0.2102 -> 0.2161
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Fermín López', 'matchup_zone': 'Holding Mid vs Central Mid — Midfield Anchor Duel', 'duel_win_rate': 0.238, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.7623, 'skill_edge': -0.1697, 'height_edge': 0.008, 'attacker_skill': 0.5211, 'defender_skill': 0.8981, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Fermín López', 'market_value': 113000000, 'role': 'CM', 'price_index': 0.9, 'adjusted_quality': 0.8981}], 'xg_override': 0.0656}	aerialWon: 0.3677 -> 1.0 totalShots: 1.2127 -> 0.9244 shotOffTarget: 0.829 -> 0.6319 onTargetScoringAttempt: 0.3684 -> 0.2808 goals: 0.1059 -> 0.0807 expectedGoals: 0.086 -> 0.0656 expectedGoalsOnTarget: 0.0624 -> 0.0476
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	expected_goals: 0.0656 -> 0.0797 expected_assists: 0.141 -> 0.1714
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 85.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0766` `goal_rate=0.0807` `scorer_weight=0.0901`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1575` `assist_rate=0.1083` `provider_weight=0.1421`	-

Valentín Gómez D 90 0 0.01 0.00 416% 0 0.1 6.8 38.2 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0013, quality 0.536)	`market_value_eur=11200000` `quality_factor=0.5365` `boost_factor=1.0013` `reference_quality=0.505` `history_matches=31` `history_shrink=0.1389`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0060)	`blend_weight=0.006` `history_matches=31` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.0667` `competition=La Liga`	wonTackle: 0.7676 -> 0.781 interceptionWon: 0.8752 -> 0.882
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Gabriel Jesus', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 76.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0147` `goal_rate=0.0142` `scorer_weight=0.0153`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0159` `assist_rate=0.0` `provider_weight=0.0128`	-

Diego Llorente D 81 0 0.02 0.01 103% 0 0.2 6.8 48.7 26 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9961, quality 0.437)	`market_value_eur=2600000` `quality_factor=0.4374` `boost_factor=0.9961` `reference_quality=0.505` `history_matches=21` `history_shrink=0.1923`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0083)	`blend_weight=0.0083` `history_matches=21` `season_overlap_discount=1.0` `per90_to_match_factor=0.9038` `prior_reliability=0.0667` `competition=La Liga`	interceptionWon: 1.1781 -> 1.206 wonTackle: 0.7887 -> 0.7821
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Gabriel Jesus', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 79.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0227` `goal_rate=0.0` `scorer_weight=0.0184`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0047` `assist_rate=0.0` `provider_weight=0.0044`	-

Héctor Bellerín D 88 0 0.09 0.03 181% 1 0.3 6.7 52.0 27 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9975, quality 0.437)	`market_value_eur=2500000` `quality_factor=0.4374` `boost_factor=0.9975` `reference_quality=0.505` `history_matches=35` `history_shrink=0.125`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0119)	`blend_weight=0.0119` `history_matches=35` `season_overlap_discount=1.0` `per90_to_match_factor=0.9743` `prior_reliability=0.1467` `competition=La Liga`	totalCross: 1.5259 -> 1.5394 interceptionWon: 0.9697 -> 0.9581 wonTackle: 1.2641 -> 1.2543 fouls: 0.7349 -> 0.7261 totalShots: 0.5385 -> 0.5321
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Karim Adeyemi', 'matchup_zone': 'Right Wing vs Left Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	expected_goals: 0.0712 -> 0.0865 expected_assists: 0.0866 -> 0.1053
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 16 by projected rating, fifa_overall 76.0)	`match_squad_rank=16` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0829` `goal_rate=0.0487` `scorer_weight=0.0704`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0999` `assist_rate=0.1912` `provider_weight=0.1316`	-

Álvaro Valles G 90 0 0.00 - n/a 0.0 7.4 60.0 32 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0003, quality 0.514)	`market_value_eur=9300000` `quality_factor=0.5142` `boost_factor=1.0003` `reference_quality=0.505` `history_matches=36` `history_shrink=0.122`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0159)	`blend_weight=0.0159` `history_matches=36` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.2` `competition=La Liga`	saves: 2.9961 -> 3.0014
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Lamine Yamal, Karim Adeyemi', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 3.3757}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	shots_faced: 32.1 -> 36.2 expected_saves: 7.86 -> 8.86 high_pressure_saves: 2.7 -> 3.05
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 77.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.5951`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=1.0141`	-

Marc Bartra D 74 0 0.07 0.03 123% 1 0.6 6.9 45.7 26 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9971, quality 0.437)	`market_value_eur=930000` `quality_factor=0.4374` `boost_factor=0.9971` `reference_quality=0.505` `history_matches=30` `history_shrink=0.1429`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0124)	`blend_weight=0.0124` `history_matches=30` `season_overlap_discount=1.0` `per90_to_match_factor=0.8259` `prior_reliability=0.1333` `competition=La Liga`	interceptionWon: 0.949 -> 0.9577 totalShots: 0.6391 -> 0.6465 totalCross: 0.3083 -> 0.3147
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Lamine Yamal', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.49 (squad xG sum 1.2283, xG scale 1.2155, xA scale 1.2155, team xA 1.08 = 0.72 x lambda)	`scoreline_reset=true` `team_lambda=1.493` `squad_xg_sum=1.2283` `squad_xa_sum=0.8898` `team_expected_assists=1.0815` `team_xa_to_xg_ratio=0.7244`	expected_goals: 0.0599 -> 0.0728
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 89.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0702` `goal_rate=0.1247` `scorer_weight=0.091`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0152` `assist_rate=0.0` `provider_weight=0.0122`	-

Team totals 1 1.49 14 7.0

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Diego Conde	G	4.4m	10.9%	Market-value substitute prior
Natan	D	14.7m	24.1%	Market-value substitute prior
Fran García	D	23.8m	35.1%	Market-value substitute prior
Sergi Altimira	M	22.7m	53.2%	Market-value substitute prior
Nelson Deossa	M	19.9m	57.0%	Market-value substitute prior
Troy Parrott	F	22.8m	54.4%	Market-value substitute prior

Barcelona — projected line vs season baseline (4-3-3)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Gerard Martín D 84 0 0.02 0.01 132% 0 0.2 7.2 55.0 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0086, quality 0.758)	`market_value_eur=36000000` `quality_factor=0.7581` `boost_factor=1.0086` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0098)	`blend_weight=0.0098` `history_matches=39` `season_overlap_discount=1.0` `per90_to_match_factor=0.9288` `prior_reliability=0.1333` `competition=La Liga`	fouls: 1.0428 -> 1.0371
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Abdessamad Ezzalzouli', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0121 -> 0.0196 expected_assists: 0.056 -> 0.1027
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 79.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0194` `goal_rate=0.0` `scorer_weight=0.0158`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.0976` `assist_rate=0.0` `provider_weight=0.0822`	-

Pau Cubarsí D 85 0 0.07 0.02 205% 0 0.2 7.2 52.0 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0141, quality 0.891)	`market_value_eur=94000000` `quality_factor=0.891` `boost_factor=1.0141` `reference_quality=0.505` `history_matches=36` `history_shrink=0.122`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0106)	`blend_weight=0.0106` `history_matches=36` `season_overlap_discount=1.0` `per90_to_match_factor=0.9471` `prior_reliability=0.1333` `competition=La Liga`	wonTackle: 0.6904 -> 0.6981 interceptionWon: 1.1594 -> 1.1522 totalShots: 0.4118 -> 0.4175
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Abdessamad Ezzalzouli', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0427 -> 0.0691 expected_assists: 0.0202 -> 0.037
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 86.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0668` `goal_rate=0.0113` `scorer_weight=0.0553`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.0363` `assist_rate=0.0` `provider_weight=0.0296`	-

Jules Koundé D 87 0 0.29 0.08 272% 1 0.4 7.1 48.4 26 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0129, quality 0.848)	`market_value_eur=58000000` `quality_factor=0.8478` `boost_factor=1.0129` `reference_quality=0.505` `history_matches=35` `history_shrink=0.125`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0065)	`blend_weight=0.0065` `history_matches=35` `season_overlap_discount=1.0` `per90_to_match_factor=0.9701` `prior_reliability=0.08` `competition=La Liga`	totalCross: 1.3659 -> 1.357 interceptionWon: 0.9004 -> 0.8945 wonTackle: 0.8873 -> 0.8815
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Cucho Hernández', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1762 -> 0.2853 expected_assists: 0.1259 -> 0.2308
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 85.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2482` `goal_rate=0.1575` `scorer_weight=0.2617`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.2061` `assist_rate=0.1129` `provider_weight=0.2137`	-

Joan García G 85 0 0.00 - n/a 0.0 7.1 60.6 30 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0109, quality 0.809)	`market_value_eur=46000000` `quality_factor=0.8089` `boost_factor=1.0109` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0155)	`blend_weight=0.0155` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.9393` `prior_reliability=0.2` `competition=La Liga`	saves: 2.8607 -> 2.8407
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Abdessamad Ezzalzouli, Cucho Hernández', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.4843}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	shots_faced: 14.1 -> 14.2 expected_saves: 3.45 -> 3.48
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 86.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=4.1327`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=2.6527`	-

Xavi Espart D 89 0 0.06 0.01 477% 1 0.3 6.8 27.0 23 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0181, quality 0.686)	`market_value_eur=26000000` `quality_factor=0.6858` `boost_factor=1.0181` `reference_quality=0.505` `history_matches=10` `history_shrink=0.3333`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0433)	`blend_weight=0.0433` `history_matches=10` `season_overlap_discount=1.0` `per90_to_match_factor=0.9838` `prior_reliability=0.2` `competition=La Liga`	wasFouled: 0.9993 -> 0.9702 totalCross: 0.7355 -> 0.7605 interceptionWon: 1.1647 -> 1.1853 fouls: 1.4571 -> 1.4366 goalAssist: 0.2377 -> 0.2558 goals: 0.1911 -> 0.1828 totalShots: 0.8253 -> 0.8181
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Isco', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0347 -> 0.0562 expected_assists: 0.091 -> 0.1669
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 15 by projected rating, fifa_overall 67.0)	`match_squad_rank=15` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0546` `goal_rate=0.1828` `scorer_weight=0.0703`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.1537` `assist_rate=0.2558` `provider_weight=0.2086`	-

Rodri M 81 0 0.19 0.07 166% 1 0.7 7.5 67.0 25 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0151, quality 0.837)	`market_value_eur=54000000` `quality_factor=0.837` `boost_factor=1.0151` `reference_quality=0.505` `history_matches=28` `history_shrink=0.1515`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0039)	`blend_weight=0.0039` `history_matches=28` `season_overlap_discount=1.0` `per90_to_match_factor=0.8983` `prior_reliability=0.04` `competition=La Liga`	wonTackle: 1.7648 -> 1.7578 interceptionWon: 1.3309 -> 1.3257 wasFouled: 1.29 -> 1.2849
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Facundo Bernal', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.58, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.2359, 'skill_edge': 0.1445, 'height_edge': 0.016, 'attacker_skill': 0.8599, 'defender_skill': 0.5389, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Facundo Bernal', 'market_value': 9700000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.5389}], 'xg_override': 0.1183}	aerialWon: 1.7904 -> 1.3 totalShots: 0.9272 -> 1.1459 shotOffTarget: 0.7101 -> 0.8776 onTargetScoringAttempt: 0.2171 -> 0.2683 bigChanceMissed: 0.1332 -> 0.1646 expectedGoals: 0.0957 -> 0.1183 goals: 0.0726 -> 0.0897 expectedGoalsOnTarget: 0.0213 -> 0.0263
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1183 -> 0.1916 expected_assists: 0.1417 -> 0.2598
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 90.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1744` `goal_rate=0.0897` `scorer_weight=0.159`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.2288` `assist_rate=0.0865` `provider_weight=0.2078`	-

Fermín López M 80 0 0.44 0.27 64% 3 1.9 7.2 63.9 28 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0140, quality 0.896)	`market_value_eur=113000000` `quality_factor=0.8965` `boost_factor=1.014` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0119)	`blend_weight=0.0119` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.888` `prior_reliability=0.1533` `competition=La Liga`	onTargetScoringAttempt: 0.9624 -> 0.9738 goals: 0.2536 -> 0.2643 wasFouled: 0.9765 -> 0.9832 wonTackle: 0.6872 -> 0.6928
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Facundo Bernal', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel', 'duel_win_rate': 0.568, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.1729, 'skill_edge': 0.1617, 'height_edge': -0.044, 'attacker_skill': 0.8981, 'defender_skill': 0.5389, 'attacker_role': 'CM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'Facundo Bernal', 'market_value': 9700000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.5389}], 'xg_override': 0.2734}	totalShots: 2.2531 -> 2.6427 aerialWon: 0.5712 -> 0.9 shotOffTarget: 1.2793 -> 1.5005 onTargetScoringAttempt: 0.9738 -> 1.1422 expectedGoalsOnTarget: 0.566 -> 0.6639 goals: 0.2643 -> 0.31 expectedGoals: 0.2331 -> 0.2734 bigChanceMissed: 0.2213 -> 0.2596
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2734 -> 0.4427 expected_assists: 0.1614 -> 0.2959
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 90.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.3577` `goal_rate=0.31` `scorer_weight=0.4928`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.2561` `assist_rate=0.156` `provider_weight=0.2884`	-

Pedri M 80 0 0.15 0.05 225% 1 0.6 7.2 64.9 26 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0141, quality 0.900)	`market_value_eur=164000000` `quality_factor=0.8997` `boost_factor=1.0141` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0119)	`blend_weight=0.0119` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.8863` `prior_reliability=0.1533` `competition=La Liga`	wonTackle: 0.9158 -> 0.9232
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Facundo Bernal', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel', 'duel_win_rate': 0.56, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.1624, 'skill_edge': 0.1624, 'height_edge': -0.052, 'attacker_skill': 0.8999, 'defender_skill': 0.5389, 'attacker_role': 'CM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'Facundo Bernal', 'market_value': 9700000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.5389}], 'xg_override': 0.0932}	aerialWon: 0.1272 -> 0.9 totalShots: 0.9867 -> 1.1469 shotOffTarget: 0.7188 -> 0.8355 onTargetScoringAttempt: 0.2673 -> 0.3107 goals: 0.0987 -> 0.1147 expectedGoals: 0.0802 -> 0.0932 expectedGoalsOnTarget: 0.067 -> 0.0779 bigChanceMissed: 0.0311 -> 0.0362
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0932 -> 0.1509 expected_assists: 0.175 -> 0.3209
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 90.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1401` `goal_rate=0.1147` `scorer_weight=0.1736`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=1` `assist_probability=0.2745` `assist_rate=0.1911` `provider_weight=0.3439`	-

Karim Adeyemi F 79 1 0.67 0.43 54% 3 1.5 7.1 58.5 11 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0324, quality 0.786)	`market_value_eur=41000000` `quality_factor=0.7859` `boost_factor=1.0324` `reference_quality=0.505` `history_matches=8` `history_shrink=0.3846`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0167)	`blend_weight=0.0167` `history_matches=8` `season_overlap_discount=1.0` `per90_to_match_factor=0.8807` `prior_reliability=0.0667` `competition=La Liga`	onTargetScoringAttempt: 0.8415 -> 0.872 goalAssist: 0.2483 -> 0.2588 wasFouled: 1.1548 -> 1.1649 goals: 0.3087 -> 0.3182 fouls: 0.3879 -> 0.3961
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Héctor Bellerín', 'matchup_zone': 'Right Wing vs Left Back — Flank Battle', 'duel_win_rate': 0.642, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.2376, 'skill_edge': 0.1536, 'height_edge': 0.008, 'attacker_skill': 0.7859, 'defender_skill': 0.4445, 'attacker_role': 'RW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Héctor Bellerín', 'market_value': 2500000, 'role': 'LB', 'price_index': 0.7, 'adjusted_quality': 0.4445}], 'xg_override': 0.4114}	totalShots: 2.4178 -> 2.9923 shotOffTarget: 1.5458 -> 1.9131 onTargetScoringAttempt: 0.872 -> 1.0792 bigChanceMissed: 0.817 -> 1.0111 expectedGoals: 0.3324 -> 0.4114 goals: 0.3182 -> 0.3938 expectedGoalsOnTarget: 0.3039 -> 0.3761
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.4114 -> 0.6662 expected_assists: 0.1465 -> 0.2686
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 82.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.4863` `goal_rate=0.3938` `scorer_weight=0.6408`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.2356` `assist_rate=0.2588` `provider_weight=0.3357`	-

Gabriel Jesus F 74 0 0.54 0.11 377% 2 0.7 6.8 36.1 26 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0066, quality 0.607)	`market_value_eur=17600000` `quality_factor=0.6068` `boost_factor=1.0066` `reference_quality=0.505` `history_matches=18` `history_shrink=0.2174`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Valentín Gómez, Diego Llorente', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.211, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.0165, 'skill_edge': 0.0432, 'height_edge': -0.032, 'attacker_skill': 0.6068, 'defender_skill': 0.5108, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Valentín Gómez', 'market_value': 11200000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5785}, {'player': 'Diego Llorente', 'market_value': 2600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4432}], 'xg_override': 0.3326}	totalShots: 2.0466 -> 2.0804 bigChanceMissed: 1.4372 -> 1.4609 shotOffTarget: 1.1159 -> 1.1343 onTargetScoringAttempt: 0.9306 -> 0.946 goals: 0.349 -> 0.3548 expectedGoals: 0.3272 -> 0.3326
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.3326 -> 0.5386 expected_assists: 0.0652 -> 0.1195
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 16 by projected rating, fifa_overall 79.0)	`match_squad_rank=16` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.4164` `goal_rate=0.3548` `scorer_weight=0.5698`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.1126` `assist_rate=0.0` `provider_weight=0.0956`	-

Lamine Yamal F 84 2 1.22 0.49 149% 7 3.6 8.2 65.5 30 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0145, quality 0.900)	`market_value_eur=237000000` `quality_factor=0.9` `boost_factor=1.0145` `reference_quality=0.505` `history_matches=36` `history_shrink=0.122`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0143)	`blend_weight=0.0143` `history_matches=36` `season_overlap_discount=1.0` `per90_to_match_factor=0.9296` `prior_reliability=0.18` `competition=La Liga`	totalOffside: 0.7768 -> 0.7952 goalAssist: 0.7365 -> 0.726 fouls: 1.1745 -> 1.1676 interceptionWon: 0.2152 -> 0.222 totalCross: 3.3521 -> 3.3583
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Marc Bartra', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.688, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.306, 'skill_edge': 0.2082, 'height_edge': 0.0, 'attacker_skill': 0.9, 'defender_skill': 0.4374, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Marc Bartra', 'market_value': 930000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.4374}], 'xg_override': 0.7541}	totalShots: 5.1418 -> 6.7152 shotOffTarget: 3.2684 -> 4.2685 onTargetScoringAttempt: 1.8711 -> 2.4437 goals: 0.8192 -> 1.0699 expectedGoalsOnTarget: 0.7478 -> 0.9766 expectedGoals: 0.5774 -> 0.7541 bigChanceMissed: 0.3811 -> 0.4977
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	goals: 1.0 -> 0 assists: 1.0 -> 0
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.81 (squad xG sum 2.3560, xG scale 1.6193, xA scale 1.8335, team xA 2.48 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=3.815` `squad_xg_sum=2.356` `squad_xa_sum=1.3524` `team_expected_assists=2.4798` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.7541 -> 1.2211 expected_assists: 0.285 -> 0.5227
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 90.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=2` `goal_probability=0.7051` `goal_rate=1.0699` `scorer_weight=1.5264`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=1` `assist_probability=0.4071` `assist_rate=0.726` `provider_weight=0.6534`	-

Team totals 3 3.81 21 7.2

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Dominik Livaković	G	4.5m	11.1%	Market-value substitute prior
Alejandro Balde	D	15.2m	37.1%	Market-value substitute prior
Eric García	D	15.0m	24.6%	Market-value substitute prior
Ferran Torres	M	23.6m	57.3%	Market-value substitute prior
Dani Olmo	M	25.5m	59.7%	Market-value substitute prior
Anthony Gordon	F	26.5m	49.8%	Market-value substitute prior

Antagonist / tactical summary

Away midfield dictates play with ~54% possession expected (Facundo Bernal vs Pedri). Total antagonist xG — Home: 0.91, Away: 1.98. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Left Wing vs Right Back — Flank Battle.

Midfield — possession 45.9% / 54.1%, dominance ratio 0.36, key duel: Facundo Bernal vs Pedri

Isco xG 0.14Abdessamad Ezzalzouli xG 0.26Cucho Hernández xG 0.24Lamine Yamal xG 0.75Karim Adeyemi xG 0.41Gabriel Jesus xG 0.33

Summary

Antagonist matchup analysis for Real Betis vs Barcelona. Most likely scoreline: 1-3. Lambdas derived from 6 home and 6 away individual player duels. Head-to-head: Real Betis 0W-2D-8L vs Barcelona over 10 meetings.

Tactical parity between 4-2-3-1 and 4-3-3. Match outcome driven primarily by individual player quality. Away midfield dictates play with ~54% possession expected (Facundo Bernal vs Pedri). Total antagonist xG — Home: 0.91, Away: 1.98. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Left Wing vs Right Back — Flank Battle.

