Soccer prediction system deep audit

Slate 20261018 · immutable forensic review of 8 match prediction artifact(s)

Executive conclusion

No critical breaches. The pipeline is internally consistent across the audited slate.

At a glance

8matches audited

0critical

0errors

2warnings

0grid non-reproducible

0headline != modal

0winner disagreements

0lambda inconsistencies

0with real result

-winner hit-rate

-decisive hit-rate

75%predicted draws

-actual draws

0confident misses

-mean Brier

Matches

Brighton vs Crystal Palace

20261018_CRY@BHA · Premier League

WARNINGS risk 10 pred 1-1 (Draw)λ 1.43 / 1.22conf 0.7 warning 1 good 37

Problems (1)

Sev	Category	Rule	Message	Value	Expected
warning	provenance	`FBREF_PRIORS_ACTIVE`	FBref dataset files not found; team priors inactive.

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-competitive-floor	1.430	1.216
Pre-calibration	1.440	1.207
Final	1.430	1.216

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

Top scorelines	P
1-1	12.4%
1-0	10.0%
2-1	8.8%
0-1	8.5%
1-2	7.5%

Outcome — home 41.9% · draw 26.2% · away 31.9%

Markets — O2.5 49.3% · BTTS 53.6%

Passed checks (37)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Brighton: 11 starters.
good	structure	`FORMATION_VALID`	Brighton: formation 4-2-3-1 valid.
good	structure	`LINEUP_ELEVEN`	Crystal Palace: 11 starters.
good	structure	`FORMATION_VALID`	Crystal Palace: formation 3-4-2-1 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0000.
good	probability	`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.
good	probability	`SCORELINE_PROBS_NONNEG`	All scoreline probabilities within [0,1].
good	probability	`GRID_SUMS_ONE`	scoreline grid sums to 0.9965.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0012).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 1-1 is the modal scoreline.
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0020).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0035).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Draw' is a participating side.
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	predicted_winner 'Draw' follows the published modal scoreline 1-1. It differs from the calibrated 1X2 favorite 'Brighton', which is published separately and does not decide the call.
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.43, away 1.216).
good	provenance	`LAMBDA_FACTORS_PRESENT`	lambda_factors audit trail present.
good	model_behavior	`LAMBDA_CONSISTENCY`	Lambda identical across prediction / antagonist / factors.
good	model_behavior	`CALIBRATION_SCALE_BOUNDS`	Calibration scale 1.0 in range.
good	provenance	`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning present (XGBoost).
good	model_behavior	`LAMBDA_REASONING_CONSISTENCY`	Lambdas in xgboost_reasoning match final Poisson lambdas (1.43 / 1.216).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Brighton: player goals reconcile to team total (1).
good	integrity	`TEAM_GOALS_SUM`	Crystal Palace: player goals reconcile to team total (1).
good	structure	`POSSESSION_SUMS_100`	Possession sums to 100.0%.
good	integrity	`PLAYER_STAT_REALISM`	All 22 projected player lines pass realism checks.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Brighton: exactly one goalkeeper (Bart Verbruggen).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Brighton: 14 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Crystal Palace: exactly one goalkeeper (Walter Benítez).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Crystal Palace: 12 projected shots inside the realistic envelope.
good	provenance	`STANDING_AVAILABLE`	League-standing context available for both sides.
good	provenance	`RANKING_PROVENANCE`	22/22 starters carry a ranking source.
good	provenance	`DATA_DEPTH`	22/22 starters have adequate recent history.
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.65 ~ Poisson total 2.65.

Osasuna vs Racing Club

20261018_SAN@OSA · LaLiga

WARNINGS risk 10 pred 1-1 (Draw)λ 1.30 / 1.39conf 0.7 warning 1 good 37

Problems (1)

Sev	Category	Rule	Message	Value	Expected
warning	provenance	`FBREF_PRIORS_ACTIVE`	FBref dataset files not found; team priors inactive.

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-competitive-floor	1.305	1.390
Pre-calibration	1.343	1.351
Final	1.305	1.390