Everton vs Chelsea

Premier League · 20261017 · 4-2-3-1 vs 3-4-2-1

Predicted 1-2Winner: ChelseaConfidence 0.72λ home/away 1.78 / 2.00Most-likely P 8.1%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Everton vs Chelsea. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 3-4-2-1 (duel lambdas: Home 1.789, Away 1.982). Final calibrated Poisson lambdas: Home 1.776, Away 1.996 (most likely score 1-2).

Home Starters 11/11 XGBoost (9 FBref) Away Starters 11/11 XGBoost (9 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.79 / 1.98 Final λ (H/A) 1.78 / 2.00

FBref Team Priors: Everton (Gls/90: 1.5, GA/90: 1.5 → E[λ]: 3.5) vs Chelsea (Gls/90: 3.5, GA/90: 3.5 → E[λ]: 3.5) | Blend Weight: 0.192

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Everton	Chelsea
League position	5.00	6.00
Points per game	1.50	1.50
Attack index	0.97	1.23
Defense (concede) index	1.16	0.85
Form multiplier	1.00	1.00
Standing λ multiplier	1.05	1.02

Lambda decomposition (how the goal rates were built)

Stage	Everton λ	Chelsea λ
FBref Squad Prior E[λ] (weight 0.192)	3.500	3.500
Pre-competitive-floor	1.776	1.996
Pre-calibration (Micro-duels)	1.789	1.982
Final λ	1.776	1.996

Scoreline distribution & markets

Top 5 scorelines	P
1-1	8.2%
1-2	8.1%
2-1	7.2%
2-2	7.2%
1-3	5.4%

Home win	35.0%
Draw	21.4%
Away win	43.7%
Over 2.5	72.7%
BTTS yes	71.8%

Pregame fair line — favorite Chelsea (43.7%), fair total 3.77, fair spread -0.22

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Everton — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Thierno Barry F 84 1 0.55 0.30 85% 2 1.4 6.8 54.1 31 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0077, quality 0.758)	`market_value_eur=36000000` `quality_factor=0.7581` `boost_factor=1.0077` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0071)	`blend_weight=0.0071` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=0.9283` `prior_reliability=0.1067` `competition=Premier League`	fouls: 1.0376 -> 1.0303
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Maxence Lacroix, Josh Acheampong', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.184, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.9761, 'skill_edge': -0.0323, 'height_edge': 0.016, 'attacker_skill': 0.7581, 'defender_skill': 0.8299, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Maxence Lacroix', 'market_value': 64000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8862}, {'player': 'Josh Acheampong', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}], 'xg_override': 0.3889}	totalShots: 2.5438 -> 2.483 bigChanceMissed: 1.4368 -> 1.4025 shotOffTarget: 1.4239 -> 1.3899 onTargetScoringAttempt: 1.1199 -> 1.0931 goals: 0.51 -> 0.4978 expectedGoalsOnTarget: 0.4266 -> 0.4164 expectedGoals: 0.3984 -> 0.3889
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.3889 -> 0.5481 expected_assists: 0.0383 -> 0.0521
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 76.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.422` `goal_rate=0.4978` `scorer_weight=0.6841`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0508` `assist_rate=0.0` `provider_weight=0.0417`	-

Jack Grealish F 86 0 0.11 0.01 969% 1 0.4 7.1 35.5 24 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0068, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0068` `reference_quality=0.505` `history_matches=25` `history_shrink=0.1667`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0007)	`blend_weight=0.0007` `history_matches=25` `season_overlap_discount=1.0` `per90_to_match_factor=0.9584` `prior_reliability=0.0067` `competition=Premier League`	fouls: 0.8767 -> 0.883
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Maxence Lacroix, Josh Acheampong', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.8102, 'skill_edge': -0.0851, 'height_edge': -0.044, 'attacker_skill': 0.6408, 'defender_skill': 0.8299, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Maxence Lacroix', 'market_value': 64000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8862}, {'player': 'Josh Acheampong', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}], 'xg_override': 0.0755}	totalShots: 1.1826 -> 0.9581 shotOffTarget: 0.8614 -> 0.6979 onTargetScoringAttempt: 0.3212 -> 0.2602 bigChanceMissed: 0.1666 -> 0.135 goals: 0.1132 -> 0.0917 expectedGoals: 0.0932 -> 0.0755 expectedGoalsOnTarget: 0.0718 -> 0.0582
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0755 -> 0.1064 expected_assists: 0.1635 -> 0.2224
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 82.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1009` `goal_rate=0.0917` `scorer_weight=0.1201`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1994` `assist_rate=0.1951` `provider_weight=0.2553`	-

Kiernan Dewsbury-Hall M 89 0 0.23 0.15 55% 2 1.5 7.2 56.4 28 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0105, quality 0.800)	`market_value_eur=44000000` `quality_factor=0.8002` `boost_factor=1.0105` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0103)	`blend_weight=0.0103` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.986` `prior_reliability=0.1333` `competition=Premier League`	totalCross: 3.7027 -> 3.7103 wonTackle: 0.9281 -> 0.9338
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Maxence Lacroix', 'matchup_zone': 'LW vs CB', 'duel_win_rate': 0.325, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.8608, 'skill_edge': -0.0387, 'height_edge': -0.056, 'attacker_skill': 0.8002, 'defender_skill': 0.8862, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Maxence Lacroix', 'market_value': 64000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8862}], 'xg_override': 0.1628}	aerialWon: 0.4079 -> 1.0 totalShots: 2.287 -> 1.9686 shotOffTarget: 1.4316 -> 1.2323 onTargetScoringAttempt: 0.8549 -> 0.7359 bigChanceMissed: 0.3083 -> 0.2654 expectedGoalsOnTarget: 0.2435 -> 0.2096 goals: 0.2312 -> 0.199 expectedGoals: 0.1891 -> 0.1628
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.1628 -> 0.2294 expected_assists: 0.1705 -> 0.232
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 81.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.205` `goal_rate=0.199` `scorer_weight=0.2684`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.2071` `assist_rate=0.1343` `provider_weight=0.1939`	-

Brennan Johnson F 73 0 0.23 0.10 127% 1 0.7 6.7 50.3 35 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0063, quality 0.694)	`market_value_eur=27000000` `quality_factor=0.6941` `boost_factor=1.0063` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0019)	`blend_weight=0.0019` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.8109` `prior_reliability=0.0267` `competition=Premier League`	fouls: 0.8839 -> 0.89 totalCross: 3.0212 -> 3.0154
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Maxence Lacroix, Josh Acheampong', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.8808, 'skill_edge': -0.0611, 'height_edge': -0.02, 'attacker_skill': 0.6941, 'defender_skill': 0.8299, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Maxence Lacroix', 'market_value': 64000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8862}, {'player': 'Josh Acheampong', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}], 'xg_override': 0.1622}	totalShots: 1.3687 -> 1.2056 shotOffTarget: 0.8792 -> 0.7744 onTargetScoringAttempt: 0.4895 -> 0.4312 bigChanceMissed: 0.2056 -> 0.1811 expectedGoals: 0.1841 -> 0.1622 goals: 0.1841 -> 0.1622
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.1622 -> 0.2286 expected_assists: 0.0555 -> 0.0755
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 16 by projected rating, fifa_overall 78.0)	`match_squad_rank=16` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2044` `goal_rate=0.1622` `scorer_weight=0.2286`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0727` `assist_rate=0.0259` `provider_weight=0.0604`	-

Harrison Armstrong M 83 0 0.14 0.04 231% 1 0.4 6.8 39.0 29 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0137, quality 0.725)	`market_value_eur=31000000` `quality_factor=0.7249` `boost_factor=1.0137` `reference_quality=0.505` `history_matches=19` `history_shrink=0.2083`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0172)	`blend_weight=0.0172` `history_matches=19` `season_overlap_discount=1.0` `per90_to_match_factor=0.9256` `prior_reliability=0.1267` `competition=Premier League`	interceptionWon: 0.5497 -> 0.5737 fouls: 1.1741 -> 1.1957 totalShots: 1.1004 -> 1.0815 goalAssist: 0.0658 -> 0.073 onTargetScoringAttempt: 0.3689 -> 0.3626 wonTackle: 1.1414 -> 1.1469
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Pedro Neto, João Pedro, Moisés Caicedo', 'matchup_zone': 'Shadow Striker vs CDM + CB — Zone 14 Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0968 -> 0.1364 expected_assists: 0.0683 -> 0.0929
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 73.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1275` `goal_rate=0.096` `scorer_weight=0.1357`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0887` `assist_rate=0.073` `provider_weight=0.0966`	-

James Garner M 87 0 0.09 0.03 206% 1 0.6 7.2 58.3 31 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0101, quality 0.834)	`market_value_eur=53000000` `quality_factor=0.834` `boost_factor=1.0101` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0035)	`blend_weight=0.0035` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=0.9619` `prior_reliability=0.0533` `competition=Premier League`	wonTackle: 1.7668 -> 1.7776 fouls: 0.9488 -> 0.9539
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Morgan Rogers', 'matchup_zone': 'Midfield lane'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0626 -> 0.0882 expected_assists: 0.1373 -> 0.1868
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 82.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0844` `goal_rate=0.0599` `scorer_weight=0.0861`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1704` `assist_rate=0.0901` `provider_weight=0.1494`	-

Jarrad Branthwaite D 90 0 0.05 0.03 107% 1 0.5 6.8 50.9 22 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0217, quality 0.809)	`market_value_eur=46000000` `quality_factor=0.8089` `boost_factor=1.0217` `reference_quality=0.505` `history_matches=16` `history_shrink=0.2381`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0206)	`blend_weight=0.0206` `history_matches=16` `season_overlap_discount=1.0` `per90_to_match_factor=0.9948` `prior_reliability=0.1333` `competition=Premier League`	wasFouled: 0.7779 -> 0.7619 totalCross: 0.4421 -> 0.433 interceptionWon: 1.2145 -> 1.2202
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Pedro Neto', 'matchup_zone': 'Shadow Striker vs CDM + CB — Zone 14 Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0373 -> 0.0526
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 79.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0512` `goal_rate=0.0451` `scorer_weight=0.0573`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0127` `assist_rate=0.0` `provider_weight=0.0108`	-

Michael Keane D 90 0 0.10 0.04 128% 1 0.5 6.9 37.2 29 FBref Prior D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9974, quality 0.437)	`market_value_eur=2800000` `quality_factor=0.4374` `boost_factor=0.9974` `reference_quality=0.505` `history_matches=34` `history_shrink=0.1282`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'João Pedro', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0682 -> 0.0961 expected_assists: 0.0388 -> 0.0528
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 76.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0916` `goal_rate=0.0359` `scorer_weight=0.0769`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0514` `assist_rate=0.1089` `provider_weight=0.066`	-

James Tarkowski D 90 0 0.06 0.07 ok 1 0.7 7.0 58.0 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9987, quality 0.465)	`market_value_eur=5200000` `quality_factor=0.4646` `boost_factor=0.9987` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0090)	`blend_weight=0.009` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	interceptionWon: 1.1494 -> 1.1661 wasFouled: 1.3704 -> 1.3806
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Cole Palmer', 'matchup_zone': 'Right Wingback vs Left Back — Overlapping Run'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0447 -> 0.063 expected_assists: 0.0386 -> 0.0525
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 80.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0611` `goal_rate=0.1545` `scorer_weight=0.0788`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0511` `assist_rate=0.1971` `provider_weight=0.0656`	-

Ainsley Maitland-Niles D 84 0 0.01 0.01 47% 1 0.3 6.8 32.8 11 XGBoost D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0047, quality 0.533)	`market_value_eur=10900000` `quality_factor=0.533` `boost_factor=1.0047` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Jordan Henderson', 'matchup_zone': 'Left Wingback vs Right Back — Overlapping Run'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.2624 -> 0.357
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 77.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0108` `goal_rate=0.0` `scorer_weight=0.0094`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.3002` `assist_rate=0.3905` `provider_weight=0.4462`	-

Jordan Pickford G 90 0 0.00 - n/a 0.0 7.1 56.3 31 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0019, quality 0.567)	`market_value_eur=13900000` `quality_factor=0.5671` `boost_factor=1.0019` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0088)	`blend_weight=0.0088` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	wasFouled: 0.3851 -> 0.395 saves: 2.9185 -> 2.9281
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'João Pedro, Cole Palmer', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.9825}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.78 (squad xG sum 1.2602, xG scale 1.4093, xA scale 1.3604, team xA 1.42 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.776` `squad_xg_sum=1.2602` `squad_xa_sum=1.0444` `team_expected_assists=1.4208` `team_xa_to_xg_ratio=0.8`	shots_faced: 18.8 -> 19.0 expected_saves: 4.6 -> 4.65 high_pressure_saves: 1.59 -> 1.6
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 85.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=2.0083`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=1.4683`	-

Team totals 1 1.78 15 6.9

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Mark Travers	G	4.4m	10.9%	Market-value substitute prior
Vitaliy Mykolenko	D	14.6m	23.7%	Market-value substitute prior
Jake O'Brien	D	14.2m	32.7%	Market-value substitute prior
Hayden Hackney	M	26.4m	52.5%	Market-value substitute prior
Merlin Röhl	M	22.7m	32.1%	Market-value substitute prior
Tyrique George	F	26.3m	35.4%	Market-value substitute prior

Chelsea — projected line vs season baseline (3-4-2-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Maxence Lacroix D 88 0 0.07 0.06 ok 1 0.6 6.8 52.5 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0116, quality 0.861)	`market_value_eur=64000000` `quality_factor=0.8608` `boost_factor=1.0116` `reference_quality=0.505` `history_matches=41` `history_shrink=0.1087`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0094)	`blend_weight=0.0094` `history_matches=41` `season_overlap_discount=1.0` `per90_to_match_factor=0.973` `prior_reliability=0.1333` `competition=Premier League`	wonTackle: 0.7533 -> 0.7645
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Kiernan Dewsbury-Hall', 'matchup_zone': 'LW vs CB'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.05 -> 0.0738 expected_assists: 0.0274 -> 0.0404
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 82.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0711` `goal_rate=0.0193` `scorer_weight=0.059`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0396` `assist_rate=0.0851` `provider_weight=0.0505`	-

Josh Acheampong D 81 0 0.04 0.00 5501% 0 0.0 6.6 33.3 29 XGBoost D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0114, quality 0.710)	`market_value_eur=29000000` `quality_factor=0.71` `boost_factor=1.0114` `reference_quality=0.505` `history_matches=22` `history_shrink=0.1852`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Thierno Barry', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0287 -> 0.0423
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 17 by projected rating, fifa_overall 75.0)	`match_squad_rank=17` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0414` `goal_rate=0.0515` `scorer_weight=0.0529`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0149` `assist_rate=0.0` `provider_weight=0.0125`	-

Emiliano Martínez G 90 0 0.00 - n/a 0.0 7.1 56.1 28 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0005, quality 0.518)	`market_value_eur=9600000` `quality_factor=0.5177` `boost_factor=1.0005` `reference_quality=0.505` `history_matches=37` `history_shrink=0.119`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0052)	`blend_weight=0.0052` `history_matches=37` `season_overlap_discount=1.0` `per90_to_match_factor=0.9972` `prior_reliability=0.0667` `competition=Premier League`	wasFouled: 0.4528 -> 0.4608
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Thierno Barry, Kiernan Dewsbury-Hall', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.7894}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	shots_faced: 17.0 -> 16.9 expected_saves: 4.16 -> 4.14 high_pressure_saves: 1.43 -> 1.42
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 85.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=2.225`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=1.3374`	-

Levi Colwill D 88 0 0.00 0.00 175% 0 0.1 6.9 35.5 15 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0386, quality 0.840)	`market_value_eur=55000000` `quality_factor=0.8399` `boost_factor=1.0386` `reference_quality=0.505` `history_matches=8` `history_shrink=0.3846`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0300)	`blend_weight=0.03` `history_matches=8` `season_overlap_discount=1.0` `per90_to_match_factor=0.9789` `prior_reliability=0.12` `competition=Premier League`	wonTackle: 0.6421 -> 0.6718 wasFouled: 0.3079 -> 0.315 totalShots: 0.3545 -> 0.3603
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Jack Grealish', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0198 -> 0.0292
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 80.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0024` `goal_rate=0.0` `scorer_weight=0.0023`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0288` `assist_rate=0.0` `provider_weight=0.0234`	-

Pedro Neto F 82 0 0.17 0.09 93% 2 0.9 6.8 47.3 37 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0114, quality 0.848)	`market_value_eur=58000000` `quality_factor=0.8478` `boost_factor=1.0114` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0043)	`blend_weight=0.0043` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.9123` `prior_reliability=0.06` `competition=Premier League`	totalCross: 4.5318 -> 4.5253 totalShots: 1.5149 -> 1.5213 onTargetScoringAttempt: 0.5671 -> 0.5733
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Harrison Armstrong, Jarrad Branthwaite', 'matchup_zone': 'Shadow Striker vs CDM + CB — Zone 14 Battle', 'duel_win_rate': 0.228, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.938, 'skill_edge': 0.0178, 'height_edge': -0.06, 'attacker_skill': 0.8478, 'defender_skill': 0.8082, 'attacker_role': 'SS', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Harrison Armstrong', 'market_value': 31000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.761}, {'player': 'Jarrad Branthwaite', 'market_value': 46000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8555}], 'xg_override': 0.1135}	totalShots: 1.5213 -> 1.427 shotOffTarget: 0.9478 -> 0.889 onTargetScoringAttempt: 0.5733 -> 0.5378 bigChanceMissed: 0.2736 -> 0.2566 goals: 0.1439 -> 0.135 expectedGoals: 0.121 -> 0.1135 expectedGoalsOnTarget: 0.1195 -> 0.1121
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1135 -> 0.1675 expected_assists: 0.1285 -> 0.1896
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 81.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1542` `goal_rate=0.135` `scorer_weight=0.1895`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1727` `assist_rate=0.3076` `provider_weight=0.237`	-

Moisés Caicedo M 84 0 0.06 0.03 84% 1 0.6 7.0 54.3 26 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0146, quality 0.895)	`market_value_eur=104000000` `quality_factor=0.8945` `boost_factor=1.0146` `reference_quality=0.505` `history_matches=35` `history_shrink=0.125`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0011)	`blend_weight=0.0011` `history_matches=35` `season_overlap_discount=1.0` `per90_to_match_factor=0.9382` `prior_reliability=0.0133` `competition=Premier League`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Harrison Armstrong', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel', 'duel_win_rate': 0.483, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 1.0488, 'skill_edge': 0.0612, 'height_edge': -0.028, 'attacker_skill': 0.8969, 'defender_skill': 0.761, 'attacker_role': 'CM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'Harrison Armstrong', 'market_value': 31000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.761}], 'xg_override': 0.0379}	aerialWon: 0.3695 -> 1.0 totalShots: 0.5714 -> 0.5993 shotOffTarget: 0.4522 -> 0.4743 onTargetScoringAttempt: 0.1192 -> 0.125
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0379 -> 0.0559 expected_assists: 0.0504 -> 0.0744
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 86.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0544` `goal_rate=0.044` `scorer_weight=0.0598`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0717` `assist_rate=0.0454` `provider_weight=0.0707`	-

Jordan Henderson M 80 0 0.07 0.01 949% 1 0.1 7.1 50.7 23 XGBoost M_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9974, quality 0.437)	`market_value_eur=1300000` `quality_factor=0.4374` `boost_factor=0.9974` `reference_quality=0.505` `history_matches=34` `history_shrink=0.1282`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Ainsley Maitland-Niles', 'matchup_zone': 'Left Wingback vs Right Back — Overlapping Run', 'duel_win_rate': 0.353, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 0.9315, 'skill_edge': -0.0666, 'height_edge': 0.02, 'attacker_skill': 0.4374, 'defender_skill': 0.5854, 'attacker_role': 'LWB', 'attacker_mv_price_index': 0.7, 'defender_adjusted_parts': [{'player': 'Ainsley Maitland-Niles', 'market_value': 10900000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.5854}], 'xg_override': 0.0443}	aerialWon: 0.7888 -> 1.1 totalShots: 0.7331 -> 0.6829 shotOffTarget: 0.5314 -> 0.495 onTargetScoringAttempt: 0.2017 -> 0.1879 expectedGoalsOnTarget: 0.0935 -> 0.0871
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0443 -> 0.0654 expected_assists: 0.1023 -> 0.1509
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 78.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0633` `goal_rate=0.0651` `scorer_weight=0.0798`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1401` `assist_rate=0.0965` `provider_weight=0.1452`	-

João Pedro F 85 1 0.58 0.42 41% 3 2.0 7.1 57.4 27 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0128, quality 0.890)	`market_value_eur=92000000` `quality_factor=0.89` `boost_factor=1.0128` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0096)	`blend_weight=0.0096` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.943` `prior_reliability=0.1333` `competition=Premier League`	totalShots: 2.4268 -> 2.4443 totalCross: 0.6328 -> 0.6267 goalAssist: 0.3302 -> 0.3361 onTargetScoringAttempt: 0.8183 -> 0.824 goals: 0.424 -> 0.429
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Harrison Armstrong, Jarrad Branthwaite', 'matchup_zone': 'Shadow Striker vs CDM + CB — Zone 14 Battle', 'duel_win_rate': 0.291, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 1.0306, 'skill_edge': 0.0368, 'height_edge': -0.016, 'attacker_skill': 0.89, 'defender_skill': 0.8082, 'attacker_role': 'SS', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Harrison Armstrong', 'market_value': 31000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.761}, {'player': 'Jarrad Branthwaite', 'market_value': 46000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8555}], 'xg_override': 0.3965}	totalShots: 2.4443 -> 2.5191 shotOffTarget: 1.6085 -> 1.6577 onTargetScoringAttempt: 0.824 -> 0.8492 expectedGoalsOnTarget: 0.5761 -> 0.5937 goals: 0.429 -> 0.4421 expectedGoals: 0.3847 -> 0.3965 bigChanceMissed: 0.3061 -> 0.3155
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.3965 -> 0.5848 expected_assists: 0.0615 -> 0.0907
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 83.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.4428` `goal_rate=0.4421` `scorer_weight=0.6445`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0867` `assist_rate=0.3361` `provider_weight=0.1134`	-

Morgan Rogers M 88 0 0.38 0.21 78% 3 2.3 7.2 59.1 35 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0122, quality 0.897)	`market_value_eur=115000000` `quality_factor=0.8968` `boost_factor=1.0122` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0086)	`blend_weight=0.0086` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.9734` `prior_reliability=0.1267` `competition=Premier League`	totalShots: 2.4401 -> 2.4456
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Harrison Armstrong', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel', 'duel_win_rate': 0.52, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 1.1026, 'skill_edge': 0.0618, 'height_edge': 0.008, 'attacker_skill': 0.8983, 'defender_skill': 0.761, 'attacker_role': 'CM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'Harrison Armstrong', 'market_value': 31000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.761}], 'xg_override': 0.2555}	totalShots: 2.4456 -> 2.6965 shotOffTarget: 1.5084 -> 1.6632 aerialWon: 1.3118 -> 1.2 onTargetScoringAttempt: 0.9281 -> 1.0233 goals: 0.2803 -> 0.3091 expectedGoalsOnTarget: 0.2703 -> 0.298 expectedGoals: 0.2317 -> 0.2555 bigChanceMissed: 0.1622 -> 0.1788
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2555 -> 0.377 expected_assists: 0.1535 -> 0.2265
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 84.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.3141` `goal_rate=0.3091` `scorer_weight=0.4432`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.2027` `assist_rate=0.1189` `provider_weight=0.188`	-

Cole Palmer M 88 1 0.42 0.31 37% 3 2.3 7.1 59.5 34 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0158, quality 0.894)	`market_value_eur=101000000` `quality_factor=0.8936` `boost_factor=1.0158` `reference_quality=0.505` `history_matches=32` `history_shrink=0.1351`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0111)	`blend_weight=0.0111` `history_matches=32` `season_overlap_discount=1.0` `per90_to_match_factor=0.9779` `prior_reliability=0.1267` `competition=Premier League`	totalShots: 2.3358 -> 2.3611 goals: 0.258 -> 0.2666 onTargetScoringAttempt: 0.8167 -> 0.8247 totalCross: 0.9982 -> 1.0043
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'James Tarkowski', 'matchup_zone': 'Right Wingback vs Left Back — Overlapping Run', 'duel_win_rate': 0.583, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.2695, 'skill_edge': 0.1834, 'height_edge': 0.0, 'attacker_skill': 0.8993, 'defender_skill': 0.4918, 'attacker_role': 'RWB', 'attacker_mv_price_index': 0.7, 'defender_adjusted_parts': [{'player': 'James Tarkowski', 'market_value': 5200000, 'role': 'LB', 'price_index': 0.7, 'adjusted_quality': 0.4918}], 'xg_override': 0.2874}	aerialWon: 0.0801 -> 1.2 totalShots: 2.3611 -> 2.9974 shotOffTarget: 1.519 -> 1.9284 onTargetScoringAttempt: 0.8247 -> 1.047 goals: 0.2666 -> 0.3384 expectedGoals: 0.2264 -> 0.2874 bigChanceMissed: 0.2079 -> 0.2639 expectedGoalsOnTarget: 0.1781 -> 0.2261
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2874 -> 0.424 expected_assists: 0.1539 -> 0.2271
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 85.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.3456` `goal_rate=0.3384` `scorer_weight=0.4881`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.2032` `assist_rate=0.1364` `provider_weight=0.2076`	-

Pep Chavarría D 89 0 0.02 0.01 138% 0 0.3 6.7 47.2 34 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0034, quality 0.606)	`market_value_eur=17500000` `quality_factor=0.6058` `boost_factor=1.0034` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0025)	`blend_weight=0.0025` `history_matches=39` `season_overlap_discount=1.0` `per90_to_match_factor=0.9897` `prior_reliability=0.0333` `competition=Premier League`	totalCross: 3.5282 -> 3.5195
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Thierno Barry', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 2 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=2`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 2.00 (squad xG sum 1.3529, xG scale 1.4753, xA scale 1.4753, team xA 1.30 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.996` `squad_xg_sum=1.3529` `squad_xa_sum=0.8794` `team_expected_assists=1.2975` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0157 -> 0.0232 expected_assists: 0.1193 -> 0.176
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 14 by projected rating, fifa_overall 83.0)	`match_squad_rank=14` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 2 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=2` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0229` `goal_rate=0.0023` `scorer_weight=0.0186`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1614` `assist_rate=0.1589` `provider_weight=0.2172`	-

Team totals 2 2.00 16 6.9

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Mike Penders	G	5.3m	15.0%	Market-value substitute prior
Reece James	D	19.5m	44.4%	Market-value substitute prior
Jorrel Hato	D	15.1m	24.7%	Market-value substitute prior
Marco Palestra	M	23.5m	56.6%	Market-value substitute prior
Valentín Barco	M	15.5m	72.1%	Market-value substitute prior
Estêvão	F	27.1m	63.9%	Market-value substitute prior

Antagonist / tactical summary

Midfield is evenly contested — James Garner vs Morgan Rogers will be the pivotal duel. Total antagonist xG — Home: 0.79, Away: 1.14. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Shadow Striker vs CDM + CB — Zone 14 Battle.

Midfield — possession 50.1% / 49.9%, dominance ratio 0.49, key duel: James Garner vs Morgan Rogers

Kiernan Dewsbury-Hall xG 0.16Thierno Barry xG 0.39Jack Grealish xG 0.08Pedro Neto xG 0.11João Pedro xG 0.40Jordan Henderson xG 0.04

Summary

Antagonist matchup analysis for Everton vs Chelsea. Most likely scoreline: 1-2. Lambdas derived from 4 home and 6 away individual player duels. Head-to-head: Everton 3W-3D-4L vs Chelsea over 10 meetings.

Tactical parity between 4-2-3-1 and 3-4-2-1. Match outcome driven primarily by individual player quality. Midfield is evenly contested — James Garner vs Morgan Rogers will be the pivotal duel. Total antagonist xG — Home: 0.79, Away: 1.14. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Shadow Striker vs CDM + CB — Zone 14 Battle.

Villarreal vs Elche

LaLiga · 20261017 · 4-4-2 vs 4-2-3-1

Predicted 1-0Winner: VillarrealWinning H 1-0 (11.3%)Winning A 0-1 (5.2%)Confidence 0.77λ home/away 1.95 / 0.90Most-likely P 11.3%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Villarreal vs Elche. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-4-2 vs 4-2-3-1 (duel lambdas: Home 1.839, Away 0.989). Final calibrated Poisson lambdas: Home 1.948, Away 0.898 (most likely score 1-0).

Home Starters 11/11 XGBoost (8 FBref) Away Starters 11/11 XGBoost (8 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.84 / 0.99 Final λ (H/A) 1.95 / 0.90

FBref Team Priors: Villarreal (Gls/90: 1.3333333333333333, GA/90: 1.33 → E[λ]: 0.988) vs Elche (Gls/90: 1.0, GA/90: 1.0 → E[λ]: 0.985) | Blend Weight: 0.228

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Villarreal	Elche
League position	11.00	18.00
Points per game	1.16	1.01
Attack index	1.02	0.96
Defense (concede) index	1.15	1.02
Form multiplier	1.01	0.99
Standing λ multiplier	1.07	0.99

Lambda decomposition (how the goal rates were built)

Stage	Villarreal λ	Elche λ
FBref Squad Prior E[λ] (weight 0.228)	0.988	0.985
Pre-mismatch	1.878	0.968
Pre-competitive-floor	2.048	0.944
Pre-calibration (Micro-duels)	1.839	0.989
Final λ	1.948	0.898

Scoreline distribution & markets

Top 5 scorelines	P
1-0	11.3%
2-0	11.0%
1-1	10.2%
2-1	9.9%
3-0	7.2%

Home win	61.9%
Draw	21.5%
Away win	16.6%
Over 2.5	54.1%
BTTS yes	50.9%

Pregame fair line — favorite Villarreal (61.9%), fair total 2.85, fair spread 1.05

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Villarreal — projected line vs season baseline (4-4-2)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Tani Oluwaseyi F 71 0 0.28 0.01 2978% 1 0.1 6.7 32.1 32 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9995, quality 0.493)	`market_value_eur=7500000` `quality_factor=0.4927` `boost_factor=0.9995` `reference_quality=0.505` `history_matches=31` `history_shrink=0.1389`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0012)	`blend_weight=0.0012` `history_matches=31` `season_overlap_discount=1.0` `per90_to_match_factor=0.7882` `prior_reliability=0.0133` `competition=La Liga`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Buba Sangaré, Roy Revivo', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.224, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.0358, 'skill_edge': 0.0064, 'height_edge': 0.018, 'attacker_skill': 0.4927, 'defender_skill': 0.4785, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Buba Sangaré', 'market_value': 4900000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4809}, {'player': 'Roy Revivo', 'market_value': 4600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4761}], 'xg_override': 0.18}	totalShots: 1.4021 -> 1.4523 shotOffTarget: 0.9668 -> 1.0014 onTargetScoringAttempt: 0.4353 -> 0.4509 bigChanceMissed: 0.3955 -> 0.4097 expectedGoals: 0.1738 -> 0.18
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.18 -> 0.2796 expected_assists: 0.0431 -> 0.067
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 18 by projected rating, fifa_overall 71.0)	`match_squad_rank=18` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2439` `goal_rate=0.1408` `scorer_weight=0.232`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0648` `assist_rate=0.1197` `provider_weight=0.0838`	-

Ayoze Pérez F 71 1 0.49 0.25 96% 2 1.2 6.8 55.4 29 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9985, quality 0.467)	`market_value_eur=5400000` `quality_factor=0.4671` `boost_factor=0.9985` `reference_quality=0.505` `history_matches=33` `history_shrink=0.1316`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0091)	`blend_weight=0.0091` `history_matches=33` `season_overlap_discount=1.0` `per90_to_match_factor=0.7898` `prior_reliability=0.1067` `competition=La Liga`	wasFouled: 1.5266 -> 1.5352 totalShots: 2.1357 -> 2.1432 onTargetScoringAttempt: 0.8798 -> 0.8852
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Buba Sangaré, Roy Revivo', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.181, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9719, 'skill_edge': -0.0051, 'height_edge': -0.014, 'attacker_skill': 0.4671, 'defender_skill': 0.4785, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Buba Sangaré', 'market_value': 4900000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4809}, {'player': 'Roy Revivo', 'market_value': 4600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4761}], 'xg_override': 0.3179}	totalShots: 2.1432 -> 2.083 shotOffTarget: 1.2559 -> 1.2206 bigChanceMissed: 1.027 -> 0.9981 onTargetScoringAttempt: 0.8852 -> 0.8603 goals: 0.4035 -> 0.3922 expectedGoals: 0.3271 -> 0.3179 expectedGoalsOnTarget: 0.2786 -> 0.2708
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.3179 -> 0.4939 expected_assists: 0.0633 -> 0.0983
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 79.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.3898` `goal_rate=0.3922` `scorer_weight=0.5937`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0936` `assist_rate=0.304` `provider_weight=0.1229`	-

Alberto Moleiro F 79 0 0.45 0.13 256% 3 1.0 7.3 59.6 41 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0100, quality 0.837)	`market_value_eur=54000000` `quality_factor=0.837` `boost_factor=1.01` `reference_quality=0.505` `history_matches=45` `history_shrink=0.1`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0069)	`blend_weight=0.0069` `history_matches=45` `season_overlap_discount=1.0` `per90_to_match_factor=0.8746` `prior_reliability=0.1067` `competition=La Liga`	totalShots: 2.421 -> 2.4042 wonTackle: 0.5668 -> 0.5742 onTargetScoringAttempt: 0.826 -> 0.8203
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Buba Sangaré, Roy Revivo', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.315, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.1695, 'skill_edge': 0.1613, 'height_edge': -0.046, 'attacker_skill': 0.837, 'defender_skill': 0.4785, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Buba Sangaré', 'market_value': 4900000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4809}, {'player': 'Roy Revivo', 'market_value': 4600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4761}], 'xg_override': 0.2871}	totalShots: 2.4153 -> 2.8247 shotOffTarget: 1.595 -> 1.8654 onTargetScoringAttempt: 0.8203 -> 0.9593 expectedGoalsOnTarget: 0.3666 -> 0.4287 goals: 0.2646 -> 0.3094 expectedGoals: 0.2455 -> 0.2871 bigChanceMissed: 0.138 -> 0.1614
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.2871 -> 0.446 expected_assists: 0.2051 -> 0.3187
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 79.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.3598` `goal_rate=0.3094` `scorer_weight=0.4755`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.2729` `assist_rate=0.3856` `provider_weight=0.3984`	-

Santi Comesaña M 71 0 0.09 0.08 ok 1 0.4 6.9 60.4 29 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0000, quality 0.503)	`market_value_eur=8400000` `quality_factor=0.5035` `boost_factor=1.0` `reference_quality=0.505` `history_matches=41` `history_shrink=0.1087`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0104)	`blend_weight=0.0104` `history_matches=41` `season_overlap_discount=1.0` `per90_to_match_factor=0.7852` `prior_reliability=0.1467` `competition=La Liga`	wasFouled: 1.0404 -> 1.0333 totalShots: 0.6257 -> 0.6192 fouls: 1.5567 -> 1.5628 totalCross: 0.545 -> 0.5394 wonTackle: 0.9166 -> 0.9219
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Matías Dituro', 'matchup_zone': 'Left Mid vs Right Back — Wide Channel', 'duel_win_rate': 0.439, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.0275, 'skill_edge': 0.0347, 'height_edge': -0.016, 'attacker_skill': 0.5146, 'defender_skill': 0.4374, 'attacker_role': 'LM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'Matías Dituro', 'market_value': 105000, 'role': 'GK', 'price_index': 0.55, 'adjusted_quality': 0.4374}], 'xg_override': 0.0586}	aerialWon: 0.2604 -> 1.2 totalShots: 0.6238 -> 0.641 shotOffTarget: 0.4426 -> 0.4548 onTargetScoringAttempt: 0.1812 -> 0.1862
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.0586 -> 0.091 expected_assists: 0.0832 -> 0.1292
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 79.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.087` `goal_rate=0.055` `scorer_weight=0.088`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1212` `assist_rate=0.0652` `provider_weight=0.1117`	-

Pape Gueye M 79 0 0.22 0.15 52% 2 1.2 7.1 60.0 30 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0094, quality 0.775)	`market_value_eur=39000000` `quality_factor=0.7754` `boost_factor=1.0094` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0151)	`blend_weight=0.0151` `history_matches=38` `season_overlap_discount=1.0` `per90_to_match_factor=0.8826` `prior_reliability=0.2` `competition=La Liga`	totalShots: 1.5557 -> 1.5455 totalCross: 0.6715 -> 0.6658 fouls: 0.8143 -> 0.8198
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Javi Morcillo', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel', 'duel_win_rate': 0.628, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.2615, 'skill_edge': 0.1619, 'height_edge': 0.016, 'attacker_skill': 0.7972, 'defender_skill': 0.4374, 'attacker_role': 'CM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'Javi Morcillo', 'market_value': 2100000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.4374}], 'xg_override': 0.1423}	aerialWon: 0.3941 -> 1.2 totalShots: 1.5512 -> 1.9569 shotOffTarget: 0.9674 -> 1.2204 onTargetScoringAttempt: 0.5838 -> 0.7365 goals: 0.1394 -> 0.1759 expectedGoals: 0.1128 -> 0.1423 expectedGoalsOnTarget: 0.0843 -> 0.1063 bigChanceMissed: 0.0558 -> 0.0704
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.1423 -> 0.221 expected_assists: 0.1008 -> 0.1566
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 80.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1983` `goal_rate=0.1759` `scorer_weight=0.2596`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.145` `assist_rate=0.0985` `provider_weight=0.1542`	-

Ilias Akhomach F 72 0 0.17 0.06 200% 1 0.9 7.0 50.7 27 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0014, quality 0.535)	`market_value_eur=11100000` `quality_factor=0.5353` `boost_factor=1.0014` `reference_quality=0.505` `history_matches=27` `history_shrink=0.1562`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0034)	`blend_weight=0.0034` `history_matches=27` `season_overlap_discount=1.0` `per90_to_match_factor=0.805` `prior_reliability=0.0333` `competition=La Liga`	totalCross: 2.6905 -> 2.725 interceptionWon: 0.1183 -> 0.1233
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Buba Sangaré, Roy Revivo', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.192, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9876, 'skill_edge': 0.0256, 'height_edge': -0.034, 'attacker_skill': 0.5353, 'defender_skill': 0.4785, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Buba Sangaré', 'market_value': 4900000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4809}, {'player': 'Roy Revivo', 'market_value': 4600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4761}], 'xg_override': 0.1114}	totalShots: 1.3451 -> 1.3285 shotOffTarget: 0.8737 -> 0.8629 onTargetScoringAttempt: 0.4714 -> 0.4656
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.1114 -> 0.173 expected_assists: 0.1268 -> 0.197
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 75.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1589` `goal_rate=0.179` `scorer_weight=0.2162`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1788` `assist_rate=0.2745` `provider_weight=0.2463`	-

Sergi Cardona D 85 0 0.05 0.01 266% 1 0.4 6.7 44.6 33 FBref Prior D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0000, quality 0.506)	`market_value_eur=8600000` `quality_factor=0.5059` `boost_factor=1.0` `reference_quality=0.505` `history_matches=35` `history_shrink=0.125`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Lucas Cepeda', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.0295 -> 0.0458 expected_assists: 0.0972 -> 0.151
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 16 by projected rating, fifa_overall 79.0)	`match_squad_rank=16` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0448` `goal_rate=0.0491` `scorer_weight=0.0571`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1402` `assist_rate=0.0113` `provider_weight=0.1208`	-

Renato Veiga D 90 0 0.03 0.01 197% 1 0.2 7.1 57.5 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0068, quality 0.710)	`market_value_eur=29000000` `quality_factor=0.71` `boost_factor=1.0068` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0144)	`blend_weight=0.0144` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.9983` `prior_reliability=0.2` `competition=La Liga`	wonTackle: 0.5786 -> 0.5702
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Fernando Niño', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.0186 -> 0.0289 expected_assists: 0.0895 -> 0.139
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 78.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0285` `goal_rate=0.0` `scorer_weight=0.0231`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1298` `assist_rate=0.0917` `provider_weight=0.1412`	-

Juan Foyth D 79 0 0.07 0.02 197% 1 0.3 6.9 48.4 13 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0001, quality 0.506)	`market_value_eur=8600000` `quality_factor=0.5059` `boost_factor=1.0001` `reference_quality=0.505` `history_matches=17` `history_shrink=0.2273`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0295)	`blend_weight=0.0295` `history_matches=17` `season_overlap_discount=1.0` `per90_to_match_factor=0.8741` `prior_reliability=0.2` `competition=La Liga`	interceptionWon: 1.1627 -> 1.1886 totalShots: 0.6608 -> 0.6756 wonTackle: 0.9672 -> 0.9817 fouls: 0.7009 -> 0.6888 wasFouled: 0.3366 -> 0.3439
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Thomas Lemar', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.0473 -> 0.0735 expected_assists: 0.049 -> 0.0761
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 79.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0709` `goal_rate=0.0` `scorer_weight=0.0588`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0733` `assist_rate=0.0` `provider_weight=0.0609`	-

Alexander Freeman D 90 0 0.04 0.01 554% 1 0.1 6.9 35.9 29 XGBoost D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0044, quality 0.566)	`market_value_eur=13800000` `quality_factor=0.566` `boost_factor=1.0044` `reference_quality=0.505` `history_matches=16` `history_shrink=0.2381`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Fernando Niño', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	expected_goals: 0.0281 -> 0.0437 expected_assists: 0.0335 -> 0.052
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 81.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0428` `goal_rate=0.0` `scorer_weight=0.035`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0507` `assist_rate=0.0` `provider_weight=0.0416`	-

Péter Gulácsi G 90 0 0.00 - n/a 0.0 7.1 27.2 11 FBref Prior G_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9887, quality 0.437)	`market_value_eur=970000` `quality_factor=0.4374` `boost_factor=0.9887` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Fernando Niño, Lucas Cepeda', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 0.9885}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.95 (squad xG sum 1.2540, xG scale 1.5534, xA scale 1.5534, team xA 1.43 = 0.73 x lambda)	`scoreline_reset=true` `team_lambda=1.948` `squad_xg_sum=1.254` `squad_xa_sum=0.9174` `team_expected_assists=1.4251` `team_xa_to_xg_ratio=0.7316`	shots_faced: 9.4 -> 8.5 expected_saves: 2.3 -> 2.08 high_pressure_saves: 0.79 -> 0.72
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 82.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=2.1033`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=1.5258`	-