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

Top scorelines	P
1-1	12.4%
0-1	9.3%
1-0	8.7%
1-2	8.5%
2-1	8.0%

Outcome — home 35.0% · draw 26.1% · away 38.9%

Markets — O2.5 50.5% · BTTS 54.9%

Passed checks (37)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Osasuna: 11 starters.
good	structure	`FORMATION_VALID`	Osasuna: formation 4-3-3 valid.
good	structure	`LINEUP_ELEVEN`	Racing Club: 11 starters.
good	structure	`FORMATION_VALID`	Racing Club: formation 4-2-3-1 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0000.
good	probability	`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.
good	probability	`SCORELINE_PROBS_NONNEG`	All scoreline probabilities within [0,1].
good	probability	`GRID_SUMS_ONE`	scoreline grid sums to 0.9968.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0012).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 1-1 is the modal scoreline.
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0018).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0031).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Draw' is a participating side.
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	predicted_winner 'Draw' follows the published modal scoreline 1-1. It differs from the calibrated 1X2 favorite 'Racing Club', which is published separately and does not decide the call.
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.305, away 1.39).
good	provenance	`LAMBDA_FACTORS_PRESENT`	lambda_factors audit trail present.
good	model_behavior	`LAMBDA_CONSISTENCY`	Lambda identical across prediction / antagonist / factors.
good	model_behavior	`CALIBRATION_SCALE_BOUNDS`	Calibration scale 1.0 in range.
good	provenance	`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning present (XGBoost).
good	model_behavior	`LAMBDA_REASONING_CONSISTENCY`	Lambdas in xgboost_reasoning match final Poisson lambdas (1.305 / 1.39).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Osasuna: player goals reconcile to team total (1).
good	integrity	`TEAM_GOALS_SUM`	Racing Club: player goals reconcile to team total (1).
good	structure	`POSSESSION_SUMS_100`	Possession sums to 100.0%.
good	integrity	`PLAYER_STAT_REALISM`	All 22 projected player lines pass realism checks.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Osasuna: exactly one goalkeeper (Aitor Fernández).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Osasuna: 10 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Racing Club: exactly one goalkeeper (Julen Agirrezabala).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Racing Club: 15 projected shots inside the realistic envelope.
good	provenance	`STANDING_AVAILABLE`	League-standing context available for both sides.
good	provenance	`RANKING_PROVENANCE`	22/22 starters carry a ranking source.
good	provenance	`DATA_DEPTH`	21/22 starters have adequate recent history.
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.69 ~ Poisson total 2.69.

Celta vs Alavés

20261018_ALA@CEL · LaLiga

CLEAN risk 0 pred 0-0 (Draw)λ 0.86 / 0.85conf 0.69 good 38

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-competitive-floor	0.859	0.849
Pre-calibration	0.835	0.874
Final	0.859	0.849

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

Top scorelines	P
0-0	18.3%
1-0	15.4%
0-1	15.2%
1-1	13.4%
2-0	6.7%

Outcome — home 33.1% · draw 34.3% · away 32.6%

Markets — O2.5 24.5% · BTTS 33.1%

Passed checks (38)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Celta: 11 starters.
good	structure	`FORMATION_VALID`	Celta: formation 3-4-3 valid.
good	structure	`LINEUP_ELEVEN`	Alavés: 11 starters.
good	structure	`FORMATION_VALID`	Alavés: formation 5-3-2 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0000.
good	probability	`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.
good	probability	`SCORELINE_PROBS_NONNEG`	All scoreline probabilities within [0,1].
good	probability	`GRID_SUMS_ONE`	scoreline grid sums to 0.9980.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0016).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 0-0 is the modal scoreline.
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0010).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0020).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Draw' is a participating side.
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	predicted_winner 'Draw' follows the published modal scoreline 0-0. It differs from the calibrated 1X2 favorite 'Celta', which is published separately and does not decide the call.
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 0.859, away 0.849).
good	provenance	`LAMBDA_FACTORS_PRESENT`	lambda_factors audit trail present.
good	model_behavior	`LAMBDA_CONSISTENCY`	Lambda identical across prediction / antagonist / factors.
good	model_behavior	`CALIBRATION_SCALE_BOUNDS`	Calibration scale 1.0 in range.
good	provenance	`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning present (XGBoost).
good	model_behavior	`LAMBDA_REASONING_CONSISTENCY`	Lambdas in xgboost_reasoning match final Poisson lambdas (0.859 / 0.849).
good	provenance	`FBREF_PRIORS_ACTIVE`	FBref team priors applied (blend weight 0.228, home E[lambda]=0.309, away E[lambda]=0.309).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Celta: player goals reconcile to team total (0).
good	integrity	`TEAM_GOALS_SUM`	Alavés: player goals reconcile to team total (0).
good	structure	`POSSESSION_SUMS_100`	Possession sums to 100.0%.
good	integrity	`PLAYER_STAT_REALISM`	All 22 projected player lines pass realism checks.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Celta: exactly one goalkeeper (Ionuț Radu).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Celta: 11 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Alavés: exactly one goalkeeper (Antonio Sivera).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Alavés: 10 projected shots inside the realistic envelope.
good	provenance	`STANDING_AVAILABLE`	League-standing context available for both sides.
good	provenance	`RANKING_PROVENANCE`	22/22 starters carry a ranking source.
good	provenance	`DATA_DEPTH`	21/22 starters have adequate recent history.
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 1.71 ~ Poisson total 1.71.