Team totals 1 1.95 14 7.0

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Luiz Júnior	G	4.8m	12.9%	Market-value substitute prior
Carlos Romero	D	14.5m	23.6%	Market-value substitute prior
Santiago Mouriño	D	14.3m	23.2%	Market-value substitute prior
Carlos Maciá	M	23.3m	70.3%	Market-value substitute prior
Nathan-Dylan Saliba	M	22.5m	52.2%	Market-value substitute prior
Georges Mikautadze	F	26.7m	36.3%	Market-value substitute prior

Elche — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Matías Dituro G 90 0 0.00 - n/a 0.0 7.0 52.8 32 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9971, quality 0.437)	`market_value_eur=105000` `quality_factor=0.4374` `boost_factor=0.9971` `reference_quality=0.505` `history_matches=30` `history_shrink=0.1429`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0186)	`blend_weight=0.0186` `history_matches=30` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.2` `competition=La Liga`	saves: 2.8878 -> 2.8775
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Ayoze Pérez, Alberto Moleiro', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.8387}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	shots_faced: 17.5 -> 18.5 expected_saves: 4.29 -> 4.53 high_pressure_saves: 1.47 -> 1.56
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 76.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.0138`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.7026`	-

Buba Sangaré D 72 0 0.01 0.01 ok 0 0.1 6.6 51.1 24 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9972, quality 0.461)	`market_value_eur=4900000` `quality_factor=0.4609` `boost_factor=0.9972` `reference_quality=0.505` `history_matches=19` `history_shrink=0.2083`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0262)	`blend_weight=0.0262` `history_matches=19` `season_overlap_discount=1.0` `per90_to_match_factor=0.8056` `prior_reliability=0.1933` `competition=La Liga`	totalCross: 1.152 -> 1.1364 wonTackle: 1.1178 -> 1.1322 interceptionWon: 0.6736 -> 0.6632 goals: 0.0856 -> 0.0906
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Tani Oluwaseyi', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 18 by projected rating, fifa_overall 69.0)	`match_squad_rank=18` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0053` `goal_rate=0.0905` `scorer_weight=0.0066`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0145` `assist_rate=0.0` `provider_weight=0.0117`	-

Javi Morcillo M 86 0 0.02 0.01 62% 0 0.1 6.7 36.5 20 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9937, quality 0.437)	`market_value_eur=2100000` `quality_factor=0.4374` `boost_factor=0.9937` `reference_quality=0.505` `history_matches=11` `history_shrink=0.3125`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0054)	`blend_weight=0.0054` `history_matches=11` `season_overlap_discount=1.0` `per90_to_match_factor=0.9571` `prior_reliability=0.0267` `competition=La Liga`	fouls: 0.8538 -> 0.8621 interceptionWon: 1.4397 -> 1.4319
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Pape Gueye', 'matchup_zone': 'Holding Mid vs Central Mid — Midfield Anchor Duel', 'duel_win_rate': 0.222, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.7385, 'skill_edge': -0.1619, 'height_edge': -0.016, 'attacker_skill': 0.4374, 'defender_skill': 0.7972, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Pape Gueye', 'market_value': 39000000, 'role': 'CM', 'price_index': 0.9, 'adjusted_quality': 0.7972}], 'xg_override': 0.0217}	aerialWon: 0.8001 -> 1.1 totalShots: 0.5173 -> 0.3821 shotOffTarget: 0.3967 -> 0.293 onTargetScoringAttempt: 0.1206 -> 0.0891 goals: 0.0433 -> 0.032 expectedGoals: 0.0294 -> 0.0217
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.0284 -> 0.0232
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 13 by projected rating, fifa_overall 79.0)	`match_squad_rank=13` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0215` `goal_rate=0.032` `scorer_weight=0.0248`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0229` `assist_rate=0.0362` `provider_weight=0.0255`	-

Víctor Chust D 89 0 0.05 0.02 102% 1 0.4 7.0 61.2 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9977, quality 0.437)	`market_value_eur=2700000` `quality_factor=0.4374` `boost_factor=0.9977` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0148)	`blend_weight=0.0148` `history_matches=39` `season_overlap_discount=1.0` `per90_to_match_factor=0.9941` `prior_reliability=0.2` `competition=La Liga`	totalCross: 1.1006 -> 1.1137 wonTackle: 0.9657 -> 0.971
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Alberto Moleiro', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.0414 -> 0.0338
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 74.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0442` `goal_rate=0.0413` `scorer_weight=0.0433`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0332` `assist_rate=0.0775` `provider_weight=0.0422`	-

Fernando Niño F 75 0 0.32 0.27 ok 2 1.1 6.9 51.9 11 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9931, quality 0.445)	`market_value_eur=3600000` `quality_factor=0.4449` `boost_factor=0.9931` `reference_quality=0.505` `history_matches=8` `history_shrink=0.3846`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0417)	`blend_weight=0.0417` `history_matches=8` `season_overlap_discount=1.0` `per90_to_match_factor=0.8302` `prior_reliability=0.1667` `competition=La Liga`	totalShots: 2.2673 -> 2.2278 fouls: 0.8197 -> 0.7994 totalOffside: 0.1347 -> 0.1291
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Renato Veiga, Alexander Freeman', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 0.8409, 'skill_edge': -0.1123, 'height_edge': 0.004, 'attacker_skill': 0.4449, 'defender_skill': 0.6943, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Renato Veiga', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}, {'player': 'Alexander Freeman', 'market_value': 13800000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.615}], 'xg_override': 0.3157}	totalShots: 2.2684 -> 1.9075 shotOffTarget: 1.3047 -> 1.0971 onTargetScoringAttempt: 0.9637 -> 0.8104 expectedGoalsOnTarget: 0.6743 -> 0.567 goals: 0.4299 -> 0.3615 expectedGoals: 0.3754 -> 0.3157 bigChanceMissed: 0.1971 -> 0.1657
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.0433 -> 0.0353
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 81.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.2709` `goal_rate=0.3615` `scorer_weight=0.3555`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0347` `assist_rate=0.0` `provider_weight=0.0282`	-

Germán Valera F 84 0 0.09 0.04 128% 1 0.5 6.8 56.3 41 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9991, quality 0.477)	`market_value_eur=6200000` `quality_factor=0.4769` `boost_factor=0.9991` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Renato Veiga, Alexander Freeman', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 0.768, 'skill_edge': -0.0979, 'height_edge': -0.06, 'attacker_skill': 0.4769, 'defender_skill': 0.6943, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Renato Veiga', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}, {'player': 'Alexander Freeman', 'market_value': 13800000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.615}], 'xg_override': 0.0879}	totalShots: 1.051 -> 0.8072 shotOffTarget: 0.6126 -> 0.4705 onTargetScoringAttempt: 0.4383 -> 0.3366 goals: 0.156 -> 0.1198 expectedGoals: 0.1145 -> 0.0879 bigChanceMissed: 0.0788 -> 0.0605 expectedGoalsOnTarget: 0.0361 -> 0.0277
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1399 -> 0.1142
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 82.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0841` `goal_rate=0.1198` `scorer_weight=0.1082`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.1079` `assist_rate=0.0484` `provider_weight=0.0914`	-

Thomas Lemar M 73 0 0.07 0.07 ok 1 0.6 7.0 58.8 28 XGBoost M_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9973, quality 0.437)	`market_value_eur=1900000` `quality_factor=0.4374` `boost_factor=0.9973` `reference_quality=0.505` `history_matches=32` `history_shrink=0.1351`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Juan Foyth', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.37, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 0.8381, 'skill_edge': -0.0502, 'height_edge': -0.06, 'attacker_skill': 0.4374, 'defender_skill': 0.5489, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Juan Foyth', 'market_value': 8600000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.5489}], 'xg_override': 0.0748}	aerialWon: 0.0282 -> 0.8 totalShots: 1.024 -> 0.8582 shotOffTarget: 0.6488 -> 0.5438 onTargetScoringAttempt: 0.3751 -> 0.3144 goals: 0.111 -> 0.093 expectedGoalsOnTarget: 0.1022 -> 0.0857 expectedGoals: 0.0893 -> 0.0748 bigChanceMissed: 0.0374 -> 0.0313
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1178 -> 0.0961
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 77.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0721` `goal_rate=0.093` `scorer_weight=0.0857`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0916` `assist_rate=0.0922` `provider_weight=0.0814`	-

Lucas Cepeda F 80 0 0.10 0.05 103% 1 0.8 6.9 42.2 29 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9956, quality 0.437)	`market_value_eur=2600000` `quality_factor=0.4374` `boost_factor=0.9956` `reference_quality=0.505` `history_matches=18` `history_shrink=0.2174`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0085)	`blend_weight=0.0085` `history_matches=18` `season_overlap_discount=1.0` `per90_to_match_factor=0.8836` `prior_reliability=0.06` `competition=La Liga`	totalCross: 1.7003 -> 1.7358 totalShots: 1.386 -> 1.4164 fouls: 1.0589 -> 1.0749 interceptionWon: 0.4333 -> 0.4463 onTargetScoringAttempt: 0.4726 -> 0.4854 wasFouled: 2.4326 -> 2.4203 wonTackle: 1.0359 -> 1.0271 goalAssist: 0.0 -> 0.0083
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Renato Veiga, Alexander Freeman', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 0.7595, 'skill_edge': -0.1156, 'height_edge': -0.048, 'attacker_skill': 0.4374, 'defender_skill': 0.6943, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Renato Veiga', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}, {'player': 'Alexander Freeman', 'market_value': 13800000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.615}], 'xg_override': 0.0956}	totalShots: 1.4164 -> 1.0758 shotOffTarget: 0.9134 -> 0.6937 onTargetScoringAttempt: 0.4854 -> 0.3687 goals: 0.1604 -> 0.1218 expectedGoals: 0.1259 -> 0.0956 expectedGoalsOnTarget: 0.0306 -> 0.0232
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1047 -> 0.0855
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 71.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0912` `goal_rate=0.1218` `scorer_weight=0.1128`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0819` `assist_rate=0.0083` `provider_weight=0.0684`	-

Marc Aguado M 76 0 0.02 0.01 112% 0 0.2 6.8 54.4 30 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9980, quality 0.446)	`market_value_eur=3700000` `quality_factor=0.4461` `boost_factor=0.998` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0106)	`blend_weight=0.0106` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.8427` `prior_reliability=0.1467` `competition=La Liga`	interceptionWon: 1.047 -> 1.0359 fouls: 1.3075 -> 1.3018
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Pape Gueye', 'matchup_zone': 'Holding Mid vs Central Mid — Midfield Anchor Duel', 'duel_win_rate': 0.21, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.7202, 'skill_edge': -0.1544, 'height_edge': -0.036, 'attacker_skill': 0.4542, 'defender_skill': 0.7972, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Pape Gueye', 'market_value': 39000000, 'role': 'CM', 'price_index': 0.9, 'adjusted_quality': 0.7972}], 'xg_override': 0.0201}	aerialWon: 0.2122 -> 1.0 totalShots: 0.5041 -> 0.3631 shotOffTarget: 0.381 -> 0.2744 onTargetScoringAttempt: 0.1231 -> 0.0887 goals: 0.0511 -> 0.0368 expectedGoalsOnTarget: 0.0495 -> 0.0356 expectedGoals: 0.0279 -> 0.0201
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.0339 -> 0.0277
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 74.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0199` `goal_rate=0.0368` `scorer_weight=0.0249`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0273` `assist_rate=0.0393` `provider_weight=0.0295`	-

Gonzalo Villar M 79 0 0.05 0.02 161% 1 0.3 6.9 51.2 27 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9959, quality 0.437)	`market_value_eur=1700000` `quality_factor=0.4374` `boost_factor=0.9959` `reference_quality=0.505` `history_matches=20` `history_shrink=0.2`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0260)	`blend_weight=0.026` `history_matches=20` `season_overlap_discount=1.0` `per90_to_match_factor=0.8811` `prior_reliability=0.2` `competition=La Liga`	wasFouled: 2.1268 -> 2.1784 totalCross: 1.5387 -> 1.5674 fouls: 1.4076 -> 1.4321 onTargetScoringAttempt: 0.2438 -> 0.2604 totalShots: 0.7359 -> 0.7472 wonTackle: 0.8492 -> 0.8424 goalAssist: 0.0833 -> 0.0888
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Pape Gueye', 'matchup_zone': 'CAM vs CM', 'duel_win_rate': 0.234, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.7267, 'skill_edge': -0.1619, 'height_edge': -0.024, 'attacker_skill': 0.4374, 'defender_skill': 0.7972, 'attacker_role': 'CAM', 'attacker_mv_price_index': 0.95, 'defender_adjusted_parts': [{'player': 'Pape Gueye', 'market_value': 39000000, 'role': 'CM', 'price_index': 0.9, 'adjusted_quality': 0.7972}], 'xg_override': 0.0514}`	totalShots: 0.7525 -> 0.5468 aerialWon: 0.9217 -> 1.1 shotOffTarget: 0.4921 -> 0.3576 onTargetScoringAttempt: 0.2604 -> 0.1892 expectedGoalsOnTarget: 0.0928 -> 0.0674 goals: 0.0747 -> 0.0543 expectedGoals: 0.0707 -> 0.0514
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.0904 -> 0.0738
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 72.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0501` `goal_rate=0.0543` `scorer_weight=0.0529`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.0711` `assist_rate=0.0888` `provider_weight=0.073`	-

Roy Revivo D 88 0 0.01 0.02 36% 1 0.2 6.8 41.8 8 XGBoost D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9928, quality 0.457)	`market_value_eur=4600000` `quality_factor=0.4572` `boost_factor=0.9928` `reference_quality=0.505` `history_matches=5` `history_shrink=0.5`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Tani Oluwaseyi', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 0 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=0`	assists: 1.0 -> 0
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 0.90 (squad xG sum 0.8976, xG scale 1.0004, xA scale 0.8162, team xA 0.72 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=0.898` `squad_xg_sum=0.8976` `squad_xa_sum=0.8802` `team_expected_assists=0.7184` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1893 -> 0.1544
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 80.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation skipped: 0 projected goals	`team_goals=0` `allocation_applied=false` `goals_allocated=0` `goal_probability=0.0128` `goal_rate=0.0` `scorer_weight=0.0103`	-
`integer_assist_projection`	Team scoreline allocation skipped: 0 projected assists	`allocation_applied=false` `team_assists=0` `assists_allocated=0` `assist_probability=0.1431` `assist_rate=0.5297` `provider_weight=0.193`	-

Team totals 0 0.90 12 6.9

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Iñaki Peña	G	4.4m	11.0%	Market-value substitute prior
Kevin Lomónaco	D	13.8m	31.0%	Market-value substitute prior
Rubén Sánchez	D	13.6m	29.9%	Market-value substitute prior
Martim Neto	M	24.8m	50.0%	Market-value substitute prior
Federico Redondo	M	22.0m	30.7%	Market-value substitute prior
Facundo Buonanotte	F	26.4m	53.4%	Market-value substitute prior

Antagonist / tactical summary

Home midfield dominates with ~52% possession expected (Pape Gueye vs Marc Aguado). Total antagonist xG — Home: 1.10, Away: 0.67. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Midfield — possession 52.4% / 47.6%, dominance ratio 0.57, key duel: Pape Gueye vs Marc Aguado

Tani Oluwaseyi xG 0.18Alberto Moleiro xG 0.29Ayoze Pérez xG 0.32Thomas Lemar xG 0.07Gonzalo Villar xG 0.05Fernando Niño xG 0.32

Summary

Antagonist matchup analysis for Villarreal vs Elche. Most likely scoreline: 1-0. Lambdas derived from 6 home and 7 away individual player duels. Head-to-head: Villarreal 5W-3D-2L vs Elche over 10 meetings.

Tactical parity between 4-4-2 and 4-2-3-1. Match outcome driven primarily by individual player quality. Home midfield dominates with ~52% possession expected (Pape Gueye vs Marc Aguado). Total antagonist xG — Home: 1.10, Away: 0.67. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Fulham vs Hull

Premier League · 20261017 · 4-2-3-1 vs 5-4-1

Predicted 1-1Winner: DrawWinning H 1-0 (12.4%)Winning A 0-1 (9.9%)Confidence 0.7λ home/away 1.30 / 1.04Most-likely P 13.1%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Fulham vs Hull. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 5-4-1 (duel lambdas: Home 1.275, Away 1.063). Final calibrated Poisson lambdas: Home 1.302, Away 1.040 (most likely score 1-1).

Home Starters 11/11 XGBoost (10 FBref) Away Starters 11/11 XGBoost (7 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.27 / 1.06 Final λ (H/A) 1.30 / 1.04

FBref Team Priors: Fulham (Gls/90: 1.0, GA/90: 1.0 → E[λ]: 1.034) vs Hull City (Gls/90: 1.5, GA/90: 1.5 → E[λ]: 1.034) | Blend Weight: 0.192

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Fulham	Hull
League position	20.00	8.00
Points per game	0.84	1.42
Attack index	0.89	0.97
Defense (concede) index	0.89	1.04
Form multiplier	0.97	1.03
Standing λ multiplier	0.91	1.01

Lambda decomposition (how the goal rates were built)

Stage	Fulham λ	Hull λ
FBref Squad Prior E[λ] (weight 0.192)	1.034	1.034
Pre-competitive-floor	1.302	1.040
Pre-calibration (Micro-duels)	1.275	1.063
Final λ	1.302	1.040

Scoreline distribution & markets

Top 5 scorelines	P
1-1	13.1%
1-0	12.4%
0-1	9.9%
0-0	9.7%
2-1	8.5%

Home win	42.4%
Draw	28.0%
Away win	29.6%
Over 2.5	41.5%
BTTS yes	47.2%

Pregame fair line — favorite Fulham (42.4%), fair total 2.34, fair spread 0.26

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Fulham — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Bernd Leno G 90 0 0.00 - n/a 0.0 7.0 55.7 32 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9989, quality 0.468)	`market_value_eur=5500000` `quality_factor=0.4683` `boost_factor=0.9989` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0088)	`blend_weight=0.0088` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Mohamed Belloumi, Oli McBurnie', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.0634}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	shots_faced: 10.1 -> 9.9 expected_saves: 2.47 -> 2.42 high_pressure_saves: 0.85 -> 0.83
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 79.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.373`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.7281`	-

Timothy Castagne D 85 0 0.09 0.05 62% 0 0.5 6.7 48.2 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9996, quality 0.494)	`market_value_eur=7600000` `quality_factor=0.4939` `boost_factor=0.9996` `reference_quality=0.505` `history_matches=35` `history_shrink=0.125`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0092)	`blend_weight=0.0092` `history_matches=35` `season_overlap_discount=1.0` `per90_to_match_factor=0.9479` `prior_reliability=0.1133` `competition=Premier League`	fouls: 1.249 -> 1.2632 wasFouled: 0.7292 -> 0.743
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Mohamed Belloumi', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0829 -> 0.0889
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 75.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0836` `goal_rate=0.0` `scorer_weight=0.0698`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0851` `assist_rate=0.0841` `provider_weight=0.0897`	-

Calvin Bassey D 90 0 0.03 0.03 ok 1 0.4 6.9 49.7 28 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0075, quality 0.710)	`market_value_eur=29000000` `quality_factor=0.71` `boost_factor=1.0075` `reference_quality=0.505` `history_matches=36` `history_shrink=0.122`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0106)	`blend_weight=0.0106` `history_matches=36` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	fouls: 1.0563 -> 1.0504 wonTackle: 1.0228 -> 1.0173
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Mohamed Belloumi', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 78.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0296` `goal_rate=0.0081` `scorer_weight=0.024`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0177` `assist_rate=0.0` `provider_weight=0.0143`	-

David Affengruber D 89 0 0.05 0.03 78% 1 0.5 6.8 53.7 33 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0034, quality 0.612)	`market_value_eur=18100000` `quality_factor=0.612` `boost_factor=1.0034` `reference_quality=0.505` `history_matches=42` `history_shrink=0.1064`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0138)	`blend_weight=0.0138` `history_matches=42` `season_overlap_discount=1.0` `per90_to_match_factor=0.9859` `prior_reliability=0.2` `competition=La Liga`	totalShots: 0.5691 -> 0.584 wasFouled: 0.6264 -> 0.6359 fouls: 0.9973 -> 0.9881 interceptionWon: 1.2072 -> 1.2132
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Lewie Coyle', 'matchup_zone': 'RWB vs LW + LB'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 76.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0466` `goal_rate=0.0238` `scorer_weight=0.0382`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0411` `assist_rate=0.0212` `provider_weight=0.0336`	-

Gonzalo García F 82 1 0.36 0.22 63% 2 1.1 6.7 47.8 32 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0093, quality 0.758)	`market_value_eur=36000000` `quality_factor=0.7581` `boost_factor=1.0093` `reference_quality=0.505` `history_matches=36` `history_shrink=0.122`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0106)	`blend_weight=0.0106` `history_matches=36` `season_overlap_discount=1.0` `per90_to_match_factor=0.9086` `prior_reliability=0.1333` `competition=Premier League`	fouls: 0.8198 -> 0.8303 totalShots: 2.0291 -> 2.0365
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Lucas Herrington, Semi Ajayi', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.253, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.0775, 'skill_edge': 0.0967, 'height_edge': -0.044, 'attacker_skill': 0.7581, 'defender_skill': 0.5432, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Lucas Herrington', 'market_value': 16399999, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.649}, {'player': 'Semi Ajayi', 'market_value': 930000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}], 'xg_override': 0.3764}	totalShots: 2.0365 -> 2.1943 shotOffTarget: 1.1134 -> 1.1997 onTargetScoringAttempt: 0.9108 -> 0.9814 bigChanceMissed: 0.7898 -> 0.851 goals: 0.3846 -> 0.4144 expectedGoals: 0.3493 -> 0.3764 expectedGoalsOnTarget: 0.2209 -> 0.238
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.3764 -> 0.3629
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 78.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.3043` `goal_rate=0.4144` `scorer_weight=0.3952`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0577` `assist_rate=0.0` `provider_weight=0.0475`	-

Alex Iwobi M 82 0 0.08 0.05 51% 1 0.9 7.2 48.9 35 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0051, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0051` `reference_quality=0.505` `history_matches=35` `history_shrink=0.125`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0108)	`blend_weight=0.0108` `history_matches=35` `season_overlap_discount=1.0` `per90_to_match_factor=0.9111` `prior_reliability=0.1333` `competition=Premier League`	totalCross: 3.7467 -> 3.7308 interceptionWon: 0.4164 -> 0.4267 totalShots: 1.2942 -> 1.2851
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Lucas Herrington', 'matchup_zone': 'LW vs CB', 'duel_win_rate': 0.376, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 0.9358, 'skill_edge': -0.0037, 'height_edge': -0.04, 'attacker_skill': 0.6408, 'defender_skill': 0.649, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Lucas Herrington', 'market_value': 16399999, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.649}], 'xg_override': 0.0858}	aerialWon: 0.0981 -> 1.1 totalShots: 1.2898 -> 1.207 shotOffTarget: 0.8955 -> 0.838 onTargetScoringAttempt: 0.3943 -> 0.369 bigChanceMissed: 0.1366 -> 0.1278 goals: 0.102 -> 0.0955 expectedGoals: 0.0917 -> 0.0858
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.1777 -> 0.1906
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 80.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0794` `goal_rate=0.0955` `scorer_weight=0.0886`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1735` `assist_rate=0.1088` `provider_weight=0.1525`	-

Josh King M 77 0 0.18 0.10 87% 2 0.8 6.9 47.2 29 XGBoost M_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0098, quality 0.786)	`market_value_eur=41000000` `quality_factor=0.7859` `boost_factor=1.0098` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Tim Iroegbunam', 'matchup_zone': 'CAM vs CM', 'duel_win_rate': 0.467, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.0692, 'skill_edge': 0.0871, 'height_edge': -0.04, 'attacker_skill': 0.7964, 'defender_skill': 0.6029, 'attacker_role': 'CAM', 'attacker_mv_price_index': 0.95, 'defender_adjusted_parts': [{'player': 'Tim Iroegbunam', 'market_value': 15500000, 'role': 'CM', 'price_index': 0.9, 'adjusted_quality': 0.6029}], 'xg_override': 0.1913}	aerialWon: 0.1567 -> 0.8 totalShots: 1.976 -> 2.1128 shotOffTarget: 1.2248 -> 1.3096 onTargetScoringAttempt: 0.7512 -> 0.8032 bigChanceMissed: 0.269 -> 0.2876 expectedGoalsOnTarget: 0.2348 -> 0.251 goals: 0.1805 -> 0.193 expectedGoals: 0.1789 -> 0.1913
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1913 -> 0.1845 expected_assists: 0.1147 -> 0.123
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 78.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1685` `goal_rate=0.193` `scorer_weight=0.1858`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1157` `assist_rate=0.0764` `provider_weight=0.0984`	-

Oscar Bobb F 71 0 0.15 0.08 77% 1 0.5 6.9 47.6 37 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0083, quality 0.694)	`market_value_eur=27000000` `quality_factor=0.6941` `boost_factor=1.0083` `reference_quality=0.505` `history_matches=29` `history_shrink=0.1471`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0096)	`blend_weight=0.0096` `history_matches=29` `season_overlap_discount=1.0` `per90_to_match_factor=0.793` `prior_reliability=0.1` `competition=Premier League`	totalShots: 1.1841 -> 1.1728 fouls: 0.2579 -> 0.2655
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Lucas Herrington, Semi Ajayi', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.208, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.0116, 'skill_edge': 0.0679, 'height_edge': -0.06, 'attacker_skill': 0.6941, 'defender_skill': 0.5432, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Lucas Herrington', 'market_value': 16399999, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.649}, {'player': 'Semi Ajayi', 'market_value': 930000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.4374}], 'xg_override': 0.1561}	totalShots: 1.1796 -> 1.1933 shotOffTarget: 0.713 -> 0.7213 onTargetScoringAttempt: 0.4666 -> 0.472
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1561 -> 0.1505 expected_assists: 0.1035 -> 0.111
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 76.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1397` `goal_rate=0.1541` `scorer_weight=0.149`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1051` `assist_rate=0.0119` `provider_weight=0.0888`	-

Shea Charles M 83 0 0.03 0.01 193% 1 0.2 6.9 34.5 12 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0223, quality 0.669)	`market_value_eur=24000000` `quality_factor=0.6685` `boost_factor=1.0223` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0098)	`blend_weight=0.0098` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.9253` `prior_reliability=0.0333` `competition=Premier League`	totalShots: 0.5483 -> 0.5778 totalCross: 0.2726 -> 0.2881 wonTackle: 0.8253 -> 0.8354 wasFouled: 0.6954 -> 0.6886
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Tim Iroegbunam, Regan Slater', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 73.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.033` `goal_rate=0.0454` `scorer_weight=0.0378`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0443` `assist_rate=0.051` `provider_weight=0.0496`	-

Sander Berge M 83 0 0.07 0.04 77% 1 0.5 7.0 51.4 31 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0059, quality 0.686)	`market_value_eur=26000000` `quality_factor=0.6858` `boost_factor=1.0059` `reference_quality=0.505` `history_matches=41` `history_shrink=0.1087`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0080)	`blend_weight=0.008` `history_matches=41` `season_overlap_discount=1.0` `per90_to_match_factor=0.922` `prior_reliability=0.1133` `competition=Premier League`	wonTackle: 0.8159 -> 0.8094
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Tim Iroegbunam', 'matchup_zone': 'Midfield lane'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 79.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0648` `goal_rate=0.0869` `scorer_weight=0.0768`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0645` `assist_rate=0.0579` `provider_weight=0.0641`	-

Ryan Sessegnon D 86 0 0.11 0.04 185% 1 0.6 6.7 44.0 30 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0057, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0057` `reference_quality=0.505` `history_matches=31` `history_shrink=0.1389`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0030)	`blend_weight=0.003` `history_matches=31` `season_overlap_discount=1.0` `per90_to_match_factor=0.9606` `prior_reliability=0.0333` `competition=Premier League`	totalCross: 2.5736 -> 2.6063 totalShots: 0.8694 -> 0.8795 onTargetScoringAttempt: 0.3343 -> 0.3396
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Regan Slater', 'matchup_zone': 'Left Mid vs Right Back — Wide Channel'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.30 (squad xG sum 1.3502, xG scale 0.9643, xA scale 1.0725, team xA 0.85 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.302` `squad_xg_sum=1.3502` `squad_xa_sum=0.7891` `team_expected_assists=0.8463` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 14 by projected rating, fifa_overall 77.0)	`match_squad_rank=14` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1034` `goal_rate=0.3333` `scorer_weight=0.1364`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0487` `assist_rate=0.0` `provider_weight=0.0399`	-

Team totals 1 1.30 13 6.9

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Benjamin Lecomte	G	4.4m	10.9%	Market-value substitute prior
Antonee Robinson	D	14.5m	23.6%	Market-value substitute prior
Joachim Andersen	D	14.4m	23.3%	Market-value substitute prior
Hugo Larsson	M	23.1m	54.9%	Market-value substitute prior
Emile Smith Rowe	M	18.4m	70.6%	Market-value substitute prior
Kevin	F	19.0m	73.9%	Market-value substitute prior

Hull — projected line vs season baseline (5-4-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Oli McBurnie F 86 0 0.25 0.10 146% 2 0.8 6.8 33.3 12 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9985, quality 0.494)	`market_value_eur=7600000` `quality_factor=0.4939` `boost_factor=0.9985` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Timothy Castagne, Calvin Bassey', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9151, 'skill_edge': -0.0697, 'height_edge': 0.012, 'attacker_skill': 0.4939, 'defender_skill': 0.6488, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Timothy Castagne', 'market_value': 7600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.524}, {'player': 'Calvin Bassey', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}], 'xg_override': 0.2282}	totalShots: 1.9833 -> 1.8149 shotOffTarget: 1.3452 -> 1.231 onTargetScoringAttempt: 0.6381 -> 0.5839 bigChanceMissed: 0.4282 -> 0.3918 expectedGoals: 0.2494 -> 0.2282 goals: 0.2239 -> 0.2049 expectedGoalsOnTarget: 0.1007 -> 0.0922
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2282 -> 0.2523 expected_assists: 0.0394 -> 0.0505
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 75.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.223` `goal_rate=0.2049` `scorer_weight=0.2312`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0492` `assist_rate=0.1737` `provider_weight=0.0631`	-

Sorba Thomas F 78 0 0.10 0.02 322% 1 0.4 6.8 18.5 17 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0115, quality 0.574)	`market_value_eur=14500000` `quality_factor=0.5737` `boost_factor=1.0115` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Timothy Castagne, Calvin Bassey', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9503, 'skill_edge': -0.0338, 'height_edge': 0.0, 'attacker_skill': 0.5737, 'defender_skill': 0.6488, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Timothy Castagne', 'market_value': 7600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.524}, {'player': 'Calvin Bassey', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}], 'xg_override': 0.0921}	totalShots: 1.2135 -> 1.1532 shotOffTarget: 0.8375 -> 0.7959 onTargetScoringAttempt: 0.376 -> 0.3573 goals: 0.1341 -> 0.1274
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0921 -> 0.1018 expected_assists: 0.0971 -> 0.1245
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 75.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0968` `goal_rate=0.1274` `scorer_weight=0.1271`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1171` `assist_rate=0.3174` `provider_weight=0.1556`	-

Regan Slater M 84 0 0.08 0.02 327% 1 0.3 6.8 34.0 12 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0011, quality 0.513)	`market_value_eur=9200000` `quality_factor=0.513` `boost_factor=1.0011` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0394)	`blend_weight=0.0394` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.9383` `prior_reliability=0.1333` `competition=Premier League`	interceptionWon: 0.9163 -> 0.8802 totalShots: 0.8818 -> 0.8471 fouls: 0.9701 -> 0.9873 goalAssist: 0.081 -> 0.0963 wonTackle: 1.2376 -> 1.2258 onTargetScoringAttempt: 0.2748 -> 0.264
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Ryan Sessegnon', 'matchup_zone': 'Left Mid vs Right Back — Wide Channel', 'duel_win_rate': 0.353, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 0.8433, 'skill_edge': -0.0866, 'height_edge': -0.02, 'attacker_skill': 0.5251, 'defender_skill': 0.7176, 'attacker_role': 'LM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'Ryan Sessegnon', 'market_value': 21000000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.7176}], 'xg_override': 0.0688}	totalShots: 0.871 -> 0.7345 shotOffTarget: 0.607 -> 0.5119 aerialWon: 0.9491 -> 0.9 onTargetScoringAttempt: 0.264 -> 0.2226 expectedGoals: 0.0816 -> 0.0688 goals: 0.0576 -> 0.0486
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0688 -> 0.0761 expected_assists: 0.0746 -> 0.0957
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 72.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0733` `goal_rate=0.0486` `scorer_weight=0.0632`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0913` `assist_rate=0.0963` `provider_weight=0.1124`	-

Tim Iroegbunam M 83 0 0.06 0.02 174% 1 0.4 6.7 41.3 26 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0031, quality 0.585)	`market_value_eur=15500000` `quality_factor=0.5846` `boost_factor=1.0031` `reference_quality=0.505` `history_matches=33` `history_shrink=0.1316`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0006)	`blend_weight=0.0006` `history_matches=33` `season_overlap_discount=1.0` `per90_to_match_factor=0.9207` `prior_reliability=0.0067` `competition=Premier League`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Shea Charles', 'matchup_zone': 'Central Mid vs Defensive Mid — Pressing/Creation Duel', 'duel_win_rate': 0.376, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 0.8919, 'skill_edge': -0.0455, 'height_edge': -0.028, 'attacker_skill': 0.6029, 'defender_skill': 0.704, 'attacker_role': 'CM', 'attacker_mv_price_index': 0.9, 'defender_adjusted_parts': [{'player': 'Shea Charles', 'market_value': 24000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.704}], 'xg_override': 0.0508}	aerialWon: 1.3919 -> 1.1 totalShots: 0.7177 -> 0.6401 shotOffTarget: 0.5205 -> 0.4642 onTargetScoringAttempt: 0.1972 -> 0.1759 goals: 0.0609 -> 0.0543 expectedGoals: 0.057 -> 0.0508
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0508 -> 0.0562 expected_assists: 0.0338 -> 0.0433
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 76.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0546` `goal_rate=0.0543` `scorer_weight=0.0582`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0424` `assist_rate=0.0476` `provider_weight=0.0504`	-

Mohamed Belloumi F 79 1 0.27 0.08 255% 2 1.3 6.9 38.3 21 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0132, quality 0.602)	`market_value_eur=17100000` `quality_factor=0.6016` `boost_factor=1.0132` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Timothy Castagne, Calvin Bassey', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 0.9217, 'skill_edge': -0.0212, 'height_edge': -0.032, 'attacker_skill': 0.6016, 'defender_skill': 0.6488, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Timothy Castagne', 'market_value': 7600000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.524}, {'player': 'Calvin Bassey', 'market_value': 29000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7736}], 'xg_override': 0.2441}	totalShots: 2.3231 -> 2.1412 shotOffTarget: 1.5171 -> 1.3983 onTargetScoringAttempt: 0.806 -> 0.7429 goals: 0.3453 -> 0.3183 expectedGoalsOnTarget: 0.3127 -> 0.2882 bigChanceMissed: 0.2732 -> 0.2518 expectedGoals: 0.2648 -> 0.2441
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2441 -> 0.2697 expected_assists: 0.0784 -> 0.1005
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 73.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.2364` `goal_rate=0.3183` `scorer_weight=0.3371`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0956` `assist_rate=0.4409` `provider_weight=0.1256`	-

Ryan Giles D 85 0 0.00 0.00 37% 0 0.1 6.6 33.7 12 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0094, quality 0.574)	`market_value_eur=14500000` `quality_factor=0.5737` `boost_factor=1.0094` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0374)	`blend_weight=0.0374` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.9433` `prior_reliability=0.1267` `competition=Premier League`	totalCross: 8.4791 -> 8.4033 interceptionWon: 0.747 -> 0.7376 fouls: 0.2066 -> 0.1989 wonTackle: 0.8099 -> 0.8168 wasFouled: 0.1756 -> 0.169 goalAssist: 0.1756 -> 0.169
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Ryan Sessegnon, Bernd Leno', 'matchup_zone': 'LWB vs RW + RB'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0715 -> 0.0917
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 18 by projected rating, fifa_overall 72.0)	`match_squad_rank=18` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0024` `goal_rate=0.0` `scorer_weight=0.0023`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0876` `assist_rate=0.169` `provider_weight=0.1146`	-

Lucas Herrington D 90 0 0.01 0.00 235% 0 0.1 6.6 30.3 11 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0134, quality 0.594)	`market_value_eur=16399999` `quality_factor=0.5942` `boost_factor=1.0134` `reference_quality=0.505` `history_matches=5` `history_shrink=0.5`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0152)	`blend_weight=0.0152` `history_matches=5` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.0467` `competition=Premier League`	wonTackle: 0.8468 -> 0.8339 interceptionWon: 0.7379 -> 0.7267 totalShots: 0.369 -> 0.3634
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Alex Iwobi', 'matchup_zone': 'LW vs CB'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 15 by projected rating, fifa_overall 66.0)	`match_squad_rank=15` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0071` `goal_rate=0.0` `scorer_weight=0.0063`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0135` `assist_rate=0.0` `provider_weight=0.0112`	-

Semi Ajayi D 88 0 0.09 0.05 59% 0 0.2 6.7 33.8 12 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9908, quality 0.437)	`market_value_eur=930000` `quality_factor=0.4374` `boost_factor=0.9908` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0335)	`blend_weight=0.0335` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.9769` `prior_reliability=0.1133` `competition=Premier League`	goals: 0.1059 -> 0.1216 onTargetScoringAttempt: 0.1059 -> 0.1216 wasFouled: 0.1245 -> 0.1396 wonTackle: 0.9872 -> 0.9734 totalCross: 0.7484 -> 0.7426
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Gonzalo García', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0791 -> 0.0875
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 70.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0838` `goal_rate=0.1216` `scorer_weight=0.1094`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0099` `assist_rate=0.0` `provider_weight=0.0086`	-

Lewie Coyle D 80 0 0.03 0.01 129% 0 0.1 6.6 32.1 12 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9908, quality 0.437)	`market_value_eur=2700000` `quality_factor=0.4374` `boost_factor=0.9908` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0256)	`blend_weight=0.0256` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.8914` `prior_reliability=0.0867` `competition=Premier League`	wasFouled: 0.6995 -> 0.7518 fouls: 0.469 -> 0.4921 totalCross: 2.2142 -> 2.1926 totalShots: 0.2667 -> 0.2599
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Alex Iwobi, David Affengruber', 'matchup_zone': 'RWB vs LW + LB'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 17 by projected rating, fifa_overall 70.0)	`match_squad_rank=17` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0254` `goal_rate=0.0` `scorer_weight=0.0206`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0159` `assist_rate=0.0` `provider_weight=0.0128`	-

Konstantinos Tzolakis G 90 0 0.00 - n/a 0.0 6.9 36.3 12 XGBoost G_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0185, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0185` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Gonzalo García, Josh King', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.2748}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	shots_faced: 12.1 -> 12.4 expected_saves: 2.96 -> 3.04 high_pressure_saves: 1.02 -> 1.04
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 79.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.0966`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.7667`	-

John Egan D 88 0 0.02 0.01 170% 1 0.3 6.7 34.2 12 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9908, quality 0.437)	`market_value_eur=1500000` `quality_factor=0.4374` `boost_factor=0.9908` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0394)	`blend_weight=0.0394` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.9796` `prior_reliability=0.1333` `competition=Premier League`	totalShots: 0.5239 -> 0.5033 fouls: 0.4913 -> 0.472 wasFouled: 0.1245 -> 0.1389 wonTackle: 0.843 -> 0.8291 onTargetScoringAttempt: 0.2028 -> 0.1948
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Gonzalo García', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.04 (squad xG sum 0.9407, xG scale 1.1056, xA scale 1.2822, team xA 0.68 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.04` `squad_xg_sum=0.9407` `squad_xa_sum=0.5272` `team_expected_assists=0.676` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0398 -> 0.051
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 13 by projected rating, fifa_overall 72.0)	`match_squad_rank=13` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0195` `goal_rate=0.0` `scorer_weight=0.0158`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0497` `assist_rate=0.0` `provider_weight=0.0408`	-

Team totals 1 1.04 10 6.7

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Jack Butland	G	4.4m	10.9%	Market-value substitute prior
Nobel Mendy	D	14.4m	23.3%	Market-value substitute prior
Charlie Hughes	D	13.9m	31.2%	Market-value substitute prior
Christos Mouzakitis	M	23.0m	54.3%	Market-value substitute prior
Brooke Norton-Cuffy	M	22.6m	52.5%	Market-value substitute prior
Robinio Vaz	F	26.0m	59.1%	Market-value substitute prior

Antagonist / tactical summary

Home midfield dominates with ~52% possession expected (Josh King vs Tim Iroegbunam). Total antagonist xG — Home: 0.81, Away: 0.70. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Midfield — possession 51.7% / 48.3%, dominance ratio 0.58, key duel: Josh King vs Tim Iroegbunam

Alex Iwobi xG 0.09Josh King xG 0.19Gonzalo García xG 0.38Mohamed Belloumi xG 0.24Oli McBurnie xG 0.23Sorba Thomas xG 0.09

Summary

Antagonist matchup analysis for Fulham vs Hull. Most likely scoreline: 1-1. Lambdas derived from 4 home and 7 away individual player duels. Head-to-head: Fulham 6W-1D-3L vs Hull over 10 meetings.

Tactical parity between 4-2-3-1 and 5-4-1. Match outcome driven primarily by individual player quality. Home midfield dominates with ~52% possession expected (Josh King vs Tim Iroegbunam). Total antagonist xG — Home: 0.81, Away: 0.70. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Man City vs Ipswich

Premier League · 20261017 · 4-2-3-1 vs 4-2-3-1

Predicted 3-1Winner: Man CityWinning H 3-1 (8.1%)Winning A 1-2 (3.0%)Confidence 0.79λ home/away 3.08 / 1.18Most-likely P 8.1%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Man City vs Ipswich. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 4-2-3-1 (duel lambdas: Home 2.858, Away 1.365). Final calibrated Poisson lambdas: Home 3.079, Away 1.176 (most likely score 3-1).

Home Starters 11/11 XGBoost (11 FBref) Away Starters 11/11 XGBoost (9 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 2.86 / 1.36 Final λ (H/A) 3.08 / 1.18

FBref Team Priors: Manchester City (Gls/90: 3.0, GA/90: 3.0 → E[λ]: 3.5) vs Ipswich Town (Gls/90: 2.0, GA/90: 2.0 → E[λ]: 3.5) | Blend Weight: 0.192

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Man City	Ipswich
League position	1.00	10.00
Points per game	2.17	1.42
Attack index	1.27	1.01
Defense (concede) index	1.12	0.85
Form multiplier	1.04	0.96
Standing λ multiplier	1.16	0.93

Lambda decomposition (how the goal rates were built)

Stage	Man City λ	Ipswich λ
FBref Squad Prior E[λ] (weight 0.192)	3.500	3.500
Pre-mismatch	2.919	1.336
Pre-calibration (Micro-duels)	2.858	1.365
Final λ	3.079	1.176

Scoreline distribution & markets

Top 5 scorelines	P
3-1	8.1%
2-1	7.9%
3-0	6.9%
2-0	6.7%
4-1	6.3%

Home win	75.2%
Draw	13.6%
Away win	11.2%
Over 2.5	79.7%
BTTS yes	66.0%

Pregame fair line — favorite Man City (75.2%), fair total 4.25, fair spread 1.90

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Man City — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Erling Haaland F 87 2 1.07 0.47 128% 4 2.7 7.0 54.3 30 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0129, quality 0.900)	`market_value_eur=240000000` `quality_factor=0.9` `boost_factor=1.0129` `reference_quality=0.505` `history_matches=41` `history_shrink=0.1087`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0094)	`blend_weight=0.0094` `history_matches=41` `season_overlap_discount=1.0` `per90_to_match_factor=0.9709` `prior_reliability=0.1333` `competition=Premier League`	totalShots: 2.8464 -> 2.8653 onTargetScoringAttempt: 1.326 -> 1.3364 fouls: 1.0655 -> 1.0555
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Leif Davis, Issa Diop', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.337, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.2015, 'skill_edge': 0.1091, 'height_edge': 0.028, 'attacker_skill': 0.9, 'defender_skill': 0.6575, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Leif Davis', 'market_value': 26000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7498}, {'player': 'Issa Diop', 'market_value': 10300000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5652}], 'xg_override': 0.5816}	totalShots: 2.8653 -> 3.4427 shotOffTarget: 1.5204 -> 1.8268 onTargetScoringAttempt: 1.3364 -> 1.6057 bigChanceMissed: 0.9031 -> 1.0851 expectedGoalsOnTarget: 0.7622 -> 0.9158 goals: 0.6503 -> 0.7813 expectedGoals: 0.4841 -> 0.5816
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	goals: 1.0 -> 0
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.5816 -> 1.0677 expected_assists: 0.0592 -> 0.1087
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 91.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=2` `goal_probability=0.6562` `goal_rate=0.7813` `scorer_weight=1.3346`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.103` `assist_rate=0.0931` `provider_weight=0.1359`	-

Antoine Semenyo F 85 0 0.28 0.13 119% 2 1.3 7.1 49.8 42 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0119, quality 0.886)	`market_value_eur=85000000` `quality_factor=0.8859` `boost_factor=1.0119` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0090)	`blend_weight=0.009` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.9434` `prior_reliability=0.1333` `competition=Premier League`	wasFouled: 0.6306 -> 0.6419
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Leif Davis, Issa Diop', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.291, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.1334, 'skill_edge': 0.1028, 'height_edge': -0.012, 'attacker_skill': 0.8859, 'defender_skill': 0.6575, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Leif Davis', 'market_value': 26000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7498}, {'player': 'Issa Diop', 'market_value': 10300000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5652}], 'xg_override': 0.1533}	totalShots: 1.7591 -> 1.9938 shotOffTarget: 1.0208 -> 1.157 onTargetScoringAttempt: 0.7383 -> 0.8368 goals: 0.2334 -> 0.2645 expectedGoalsOnTarget: 0.1755 -> 0.1989 expectedGoals: 0.1353 -> 0.1533
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.1533 -> 0.2814 expected_assists: 0.1746 -> 0.3205
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 85.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2453` `goal_rate=0.2645` `scorer_weight=0.3518`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=1` `assist_probability=0.2742` `assist_rate=0.3123` `provider_weight=0.4006`	-

Rayan Cherki M 78 1 0.47 0.13 256% 3 1.7 7.3 59.2 32 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0134, quality 0.898)	`market_value_eur=120000000` `quality_factor=0.8975` `boost_factor=1.0134` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0059)	`blend_weight=0.0059` `history_matches=39` `season_overlap_discount=1.0` `per90_to_match_factor=0.8623` `prior_reliability=0.08` `competition=Premier League`	fouls: 0.8003 -> 0.8126 totalCross: 2.5118 -> 2.5224 onTargetScoringAttempt: 0.8554 -> 0.863 goalAssist: 0.2098 -> 0.2171 goals: 0.3062 -> 0.3129 wasFouled: 1.0422 -> 1.036
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Christian Walton', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.557, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 1.202, 'skill_edge': 0.1974, 'height_edge': -0.06, 'attacker_skill': 0.8975, 'defender_skill': 0.4588, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Christian Walton', 'market_value': 2600000, 'role': 'GK', 'price_index': 0.55, 'adjusted_quality': 0.4588}], 'xg_override': 0.2537}	aerialWon: 0.0 -> 0.9 totalShots: 2.8943 -> 3.4789 shotOffTarget: 2.0313 -> 2.4416 onTargetScoringAttempt: 0.863 -> 1.0373 expectedGoalsOnTarget: 0.4325 -> 0.5199 goals: 0.3129 -> 0.3761 expectedGoals: 0.2111 -> 0.2537 bigChanceMissed: 0.0503 -> 0.0605
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.2537 -> 0.4657 expected_assists: 0.2149 -> 0.3944
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 86.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.3723` `goal_rate=0.3761` `scorer_weight=0.5821`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=1` `assist_probability=0.3259` `assist_rate=0.2171` `provider_weight=0.3977`	-

Iliman Ndiaye F 82 0 0.21 0.11 83% 2 1.0 6.8 46.8 38 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0123, quality 0.859)	`market_value_eur=63000000` `quality_factor=0.8589` `boost_factor=1.0123` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0101)	`blend_weight=0.0101` `history_matches=38` `season_overlap_discount=1.0` `per90_to_match_factor=0.9149` `prior_reliability=0.1333` `competition=Premier League`	totalCross: 2.075 -> 2.0633 onTargetScoringAttempt: 0.4881 -> 0.497 wasFouled: 1.251 -> 1.243 totalShots: 1.3186 -> 1.3238
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Leif Davis, Issa Diop', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.259, 'pressing_faced': 'LOW', 'numerical_situation': '1v2', 'xg_multiplier': 1.0862, 'skill_edge': 0.0906, 'height_edge': -0.032, 'attacker_skill': 0.8589, 'defender_skill': 0.6575, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Leif Davis', 'market_value': 26000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7498}, {'player': 'Issa Diop', 'market_value': 10300000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.5652}], 'xg_override': 0.1139}	totalShots: 1.3275 -> 1.4419 shotOffTarget: 0.8305 -> 0.9021 onTargetScoringAttempt: 0.497 -> 0.5398 expectedGoalsOnTarget: 0.1616 -> 0.1755 goals: 0.1472 -> 0.1599 bigChanceMissed: 0.1087 -> 0.1181 expectedGoals: 0.1049 -> 0.1139
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.1139 -> 0.2091 expected_assists: 0.1087 -> 0.1995
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 16 by projected rating, fifa_overall 82.0)	`match_squad_rank=16` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1887` `goal_rate=0.1599` `scorer_weight=0.2614`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.1809` `assist_rate=0.1038` `provider_weight=0.1933`	-

Elliot Anderson M 86 0 0.13 0.04 202% 1 1.0 7.4 60.4 31 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0120, quality 0.896)	`market_value_eur=109000000` `quality_factor=0.8957` `boost_factor=1.012` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0071)	`blend_weight=0.0071` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=0.9551` `prior_reliability=0.1067` `competition=Premier League`	interceptionWon: 1.3127 -> 1.3245 wasFouled: 3.1124 -> 3.103 wonTackle: 1.3179 -> 1.3255 fouls: 1.3176 -> 1.3252
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Exequiel Palacios', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.507, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 1.1285, 'skill_edge': 0.0794, 'height_edge': 0.008, 'attacker_skill': 0.8984, 'defender_skill': 0.7219, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Exequiel Palacios', 'market_value': 26000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.7219}], 'xg_override': 0.0697}	aerialWon: 1.2771 -> 1.0 totalShots: 0.9561 -> 1.079 shotOffTarget: 0.7191 -> 0.8115 onTargetScoringAttempt: 0.237 -> 0.2675 goals: 0.0947 -> 0.1069 expectedGoals: 0.0618 -> 0.0697 expectedGoalsOnTarget: 0.0555 -> 0.0626
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.0697 -> 0.1279 expected_assists: 0.1335 -> 0.2451
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 84.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1201` `goal_rate=0.1069` `scorer_weight=0.1599`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.2174` `assist_rate=0.1336` `provider_weight=0.2452`	-

Nico O'Reilly D 82 0 0.37 0.16 132% 1 1.0 6.8 53.3 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0127, quality 0.878)	`market_value_eur=76000000` `quality_factor=0.8781` `boost_factor=1.0127` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0084)	`blend_weight=0.0084` `history_matches=39` `season_overlap_discount=1.0` `per90_to_match_factor=0.9091` `prior_reliability=0.1133` `competition=Premier League`	totalCross: 1.332 -> 1.3388 interceptionWon: 0.7992 -> 0.7925
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Emersonn', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.2031 -> 0.3728 expected_assists: 0.111 -> 0.2038
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 17 by projected rating, fifa_overall 83.0)	`match_squad_rank=17` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.3112` `goal_rate=0.1168` `scorer_weight=0.2982`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.1844` `assist_rate=0.0081` `provider_weight=0.163`	-

Joško Gvardiol D 84 0 0.12 0.08 58% 1 0.6 6.9 45.0 16 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0188, quality 0.869)	`market_value_eur=69000000` `quality_factor=0.8692` `boost_factor=1.0188` `reference_quality=0.505` `history_matches=24` `history_shrink=0.1724`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0134)	`blend_weight=0.0134` `history_matches=24` `season_overlap_discount=1.0` `per90_to_match_factor=0.9321` `prior_reliability=0.12` `competition=Premier League`	wasFouled: 0.8222 -> 0.8111 totalCross: 0.3341 -> 0.3435 wonTackle: 0.9938 -> 0.9874 goals: 0.1361 -> 0.1412
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Jaden Philogene-Bidace', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.065 -> 0.1193 expected_assists: 0.128 -> 0.235
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 85.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1125` `goal_rate=0.1412` `scorer_weight=0.1491`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.2094` `assist_rate=0.3184` `provider_weight=0.2937`	-

Rúben Dias D 88 0 0.08 0.02 314% 1 0.2 7.1 41.6 24 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0136, quality 0.840)	`market_value_eur=55000000` `quality_factor=0.8399` `boost_factor=1.0136` `reference_quality=0.505` `history_matches=32` `history_shrink=0.1351`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0117)	`blend_weight=0.0117` `history_matches=32` `season_overlap_discount=1.0` `per90_to_match_factor=0.9812` `prior_reliability=0.1333` `competition=Premier League`	fouls: 1.296 -> 1.2808
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Emersonn', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.0413 -> 0.0758 expected_assists: 0.0709 -> 0.1301
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 87.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.073` `goal_rate=0.0186` `scorer_weight=0.0606`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.122` `assist_rate=0.0` `provider_weight=0.1041`	-

Matheus Nunes D 87 0 0.01 0.00 91% 0 0.1 6.9 50.0 32 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0101, quality 0.796)	`market_value_eur=43000000` `quality_factor=0.7957` `boost_factor=1.0101` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0020)	`blend_weight=0.002` `history_matches=38` `season_overlap_discount=1.0` `per90_to_match_factor=0.9672` `prior_reliability=0.0267` `competition=Premier League`	totalCross: 1.6087 -> 1.6152
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Julio Enciso', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_assists: 0.0381 -> 0.0699
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 83.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0053` `goal_rate=0.0007` `scorer_weight=0.0051`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.0675` `assist_rate=0.0584` `provider_weight=0.086`	-

Gianluigi Donnarumma G 90 0 0.00 - n/a 0.0 6.9 58.0 31 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0101, quality 0.809)	`market_value_eur=46000000` `quality_factor=0.8089` `boost_factor=1.0101` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0096)	`blend_weight=0.0096` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	saves: 2.9734 -> 2.964
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Emersonn, Julio Enciso', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.3649}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	shots_faced: 13.0 -> 11.2 expected_saves: 3.19 -> 2.74 high_pressure_saves: 1.09 -> 0.94
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 89.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=3.6309`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=2.3181`	-

Ayyoub Bouaddi M 86 0 0.13 0.00 5507% 1 0.1 6.8 20.3 10 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0727, quality 0.893)	`market_value_eur=98000000` `quality_factor=0.8926` `boost_factor=1.0727` `reference_quality=0.505` `history_matches=3` `history_shrink=0.625`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0027)	`blend_weight=0.0027` `history_matches=3` `season_overlap_discount=1.0` `per90_to_match_factor=0.9556` `prior_reliability=0.0067` `competition=Premier League`	interceptionWon: 0.3013 -> 0.3265
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Exequiel Palacios', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.531, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 1.1628, 'skill_edge': 0.0787, 'height_edge': 0.032, 'attacker_skill': 0.8969, 'defender_skill': 0.7219, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Exequiel Palacios', 'market_value': 26000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.7219}], 'xg_override': 0.0715}	aerialWon: 0.0 -> 1.2 totalShots: 0.7328 -> 0.8521 shotOffTarget: 0.5547 -> 0.645 onTargetScoringAttempt: 0.1781 -> 0.2071 goals: 0.1044 -> 0.1214 expectedGoals: 0.0615 -> 0.0715
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 3 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=3`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 3.08 (squad xG sum 1.6774, xG scale 1.8356, xA scale 1.8356, team xA 2.27 = 0.74 x lambda)	`scoreline_reset=true` `team_lambda=3.079` `squad_xg_sum=1.6774` `squad_xa_sum=1.2357` `team_expected_assists=2.2682` `team_xa_to_xg_ratio=0.7367`	expected_goals: 0.0715 -> 0.1312 expected_assists: 0.0603 -> 0.1107
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 13 by projected rating, fifa_overall 80.0)	`match_squad_rank=13` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 3 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=3` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.123` `goal_rate=0.1214` `scorer_weight=0.164`	-
`integer_assist_projection`	Team scoreline allocation: 2 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=2` `assists_allocated=0` `assist_probability=0.1048` `assist_rate=0.0477` `provider_weight=0.0981`	-

Team totals 3 3.08 19 7.0

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Gerónimo Rulli	G	4.5m	11.5%	Market-value substitute prior
Marc Guéhi	D	15.3m	25.2%	Market-value substitute prior
Abdukodir Khusanov	D	15.1m	24.7%	Market-value substitute prior
Enzo Fernández	M	23.8m	58.1%	Market-value substitute prior
Phil Foden	M	23.7m	34.2%	Market-value substitute prior
Jérémy Doku	F	27.0m	63.7%	Market-value substitute prior

Ipswich — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Emersonn F 70 0 0.24 0.17 42% 2 0.9 6.8 34.2 12 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0223, quality 0.669)	`market_value_eur=24000000` `quality_factor=0.6685` `boost_factor=1.0223` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0236)	`blend_weight=0.0236` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.7814` `prior_reliability=0.08` `competition=Premier League`	totalCross: 0.1093 -> 0.1221 fouls: 0.1285 -> 0.1409 wasFouled: 1.7028 -> 1.6933 onTargetScoringAttempt: 0.8947 -> 0.9038 totalOffside: 0.3024 -> 0.2953 goals: 0.3739 -> 0.3805 interceptionWon: 0.2462 -> 0.2404
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': "Nico O'Reilly, Rúben Dias", 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.8307, 'skill_edge': -0.0971, 'height_edge': -0.018, 'attacker_skill': 0.6685, 'defender_skill': 0.8844, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': "Nico O'Reilly", 'market_value': 76000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8937}, {'player': 'Rúben Dias', 'market_value': 55000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8751}], 'xg_override': 0.2626}	totalShots: 2.1045 -> 1.7482 shotOffTarget: 1.2007 -> 0.9974 onTargetScoringAttempt: 0.9038 -> 0.7508 goals: 0.3805 -> 0.3161 expectedGoals: 0.3161 -> 0.2626 bigChanceMissed: 0.2889 -> 0.24 expectedGoalsOnTarget: 0.2701 -> 0.2244
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2626 -> 0.2426
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 74.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2154` `goal_rate=0.3161` `scorer_weight=0.2841`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0443` `assist_rate=0.1737` `provider_weight=0.0566`	-

Jaden Philogene-Bidace F 74 0 0.08 0.01 922% 1 0.3 6.8 17.1 15 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0214, quality 0.619)	`market_value_eur=18800000` `quality_factor=0.6191` `boost_factor=1.0214` `reference_quality=0.505` `history_matches=3` `history_shrink=0.625`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': "Nico O'Reilly, Rúben Dias", 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.7686, 'skill_edge': -0.1194, 'height_edge': -0.038, 'attacker_skill': 0.6191, 'defender_skill': 0.8844, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': "Nico O'Reilly", 'market_value': 76000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8937}, {'player': 'Rúben Dias', 'market_value': 55000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8751}], 'xg_override': 0.0918}	totalShots: 1.5476 -> 1.1895 shotOffTarget: 0.9977 -> 0.7668 onTargetScoringAttempt: 0.5499 -> 0.4227 goals: 0.1716 -> 0.1319 expectedGoals: 0.1194 -> 0.0918
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0918 -> 0.0848
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 85.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0813` `goal_rate=0.1319` `scorer_weight=0.106`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0512` `assist_rate=0.0` `provider_weight=0.0421`	-

Abdul Fatawu Issahaku F 73 0 0.06 0.02 266% 1 0.4 6.7 33.6 20 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0245, quality 0.669)	`market_value_eur=24000000` `quality_factor=0.6685` `boost_factor=1.0245` `reference_quality=0.505` `history_matches=5` `history_shrink=0.5`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0347)	`blend_weight=0.0347` `history_matches=5` `season_overlap_discount=1.0` `per90_to_match_factor=0.8151` `prior_reliability=0.1067` `competition=Premier League`	totalShots: 1.2209 -> 1.1786 fouls: 2.24 -> 2.1977 interceptionWon: 1.131 -> 1.1094 wonTackle: 1.092 -> 1.1071 totalCross: 5.49 -> 5.4763 onTargetScoringAttempt: 0.3899 -> 0.3764 goalAssist: 0.1697 -> 0.1815 totalOffside: 0.1697 -> 0.1815
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': "Nico O'Reilly, Rúben Dias", 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.7955, 'skill_edge': -0.0971, 'height_edge': -0.042, 'attacker_skill': 0.6685, 'defender_skill': 0.8844, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': "Nico O'Reilly", 'market_value': 76000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8937}, {'player': 'Rúben Dias', 'market_value': 55000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8751}], 'xg_override': 0.0682}	totalShots: 1.2074 -> 0.9605 shotOffTarget: 0.831 -> 0.6611 onTargetScoringAttempt: 0.3764 -> 0.2994 goals: 0.0906 -> 0.0721 expectedGoals: 0.0857 -> 0.0682
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0682 -> 0.063
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 16 by projected rating, fifa_overall 82.0)	`match_squad_rank=16` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv_filled` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0611` `goal_rate=0.0721` `scorer_weight=0.0651`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0943` `assist_rate=0.1815` `provider_weight=0.1237`	-

Julio Enciso M 82 0 0.16 0.11 40% 2 1.6 6.9 34.7 12 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0269, quality 0.702)	`market_value_eur=28000000` `quality_factor=0.7022` `boost_factor=1.0269` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0355)	`blend_weight=0.0355` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.9066` `prior_reliability=0.12` `competition=Premier League`	goalAssist: 0.137 -> 0.1679 wonTackle: 0.2389 -> 0.2661 totalShots: 2.5405 -> 2.5591 totalCross: 0.426 -> 0.4109 wasFouled: 1.5963 -> 1.6111 interceptionWon: 0.4135 -> 0.3988 onTargetScoringAttempt: 0.7674 -> 0.7582 goals: 0.2289 -> 0.2208
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Matheus Nunes', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.371, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.8396, 'skill_edge': -0.0691, 'height_edge': -0.04, 'attacker_skill': 0.7022, 'defender_skill': 0.8557, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Matheus Nunes', 'market_value': 43000000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.8557}], 'xg_override': 0.169}	aerialWon: 0.1787 -> 0.9 totalShots: 2.5591 -> 2.1486 shotOffTarget: 1.7731 -> 1.4887 onTargetScoringAttempt: 0.7582 -> 0.6366 expectedGoalsOnTarget: 0.2343 -> 0.1967 goals: 0.2208 -> 0.1854 expectedGoals: 0.2013 -> 0.169 bigChanceMissed: 0.1515 -> 0.1272
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.169 -> 0.1561
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 78.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1445` `goal_rate=0.1854` `scorer_weight=0.1678`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1354` `assist_rate=0.1679` `provider_weight=0.159`	-

Exequiel Palacios M 84 0 0.03 - n/a 1 0.0 7.0 35.0 10 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0271, quality 0.686)	`market_value_eur=26000000` `quality_factor=0.6858` `boost_factor=1.0271` `reference_quality=0.505` `history_matches=5` `history_shrink=0.5`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0065)	`blend_weight=0.0065` `history_matches=5` `season_overlap_discount=1.0` `per90_to_match_factor=0.9364` `prior_reliability=0.02` `competition=Premier League`	interceptionWon: 1.3156 -> 1.3476 wonTackle: 1.3156 -> 1.3273 fouls: 1.3665 -> 1.3779 wasFouled: 1.5531 -> 1.5633
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Elliot Anderson', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.333, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.8715, 'skill_edge': -0.0794, 'height_edge': -0.008, 'attacker_skill': 0.7219, 'defender_skill': 0.8984, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Elliot Anderson', 'market_value': 109000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.8984}], 'xg_override': 0.0346}	aerialWon: 0.0 -> 1.0 totalShots: 0.6761 -> 0.5892 shotOffTarget: 0.5387 -> 0.4695 onTargetScoringAttempt: 0.1374 -> 0.1197 goals: 0.0484 -> 0.0422 expectedGoals: 0.0397 -> 0.0346
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 3 by projected rating, fifa_overall 83.0)	`match_squad_rank=3` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0315` `goal_rate=0.0422` `scorer_weight=0.0349`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.046` `assist_rate=0.0448` `provider_weight=0.0454`	-

Saša Lukić M 83 0 0.03 0.03 ok 0 0.4 7.0 45.4 25 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0004, quality 0.515)	`market_value_eur=9400000` `quality_factor=0.5154` `boost_factor=1.0004` `reference_quality=0.505` `history_matches=32` `history_shrink=0.1351`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0105)	`blend_weight=0.0105` `history_matches=32` `season_overlap_discount=1.0` `per90_to_match_factor=0.9228` `prior_reliability=0.12` `competition=Premier League`	wonTackle: 0.9597 -> 0.9658 wasFouled: 0.4797 -> 0.4746
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Elliot Anderson', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.272, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.783, 'skill_edge': -0.1636, 'height_edge': 0.016, 'attacker_skill': 0.5348, 'defender_skill': 0.8984, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Elliot Anderson', 'market_value': 109000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.8984}], 'xg_override': 0.0309}	aerialWon: 0.4507 -> 1.1 totalShots: 0.5575 -> 0.4365 shotOffTarget: 0.3749 -> 0.2935 onTargetScoringAttempt: 0.181 -> 0.1417 bigChanceMissed: 0.0505 -> 0.0395 goals: 0.0479 -> 0.0375 expectedGoals: 0.0394 -> 0.0309 expectedGoalsOnTarget: 0.0237 -> 0.0186
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 78.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0281` `goal_rate=0.0375` `scorer_weight=0.0308`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0956` `assist_rate=0.1112` `provider_weight=0.1057`	-

Leif Davis D 90 1 0.26 0.14 89% 1 0.6 6.7 33.1 12 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0247, quality 0.686)	`market_value_eur=26000000` `quality_factor=0.6858` `boost_factor=1.0247` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0394)	`blend_weight=0.0394` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	totalCross: 2.8215 -> 2.7695 wonTackle: 0.8195 -> 0.7872 onTargetScoringAttempt: 0.4359 -> 0.4581 fouls: 0.4565 -> 0.4385 totalShots: 0.7332 -> 0.7437 interceptionWon: 0.7897 -> 0.798 goals: 0.3071 -> 0.3147 goalAssist: 0.1783 -> 0.1713
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Erling Haaland', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.2798 -> 0.2585
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 13 by projected rating, fifa_overall 75.0)	`match_squad_rank=13` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.2278` `goal_rate=0.3147` `scorer_weight=0.2858`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0714` `assist_rate=0.1713` `provider_weight=0.0926`	-

Dara O'Shea D 90 0 0.06 0.03 140% 1 0.2 6.9 33.1 12 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0185, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0185` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0394)	`blend_weight=0.0394` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	interceptionWon: 1.2285 -> 1.1998 fouls: 0.8747 -> 0.8993 wasFouled: 0.6135 -> 0.6287 wonTackle: 0.8444 -> 0.8308 goals: 0.2453 -> 0.2356 onTargetScoringAttempt: 0.2453 -> 0.2356
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Erling Haaland', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0665 -> 0.0614
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 6 by projected rating, fifa_overall 75.0)	`match_squad_rank=6` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0596` `goal_rate=0.2356` `scorer_weight=0.0767`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0375` `assist_rate=0.0` `provider_weight=0.0306`	-

Issa Diop D 90 0 0.06 0.03 94% 0 0.1 6.9 32.8 29 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0013, quality 0.526)	`market_value_eur=10300000` `quality_factor=0.526` `boost_factor=1.0013` `reference_quality=0.505` `history_matches=19` `history_shrink=0.2083`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0181)	`blend_weight=0.0181` `history_matches=19` `season_overlap_discount=1.0` `per90_to_match_factor=0.9968` `prior_reliability=0.1333` `competition=Premier League`	fouls: 0.7295 -> 0.7433 totalShots: 0.4854 -> 0.4766 interceptionWon: 1.0446 -> 1.0527
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Erling Haaland', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 75.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0584` `goal_rate=0.0621` `scorer_weight=0.0586`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.022` `assist_rate=0.0` `provider_weight=0.0178`	-

Abdoul Ouattara M 88 0 0.03 0.00 574% 0 0.2 6.9 45.7 18 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0123, quality 0.595)	`market_value_eur=16500000` `quality_factor=0.5953` `boost_factor=1.0123` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0059)	`blend_weight=0.0059` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.9778` `prior_reliability=0.02` `competition=Premier League`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Elliot Anderson, Ayyoub Bouaddi', 'matchup_zone': 'Attacking Mid vs Double Pivot — Tight Space Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'ULTRA_HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.7943, 'skill_edge': -0.1319, 'height_edge': -0.008, 'attacker_skill': 0.6044, 'defender_skill': 0.8976, 'attacker_role': 'CAM', 'attacker_mv_price_index': 0.95, 'defender_adjusted_parts': [{'player': 'Elliot Anderson', 'market_value': 109000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.8984}, {'player': 'Ayyoub Bouaddi', 'market_value': 98000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.8969}], 'xg_override': 0.0319}	aerialWon: 0.2438 -> 1.1 totalShots: 0.6103 -> 0.4848 shotOffTarget: 0.4763 -> 0.3783 onTargetScoringAttempt: 0.134 -> 0.1064 expectedGoalsOnTarget: 0.1051 -> 0.0835 goals: 0.0553 -> 0.0439 expectedGoals: 0.0401 -> 0.0319
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	-
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 73.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0291` `goal_rate=0.0439` `scorer_weight=0.0338`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0488` `assist_rate=0.0559` `provider_weight=0.0523`	-

Christian Walton G 90 0 0.00 - n/a 0.0 6.9 11.4 12 FBref Prior G_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9855, quality 0.437)	`market_value_eur=2600000` `quality_factor=0.4374` `boost_factor=0.9855` `reference_quality=0.505` `history_matches=2` `history_shrink=0.7143`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Erling Haaland, Rayan Cherki', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 2.8576}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.18 (squad xG sum 1.2730, xG scale 0.9238, xA scale 0.9738, team xA 0.76 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.176` `squad_xg_sum=1.273` `squad_xa_sum=0.785` `team_expected_assists=0.7644` `team_xa_to_xg_ratio=0.65`	shots_faced: 27.1 -> 29.3 expected_saves: 6.63 -> 7.18 high_pressure_saves: 2.29 -> 2.46
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 73.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.3239`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.8324`	-

Team totals 1 1.18 12 6.8

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Kjell Scherpen	G	4.8m	14.0%	Market-value substitute prior
Jacob Greaves	D	14.2m	23.0%	Market-value substitute prior
Cédric Kipré	D	13.6m	29.9%	Market-value substitute prior
Florentino Luís	M	22.6m	52.5%	Market-value substitute prior
Marcelino Núñez	M	22.5m	31.8%	Market-value substitute prior
Jack Clarke	F	22.7m	73.5%	Market-value substitute prior

Antagonist / tactical summary

Home midfield dominates with ~52% possession expected (Rayan Cherki vs Julio Enciso). Total antagonist xG — Home: 1.24, Away: 0.69. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Midfield — possession 52.2% / 47.8%, dominance ratio 0.58, key duel: Rayan Cherki vs Julio Enciso

Rayan Cherki xG 0.25Erling Haaland xG 0.58Antoine Semenyo xG 0.15Julio Enciso xG 0.17Abdoul Ouattara xG 0.03Emersonn xG 0.26

Summary

Antagonist matchup analysis for Man City vs Ipswich. Most likely scoreline: 3-1. Lambdas derived from 6 home and 7 away individual player duels. Head-to-head: Man City 6W-1D-3L vs Ipswich over 10 meetings.

Tactical parity between 4-2-3-1 and 4-2-3-1. Match outcome driven primarily by individual player quality. Home midfield dominates with ~52% possession expected (Rayan Cherki vs Julio Enciso). Total antagonist xG — Home: 1.24, Away: 0.69. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.

Brentford vs Liverpool

Premier League · 20261017 · 4-2-3-1 vs 4-2-3-1

Predicted 1-1Winner: DrawWinning H 2-1 (8.1%)Winning A 1-2 (8.7%)Confidence 0.7λ home/away 1.43 / 1.55Most-likely P 11.3%

Actual vs predicted: result not available yet (upcoming fixture).

XGBoost Reasoning & FBref Dataset Tracing

Antagonist + XGBoost match analysis for Brentford vs Liverpool. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 4-2-3-1 (duel lambdas: Home 1.475, Away 1.504). Final calibrated Poisson lambdas: Home 1.433, Away 1.547 (most likely score 1-1).

Home Starters 11/11 XGBoost (9 FBref) Away Starters 11/11 XGBoost (10 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.48 / 1.50 Final λ (H/A) 1.43 / 1.55

FBref Team Priors: Brentford (Gls/90: 2.0, GA/90: 2.0 → E[λ]: 2.759) vs Liverpool FC (Gls/90: 2.0, GA/90: 2.0 → E[λ]: 2.759) | Blend Weight: 0.192

Modeling Pipeline Stages (5 layers)

1. XGBoost player regressions for volume & event rates with market-value quality scaling
2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
5. Bivariate Poisson matrix with Dixon-Coles low-score correction

Standing & form context

Metric	Brentford	Liverpool
League position	4.00	9.00
Points per game	1.50	1.42
Attack index	1.12	0.93
Defense (concede) index	0.93	0.89
Form multiplier	1.00	0.99
Standing λ multiplier	1.02	0.93

Lambda decomposition (how the goal rates were built)

Stage	Brentford λ	Liverpool λ
FBref Squad Prior E[λ] (weight 0.192)	2.759	2.759
Pre-competitive-floor	1.433	1.547
Pre-calibration (Micro-duels)	1.475	1.504
Final λ	1.433	1.547

Scoreline distribution & markets

Top 5 scorelines	P
1-1	11.3%
1-2	8.7%
2-1	8.1%
0-1	7.8%
1-0	7.2%

Home win	35.2%
Draw	24.5%
Away win	40.2%
Over 2.5	57.2%
BTTS yes	60.0%

Pregame fair line — favorite Liverpool (40.2%), fair total 2.98, fair spread -0.11

Projected players vs season baseline

Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.

Brentford — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Keane Lewis-Potter D 81 0 0.20 0.10 93% 1 1.0 6.7 54.5 33 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0065, quality 0.710)	`market_value_eur=29000000` `quality_factor=0.71` `boost_factor=1.0065` `reference_quality=0.505` `history_matches=42` `history_shrink=0.1064`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0092)	`blend_weight=0.0092` `history_matches=42` `season_overlap_discount=1.0` `per90_to_match_factor=0.9003` `prior_reliability=0.1333` `competition=Premier League`	totalCross: 3.5196 -> 3.5287 wasFouled: 0.97 -> 0.9611
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Lewis Koumas', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.1469 -> 0.1976 expected_assists: 0.1228 -> 0.1397
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 15 by projected rating, fifa_overall 77.0)	`match_squad_rank=15` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1793` `goal_rate=0.0971` `scorer_weight=0.1581`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1304` `assist_rate=0.1335` `provider_weight=0.1483`	-

Mathias Jensen M 82 0 0.10 0.06 51% 1 0.9 7.1 57.8 27 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9999, quality 0.502)	`market_value_eur=8300000` `quality_factor=0.5023` `boost_factor=0.9999` `reference_quality=0.505` `history_matches=38` `history_shrink=0.1163`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0035)	`blend_weight=0.0035` `history_matches=38` `season_overlap_discount=1.0` `per90_to_match_factor=0.9107` `prior_reliability=0.0467` `competition=Premier League`	fouls: 1.095 -> 1.1049 totalCross: 4.4209 -> 4.4283
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Virgil van Dijk', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.353, 'pressing_faced': 'LOW', 'numerical_situation': '1v1', 'xg_multiplier': 0.813, 'skill_edge': -0.0672, 'height_edge': -0.06, 'attacker_skill': 0.5023, 'defender_skill': 0.6516, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Virgil van Dijk', 'market_value': 15500000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.6516}], 'xg_override': 0.0716}	aerialWon: 0.2414 -> 1.1 totalShots: 1.1385 -> 0.9256 shotOffTarget: 0.745 -> 0.6057 onTargetScoringAttempt: 0.3917 -> 0.3185 bigChanceMissed: 0.1847 -> 0.1502 goals: 0.0895 -> 0.0728 expectedGoals: 0.0881 -> 0.0716 expectedGoalsOnTarget: 0.0572 -> 0.0465
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0716 -> 0.0963 expected_assists: 0.1588 -> 0.1807
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 80.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0918` `goal_rate=0.0728` `scorer_weight=0.0973`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1653` `assist_rate=0.127` `provider_weight=0.1532`	-

Jaidon Anthony F 80 0 0.25 0.13 94% 1 1.2 6.8 49.3 40 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0038, quality 0.625)	`market_value_eur=19400000` `quality_factor=0.6251` `boost_factor=1.0038` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0041)	`blend_weight=0.0041` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.8931` `prior_reliability=0.06` `competition=Premier League`	fouls: 1.2353 -> 1.2424
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Jérémy Jacquet, Ronald Araújo', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 0.7865, 'skill_edge': -0.0892, 'height_edge': -0.056, 'attacker_skill': 0.6251, 'defender_skill': 0.8234, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Jérémy Jacquet', 'market_value': 67000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8886}, {'player': 'Ronald Araújo', 'market_value': 27000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7581}], 'xg_override': 0.1822}	totalShots: 1.6664 -> 1.3106 shotOffTarget: 1.0251 -> 0.8062 onTargetScoringAttempt: 0.6338 -> 0.4985 expectedGoalsOnTarget: 0.2518 -> 0.198 goals: 0.2414 -> 0.1899 bigChanceMissed: 0.2386 -> 0.1877 expectedGoals: 0.2316 -> 0.1822
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.1822 -> 0.2451 expected_assists: 0.0924 -> 0.1051
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 8 by projected rating, fifa_overall 76.0)	`match_squad_rank=8` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.2174` `goal_rate=0.1899` `scorer_weight=0.2532`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0998` `assist_rate=0.1993` `provider_weight=0.1314`	-

Vitaly Janelt M 86 0 0.06 0.05 35% 1 0.4 7.0 51.8 25 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0057, quality 0.641)	`market_value_eur=21000000` `quality_factor=0.6408` `boost_factor=1.0057` `reference_quality=0.505` `history_matches=31` `history_shrink=0.1389`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0120)	`blend_weight=0.012` `history_matches=31` `season_overlap_discount=1.0` `per90_to_match_factor=0.9608` `prior_reliability=0.1333` `competition=Premier League`	totalCross: 2.4592 -> 2.4816 interceptionWon: 2.5939 -> 2.6089 fouls: 1.4214 -> 1.4332 wonTackle: 1.1342 -> 1.1437
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Trey Nyoni', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.445, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 1.0368, 'skill_edge': 0.009, 'height_edge': 0.016, 'attacker_skill': 0.6747, 'defender_skill': 0.6546, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Trey Nyoni', 'market_value': 19100000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.6546}], 'xg_override': 0.0459}	aerialWon: 0.6594 -> 1.1 totalShots: 0.668 -> 0.6926 shotOffTarget: 0.494 -> 0.5122 onTargetScoringAttempt: 0.174 -> 0.1804
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0459 -> 0.0617 expected_assists: 0.0828 -> 0.0942
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 78.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0598` `goal_rate=0.0987` `scorer_weight=0.0771`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0899` `assist_rate=0.0751` `provider_weight=0.0887`	-

Yehor Yarmoliuk M 78 0 0.07 0.03 133% 1 0.3 6.9 54.7 31 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0066, quality 0.718)	`market_value_eur=30000000` `quality_factor=0.7176` `boost_factor=1.0066` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0023)	`blend_weight=0.0023` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.8701` `prior_reliability=0.0333` `competition=Premier League`	totalCross: 1.2485 -> 1.2535
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Trey Nyoni', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.465, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 1.0656, 'skill_edge': 0.0446, 'height_edge': 0.0, 'attacker_skill': 0.7538, 'defender_skill': 0.6546, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Trey Nyoni', 'market_value': 19100000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.6546}], 'xg_override': 0.0513}	aerialWon: 1.1571 -> 1.1 totalShots: 0.7452 -> 0.7941 shotOffTarget: 0.5072 -> 0.5405 onTargetScoringAttempt: 0.2352 -> 0.2506
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0513 -> 0.069 expected_assists: 0.0866 -> 0.0985
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 78.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0667` `goal_rate=0.0801` `scorer_weight=0.0863`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0938` `assist_rate=0.0896` `provider_weight=0.1007`	-

El Hadji Malick Diouf D 88 0 0.04 0.01 276% 1 0.3 6.7 48.5 27 FBref Prior D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0095, quality 0.758)	`market_value_eur=36000000` `quality_factor=0.7581` `boost_factor=1.0095` `reference_quality=0.505` `history_matches=35` `history_shrink=0.125`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Florian Wirtz', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0299 -> 0.0402 expected_assists: 0.0819 -> 0.0932
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 17 by projected rating, fifa_overall 78.0)	`match_squad_rank=17` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0394` `goal_rate=0.0` `scorer_weight=0.0322`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.089` `assist_rate=0.0463` `provider_weight=0.0746`	-

Kristoffer Ajer D 88 0 0.06 0.01 391% 1 0.2 6.8 45.7 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0075, quality 0.694)	`market_value_eur=27000000` `quality_factor=0.6941` `boost_factor=1.0075` `reference_quality=0.505` `history_matches=33` `history_shrink=0.1316`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0114)	`blend_weight=0.0114` `history_matches=33` `season_overlap_discount=1.0` `per90_to_match_factor=0.9781` `prior_reliability=0.1333` `competition=Premier League`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Bradley Barcola', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0435 -> 0.0585
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 77.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0568` `goal_rate=0.1531` `scorer_weight=0.0731`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0183` `assist_rate=0.0` `provider_weight=0.0148`	-

Jannik Schuster D 80 0 0.00 0.01 100% 1 0.1 6.8 24.3 12 FBref Prior D_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0210, quality 0.631)	`market_value_eur=20000000` `quality_factor=0.6311` `boost_factor=1.021` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Bradley Barcola', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_assists: 0.1785 -> 0.2032
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 67.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `goal_rate=0.0` `scorer_weight=0.001`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1839` `assist_rate=0.3373` `provider_weight=0.254`	-

Michael Kayode D 81 0 0.10 0.05 93% 1 0.3 6.7 55.3 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0085, quality 0.775)	`market_value_eur=39000000` `quality_factor=0.7754` `boost_factor=1.0085` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0081)	`blend_weight=0.0081` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.9046` `prior_reliability=0.12` `competition=Premier League`	totalShots: 0.5344 -> 0.5505 wasFouled: 0.9364 -> 0.9451
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Bradley Barcola', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.0716 -> 0.0963 expected_assists: 0.0596 -> 0.0678
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 18 by projected rating, fifa_overall 81.0)	`match_squad_rank=18` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0918` `goal_rate=0.1079` `scorer_weight=0.1204`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0656` `assist_rate=0.0251` `provider_weight=0.0542`	-

Caoimhin Kelleher G 90 0 0.00 - n/a 0.0 7.0 56.9 31 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0064, quality 0.710)	`market_value_eur=29000000` `quality_factor=0.71` `boost_factor=1.0064` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0090)	`blend_weight=0.009` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Florian Wirtz, Víctor Muñoz', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.5041}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	shots_faced: 14.3 -> 14.7 expected_saves: 3.5 -> 3.6 high_pressure_saves: 1.2 -> 1.24
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 80.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.4802`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=1.1464`	-

Igor Thiago F 89 1 0.46 0.46 ok 2 1.9 6.8 55.1 32 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0111, quality 0.868)	`market_value_eur=68000000` `quality_factor=0.8677` `boost_factor=1.0111` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0084)	`blend_weight=0.0084` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=0.9896` `prior_reliability=0.1267` `competition=Premier League`	totalShots: 2.1242 -> 2.1368 fouls: 1.949 -> 1.937 wasFouled: 0.6428 -> 0.6374
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Jérémy Jacquet, Ronald Araújo', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.22, 'pressing_faced': 'HIGH', 'numerical_situation': '1v2', 'xg_multiplier': 1.0293, 'skill_edge': 0.0199, 'height_edge': 0.0, 'attacker_skill': 0.8677, 'defender_skill': 0.8234, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Jérémy Jacquet', 'market_value': 67000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.8886}, {'player': 'Ronald Araújo', 'market_value': 27000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7581}], 'xg_override': 0.3401}	totalShots: 2.1368 -> 2.1994 shotOffTarget: 1.3504 -> 1.39 onTargetScoringAttempt: 0.776 -> 0.7987 bigChanceMissed: 0.6625 -> 0.6819 expectedGoals: 0.3304 -> 0.3401 goals: 0.3325 -> 0.3422 expectedGoalsOnTarget: 0.31 -> 0.3191
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.43 (squad xG sum 1.0653, xG scale 1.3451, xA scale 1.1378, team xA 1.15 = 0.80 x lambda)	`scoreline_reset=true` `team_lambda=1.433` `squad_xg_sum=1.0653` `squad_xa_sum=1.0076` `team_expected_assists=1.1464` `team_xa_to_xg_ratio=0.8`	expected_goals: 0.3401 -> 0.4575 expected_assists: 0.0593 -> 0.0675
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 12 by projected rating, fifa_overall 82.0)	`match_squad_rank=12` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.3671` `goal_rate=0.3422` `scorer_weight=0.46`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0653` `assist_rate=0.0022` `provider_weight=0.054`	-

Team totals 1 1.43 13 6.8

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Hákon Rafn Valdimarsson	G	4.4m	10.9%	Market-value substitute prior
Sepp van den Berg	D	14.8m	35.3%	Market-value substitute prior
Nathan Collins	D	14.7m	23.9%	Market-value substitute prior
Mamadou Sangare	M	23.5m	33.8%	Market-value substitute prior
Mikkel Damsgaard	M	18.6m	52.2%	Market-value substitute prior
Kevin Schade	F	26.9m	36.6%	Market-value substitute prior

Liverpool — projected line vs season baseline (4-2-3-1)

Player Pos Min Proj G Proj xG Season xG/g Proj Sh Season Sh/g Proj Rtg Season Rtg Rec Model / FBref Prior

Lewis Koumas F 78 0 0.16 0.03 424% 2 0.2 6.7 16.1 12 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 0.9979, quality 0.493)	`market_value_eur=7500000` `quality_factor=0.4927` `boost_factor=0.9979` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0024)	`blend_weight=0.0024` `history_matches=4` `season_overlap_discount=1.0` `per90_to_match_factor=0.8703` `prior_reliability=0.0067` `competition=Premier League`	totalShots: 1.5062 -> 1.5171
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Kristoffer Ajer, Jannik Schuster', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 0.7759, 'skill_edge': -0.1044, 'height_edge': -0.048, 'attacker_skill': 0.4927, 'defender_skill': 0.7248, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Kristoffer Ajer', 'market_value': 27000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7581}, {'player': 'Jannik Schuster', 'market_value': 20000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.6914}], 'xg_override': 0.1352}	totalShots: 1.5171 -> 1.1771 shotOffTarget: 0.8559 -> 0.6641 onTargetScoringAttempt: 0.6487 -> 0.5033 goals: 0.1918 -> 0.1488 expectedGoals: 0.1742 -> 0.1352 expectedGoalsOnTarget: 0.0576 -> 0.0447
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1352 -> 0.1603 expected_assists: 0.031 -> 0.0442
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 17 by projected rating, fifa_overall 69.0)	`match_squad_rank=17` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1481` `goal_rate=0.1488` `scorer_weight=0.1721`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0432` `assist_rate=0.0` `provider_weight=0.0354`	-

Bradley Barcola F 79 0 0.16 0.04 264% 1 0.4 6.7 30.2 13 XGBoost F_service

Step-by-step (9 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0645, quality 0.892)	`market_value_eur=97000000` `quality_factor=0.8922` `boost_factor=1.0645` `reference_quality=0.505` `history_matches=4` `history_shrink=0.5556`	-
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Kristoffer Ajer, Jannik Schuster', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.231, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 1.0461, 'skill_edge': 0.0754, 'height_edge': -0.044, 'attacker_skill': 0.8922, 'defender_skill': 0.7248, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Kristoffer Ajer', 'market_value': 27000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7581}, {'player': 'Jannik Schuster', 'market_value': 20000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.6914}], 'xg_override': 0.1323}	totalShots: 1.2592 -> 1.3173 shotOffTarget: 0.789 -> 0.8254 onTargetScoringAttempt: 0.4702 -> 0.4919 bigChanceMissed: 0.299 -> 0.3128 goals: 0.1737 -> 0.1817 expectedGoals: 0.1265 -> 0.1323
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1323 -> 0.1568 expected_assists: 0.0791 -> 0.1128
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 14 by projected rating, fifa_overall 85.0)	`match_squad_rank=14` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1451` `goal_rate=0.1817` `scorer_weight=0.196`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1067` `assist_rate=0.0` `provider_weight=0.0902`	-

Florian Wirtz M 80 1 0.23 0.18 ok 2 1.5 6.9 53.7 28 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0129, quality 0.883)	`market_value_eur=81000000` `quality_factor=0.8829` `boost_factor=1.0129` `reference_quality=0.505` `history_matches=39` `history_shrink=0.1136`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0084)	`blend_weight=0.0084` `history_matches=39` `season_overlap_discount=1.0` `per90_to_match_factor=0.8907` `prior_reliability=0.1133` `competition=Premier League`	fouls: 1.5562 -> 1.5476 totalCross: 1.8799 -> 1.8729 totalShots: 1.8723 -> 1.8787
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'El Hadji Malick Diouf', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle', 'duel_win_rate': 0.472, 'pressing_faced': 'HIGH', 'numerical_situation': '1v1', 'xg_multiplier': 0.9886, 'skill_edge': 0.0242, 'height_edge': -0.032, 'attacker_skill': 0.8829, 'defender_skill': 0.829, 'attacker_role': 'LW', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'El Hadji Malick Diouf', 'market_value': 36000000, 'role': 'RB', 'price_index': 0.7, 'adjusted_quality': 0.829}], 'xg_override': 0.1937}	aerialWon: 0.4342 -> 0.9 totalShots: 1.8787 -> 1.8573 shotOffTarget: 1.2383 -> 1.2242 onTargetScoringAttempt: 0.6375 -> 0.6302
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1937 -> 0.2296 expected_assists: 0.1438 -> 0.2051
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 5 by projected rating, fifa_overall 86.0)	`match_squad_rank=5` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=1` `goal_probability=0.2051` `goal_rate=0.1791` `scorer_weight=0.2158`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.1854` `assist_rate=0.1499` `provider_weight=0.2116`	-

Víctor Muñoz F 82 0 0.19 0.15 ok 2 1.9 6.9 55.0 33 FBref Prior F_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost F_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0090, quality 0.775)	`market_value_eur=39000000` `quality_factor=0.7754` `boost_factor=1.009` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0063)	`blend_weight=0.0063` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.9141` `prior_reliability=0.0867` `competition=Premier League`	totalCross: 2.539 -> 2.5583 totalShots: 2.0129 -> 2.0265 fouls: 1.2559 -> 1.2656 interceptionWon: 0.7519 -> 0.7604 wasFouled: 1.939 -> 1.9445
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Kristoffer Ajer, Jannik Schuster', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle', 'duel_win_rate': 0.18, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v2', 'xg_multiplier': 0.9453, 'skill_edge': 0.0228, 'height_edge': -0.06, 'attacker_skill': 0.7754, 'defender_skill': 0.7248, 'attacker_role': 'ST', 'attacker_mv_price_index': 1.0, 'defender_adjusted_parts': [{'player': 'Kristoffer Ajer', 'market_value': 27000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.7581}, {'player': 'Jannik Schuster', 'market_value': 20000000, 'role': 'CB', 'price_index': 0.75, 'adjusted_quality': 0.6914}], 'xg_override': 0.1617}	totalShots: 2.0265 -> 1.9157 shotOffTarget: 1.4025 -> 1.3258 onTargetScoringAttempt: 0.6109 -> 0.5775 bigChanceMissed: 0.4785 -> 0.4523 expectedGoalsOnTarget: 0.2667 -> 0.2521 goals: 0.1923 -> 0.1818 expectedGoals: 0.1711 -> 0.1617
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1617 -> 0.1917 expected_assists: 0.1067 -> 0.1522
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 9 by projected rating, fifa_overall 79.0)	`match_squad_rank=9` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1744` `goal_rate=0.1818` `scorer_weight=0.2099`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=1` `assist_probability=0.1412` `assist_rate=0.0311` `provider_weight=0.1218`	-

Trey Nyoni M 86 0 0.05 0.01 335% 1 0.0 6.7 14.7 31 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0117, quality 0.622)	`market_value_eur=19100000` `quality_factor=0.6221` `boost_factor=1.0117` `reference_quality=0.505` `history_matches=10` `history_shrink=0.3333`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0029)	`blend_weight=0.0029` `history_matches=10` `season_overlap_discount=1.0` `per90_to_match_factor=0.9556` `prior_reliability=0.0133` `competition=Premier League`	interceptionWon: 0.4656 -> 0.478
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Vitaly Janelt', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.395, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 0.9632, 'skill_edge': -0.009, 'height_edge': -0.016, 'attacker_skill': 0.6546, 'defender_skill': 0.6747, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Vitaly Janelt', 'market_value': 21000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.6747}], 'xg_override': 0.0438}	aerialWon: 0.0 -> 1.1 totalShots: 0.6765 -> 0.6516 shotOffTarget: 0.4989 -> 0.4805 onTargetScoringAttempt: 0.1776 -> 0.1711
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0438 -> 0.0519 expected_assists: 0.0405 -> 0.0578
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 11 by projected rating, fifa_overall 67.0)	`match_squad_rank=11` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0506` `goal_rate=0.0672` `scorer_weight=0.0648`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0562` `assist_rate=0.0423` `provider_weight=0.059`	-

Alexis Mac Allister M 87 0 0.14 0.08 74% 2 1.1 6.9 55.2 32 FBref Prior M_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost M_service raw output over 8 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0117, quality 0.879)	`market_value_eur=77000000` `quality_factor=0.8792` `boost_factor=1.0117` `reference_quality=0.505` `history_matches=43` `history_shrink=0.1042`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0054)	`blend_weight=0.0054` `history_matches=43` `season_overlap_discount=1.0` `per90_to_match_factor=0.9719` `prior_reliability=0.08` `competition=Premier League`	wasFouled: 0.8219 -> 0.8306 totalShots: 1.3878 -> 1.3803
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` duel_context={'opponent_faced': 'Vitaly Janelt', 'matchup_zone': 'CDM vs CDM', 'duel_win_rate': 0.485, 'pressing_faced': 'MEDIUM', 'numerical_situation': '1v1', 'xg_multiplier': 1.0949, 'skill_edge': 0.0966, 'height_edge': -0.032, 'attacker_skill': 0.8893, 'defender_skill': 0.6747, 'attacker_role': 'CDM', 'attacker_mv_price_index': 0.85, 'defender_adjusted_parts': [{'player': 'Vitaly Janelt', 'market_value': 21000000, 'role': 'CDM', 'price_index': 0.85, 'adjusted_quality': 0.6747}], 'xg_override': 0.1182}	totalShots: 1.3853 -> 1.5168 shotOffTarget: 0.9459 -> 1.0357 onTargetScoringAttempt: 0.4394 -> 0.4811 aerialWon: 0.8584 -> 0.9 goals: 0.1183 -> 0.1295 expectedGoals: 0.108 -> 0.1182 expectedGoalsOnTarget: 0.0897 -> 0.0982
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.1182 -> 0.1401 expected_assists: 0.0733 -> 0.1046
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 7 by projected rating, fifa_overall 84.0)	`match_squad_rank=7` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.1307` `goal_rate=0.1295` `scorer_weight=0.1495`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0993` `assist_rate=0.0691` `provider_weight=0.101`	-

Miloš Kerkez D 85 0 0.04 0.02 123% 0 0.2 6.7 47.3 32 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0084, quality 0.758)	`market_value_eur=36000000` `quality_factor=0.7581` `boost_factor=1.0084` `reference_quality=0.505` `history_matches=40` `history_shrink=0.1111`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0087)	`blend_weight=0.0087` `history_matches=40` `season_overlap_discount=1.0` `per90_to_match_factor=0.9463` `prior_reliability=0.12` `competition=Premier League`	totalShots: 0.4752 -> 0.4803
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Igor Thiago', 'matchup_zone': 'Shape fallback'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0371 -> 0.044 expected_assists: 0.0175 -> 0.025
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 16 by projected rating, fifa_overall 81.0)	`match_squad_rank=16` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.043` `goal_rate=0.0525` `scorer_weight=0.0518`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0247` `assist_rate=0.0215` `provider_weight=0.0265`	-

Virgil van Dijk D 90 0 0.09 0.10 ok 1 0.7 7.1 58.5 32 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0024, quality 0.585)	`market_value_eur=15500000` `quality_factor=0.5846` `boost_factor=1.0024` `reference_quality=0.505` `history_matches=44` `history_shrink=0.102`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0088)	`blend_weight=0.0088` `history_matches=44` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	fouls: 0.6505 -> 0.6447 interceptionWon: 1.1418 -> 1.1361
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Mathias Jensen', 'matchup_zone': 'Left Wing vs Right Back — Flank Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0776 -> 0.092 expected_assists: 0.0281 -> 0.0401
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 1 by projected rating, fifa_overall 88.0)	`match_squad_rank=1` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0879` `goal_rate=0.1633` `scorer_weight=0.115`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0393` `assist_rate=0.0` `provider_weight=0.0321`	-

Jérémy Jacquet D 87 0 0.08 0.02 259% 0 0.1 7.0 37.5 12 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0492, quality 0.866)	`market_value_eur=67000000` `quality_factor=0.8661` `boost_factor=1.0492` `reference_quality=0.505` `history_matches=6` `history_shrink=0.4545`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0315)	`blend_weight=0.0315` `history_matches=6` `season_overlap_discount=1.0` `per90_to_match_factor=0.9646` `prior_reliability=0.1067` `competition=Premier League`	interceptionWon: 1.1682 -> 1.2074 wasFouled: 0.7137 -> 0.7482 totalShots: 0.5057 -> 0.4898 fouls: 1.0725 -> 1.0577 onTargetScoringAttempt: 0.2526 -> 0.2446
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Igor Thiago', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_goals: 0.0685 -> 0.0812 expected_assists: 0.021 -> 0.03
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 4 by projected rating, fifa_overall 77.0)	`match_squad_rank=4` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.078` `goal_rate=0.0` `scorer_weight=0.065`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0296` `assist_rate=0.0` `provider_weight=0.024`	-

Ronald Araújo D 78 0 0.02 0.03 ok 0 0.2 6.8 40.4 31 FBref Prior D_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost D_service raw output over 7 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0081, quality 0.694)	`market_value_eur=27000000` `quality_factor=0.6941` `boost_factor=1.0081` `reference_quality=0.505` `history_matches=30` `history_shrink=0.1429`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0025)	`blend_weight=0.0025` `history_matches=30` `season_overlap_discount=1.0` `per90_to_match_factor=0.8697` `prior_reliability=0.0267` `competition=Premier League`	fouls: 0.4416 -> 0.4513
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=true`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Igor Thiago', 'matchup_zone': 'Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle'}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	expected_assists: 0.0277 -> 0.0395
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 10 by projected rating, fifa_overall 80.0)	`match_squad_rank=10` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0246` `goal_rate=0.0868` `scorer_weight=0.0311`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0387` `assist_rate=0.0` `provider_weight=0.0316`	-

Alisson G 90 0 0.00 - n/a 0.0 7.0 51.8 25 FBref Prior G_service

Step-by-step (10 steps)

Step	Description	Key params	Changes
`xgboost_raw`	XGBoost G_service raw output over 3 target models	-	-
`market_value_boost`	Market-value quality scaling (boost 1.0037, quality 0.595)	`market_value_eur=16500000` `quality_factor=0.5953` `boost_factor=1.0037` `reference_quality=0.505` `history_matches=32` `history_shrink=0.1351`	-
`fbref_prior_blend`	FBref 2026-27 prior blended (weight 0.0117)	`blend_weight=0.0117` `history_matches=32` `season_overlap_discount=1.0` `per90_to_match_factor=1.0` `prior_reliability=0.1333` `competition=Premier League`	saves: 2.9224 -> 2.9116
`final_cap`	Stats clamped to historical constraints and shot-component identity	`stats_clamped=0` `shot_identity_applied=false`	-
`enriched_model_output`	Model output after duel-context enrichment (pre router post-processing)	`duel_context_applied=true` `duel_context={'opponent_faced': 'Igor Thiago, Jaidon Anthony', 'matchup_zone': 'Goalkeeper shot-stopping context', 'opponent_lambda': 1.4753}`	-
`scoreline_reset`	Team scoreline allocation: goal/assist counters reset to 0; the team's 1 goals are handed out as whole numbers by integer_goal_projection / integer_assist_projection	`scoreline_reset=true` `team_goals=1`	-
`team_lambda_rescale`	Player expected_goals/expected_assists rescaled to unconditional contributions so squad sum equals team lambda 1.55 (squad xG sum 1.3050, xG scale 1.1854, xA scale 1.4265, team xA 1.01 = 0.65 x lambda)	`scoreline_reset=true` `team_lambda=1.547` `squad_xg_sum=1.305` `squad_xa_sum=0.7049` `team_expected_assists=1.0055` `team_xa_to_xg_ratio=0.65`	shots_faced: 14.0 -> 13.6 expected_saves: 3.42 -> 3.33 high_pressure_saves: 1.18 -> 1.15
`rankings_attached`	Display-only provenance: no published number changes. Global ranking attached (match squad rank 2 by projected rating, fifa_overall 87.0)	`match_squad_rank=2` `squad_role=starter` `squad_rank_basis=projected_rating` `affects_projection=false` `ranking_source=sofascore_plus_fifa_csv` `ranking_status=combined`	-
`integer_goal_projection`	Team scoreline allocation: 1 projected goals distributed by scorer weight (xG x finishing x appearance)	`team_goals=1` `allocation_applied=true` `goals_allocated=0` `goal_probability=0.0` `scorer_weight=0.0` `squad_weight_total=1.6711`	-
`integer_assist_projection`	Team scoreline allocation: 1 projected assists distributed by provider weight (xA x creation x appearance; nobody assists his own goal)	`allocation_applied=true` `team_assists=1` `assists_allocated=0` `assist_probability=0.0` `provider_weight=0.0` `squad_weight_total=0.9033`	-

Team totals 1 1.55 15 6.9

Bench Candidates (7 modeled substitutes with empirical playing time)

Player	Pos	Exp Min	App Prob	Substitute Modeling Details
Giorgi Mamardashvili	G	5.5m	15.8%	Market-value substitute prior
Jeremie Frimpong	D	14.8m	35.6%	Market-value substitute prior
Curtis Jones	D	14.8m	35.3%	Market-value substitute prior
Dominik Szoboszlai	M	23.8m	58.0%	Market-value substitute prior
Ryan Gravenberch	M	23.8m	58.0%	Market-value substitute prior
Alexander Isak	F	27.1m	64.0%	Market-value substitute prior

Antagonist / tactical summary

Away midfield dictates play with ~51% possession expected (Yehor Yarmoliuk vs Florian Wirtz). Total antagonist xG — Home: 0.69, Away: 0.78. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Left Wing vs Right Back — Flank Battle.

Midfield — possession 49.1% / 50.9%, dominance ratio 0.44, key duel: Yehor Yarmoliuk vs Florian Wirtz

Mathias Jensen xG 0.07Igor Thiago xG 0.34Jaidon Anthony xG 0.18Florian Wirtz xG 0.19Bradley Barcola xG 0.13Lewis Koumas xG 0.14

Summary

Antagonist matchup analysis for Brentford vs Liverpool. Most likely scoreline: 1-1. Lambdas derived from 5 home and 6 away individual player duels. Head-to-head: Brentford 2W-2D-6L vs Liverpool over 10 meetings.

Tactical parity between 4-2-3-1 and 4-2-3-1. Match outcome driven primarily by individual player quality. Away midfield dictates play with ~51% possession expected (Yehor Yarmoliuk vs Florian Wirtz). Total antagonist xG — Home: 0.69, Away: 0.78. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Left Wing vs Right Back — Flank Battle.

Generated 2026-10-17 13:38:34 UTC · Soccer prediction reporting