Forest vs Arsenal

20261018_ARS@NFO · Premier League

CLEAN risk 0 pred 1-1 (Draw)λ 1.12 / 1.79conf 0.75 good 38

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-competitive-floor	1.117	1.792
Pre-calibration	1.150	1.742
Final	1.117	1.792

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

Top scorelines	P
1-1	11.0%
1-2	9.8%
0-1	9.7%
0-2	8.8%
2-1	6.1%

Outcome — home 23.5% · draw 23.3% · away 53.1%

Markets — O2.5 55.6% · BTTS 56.1%

Passed checks (38)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Forest: 11 starters.
good	structure	`FORMATION_VALID`	Forest: formation 3-4-2-1 valid.
good	structure	`LINEUP_ELEVEN`	Arsenal: 11 starters.
good	structure	`FORMATION_VALID`	Arsenal: formation 4-2-3-1 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0000.
good	probability	`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.
good	probability	`SCORELINE_PROBS_NONNEG`	All scoreline probabilities within [0,1].
good	probability	`GRID_SUMS_ONE`	scoreline grid sums to 0.9969.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0007).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 1-1 is the modal scoreline.
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0015).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0030).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Draw' is a participating side.
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	predicted_winner 'Draw' follows the published modal scoreline 1-1. It differs from the calibrated 1X2 favorite 'Arsenal', which is published separately and does not decide the call.
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.117, away 1.792).
good	provenance	`LAMBDA_FACTORS_PRESENT`	lambda_factors audit trail present.
good	model_behavior	`LAMBDA_CONSISTENCY`	Lambda identical across prediction / antagonist / factors.
good	model_behavior	`CALIBRATION_SCALE_BOUNDS`	Calibration scale 1.0 in range.
good	provenance	`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning present (XGBoost).
good	model_behavior	`LAMBDA_REASONING_CONSISTENCY`	Lambdas in xgboost_reasoning match final Poisson lambdas (1.117 / 1.792).
good	provenance	`FBREF_PRIORS_ACTIVE`	FBref team priors applied (blend weight 0.192, home E[lambda]=1.379, away E[lambda]=1.379).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Forest: player goals reconcile to team total (1).
good	integrity	`TEAM_GOALS_SUM`	Arsenal: player goals reconcile to team total (1).
good	structure	`POSSESSION_SUMS_100`	Possession sums to 100.0%.
good	integrity	`PLAYER_STAT_REALISM`	All 22 projected player lines pass realism checks.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Forest: exactly one goalkeeper (Matz Sels).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Forest: 12 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Arsenal: exactly one goalkeeper (David Raya).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Arsenal: 16 projected shots inside the realistic envelope.
good	provenance	`STANDING_AVAILABLE`	League-standing context available for both sides.
good	provenance	`RANKING_PROVENANCE`	22/22 starters carry a ranking source.
good	provenance	`DATA_DEPTH`	22/22 starters have adequate recent history.
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.91 ~ Poisson total 2.91.

Leeds vs Man Utd

20261018_MUN@LEE · Premier League

CLEAN risk 0 pred 1-1 (Draw)λ 1.21 / 1.52conf 0.72 good 38

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-competitive-floor	1.207	1.524
Pre-calibration	1.281	1.347
Final	1.207	1.524

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

Top scorelines	P
1-1	12.1%
0-1	9.8%
1-2	9.1%
1-0	7.8%
0-2	7.6%

Outcome — home 30.0% · draw 25.5% · away 44.5%

Markets — O2.5 51.4% · BTTS 54.9%

Passed checks (38)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Leeds: 11 starters.
good	structure	`FORMATION_VALID`	Leeds: formation 3-4-2-1 valid.
good	structure	`LINEUP_ELEVEN`	Man Utd: 11 starters.
good	structure	`FORMATION_VALID`	Man Utd: formation 4-2-3-1 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0000.
good	probability	`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.
good	probability	`SCORELINE_PROBS_NONNEG`	All scoreline probabilities within [0,1].
good	probability	`GRID_SUMS_ONE`	scoreline grid sums to 0.9963.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0011).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 1-1 is the modal scoreline.
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0018).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0037).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Draw' is a participating side.
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	predicted_winner 'Draw' follows the published modal scoreline 1-1. It differs from the calibrated 1X2 favorite 'Man Utd', which is published separately and does not decide the call.
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.207, away 1.524).
good	provenance	`LAMBDA_FACTORS_PRESENT`	lambda_factors audit trail present.
good	model_behavior	`LAMBDA_CONSISTENCY`	Lambda identical across prediction / antagonist / factors.
good	model_behavior	`CALIBRATION_SCALE_BOUNDS`	Calibration scale 1.0 in range.
good	provenance	`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning present (XGBoost).
good	model_behavior	`LAMBDA_REASONING_CONSISTENCY`	Lambdas in xgboost_reasoning match final Poisson lambdas (1.207 / 1.524).
good	provenance	`FBREF_PRIORS_ACTIVE`	FBref team priors applied (blend weight 0.192, home E[lambda]=1.724, away E[lambda]=1.379).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Leeds: player goals reconcile to team total (1).
good	integrity	`TEAM_GOALS_SUM`	Man Utd: player goals reconcile to team total (1).
good	structure	`POSSESSION_SUMS_100`	Possession sums to 100.0%.
good	integrity	`PLAYER_STAT_REALISM`	All 22 projected player lines pass realism checks.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Leeds: exactly one goalkeeper (James Trafford).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Leeds: 13 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Man Utd: exactly one goalkeeper (Senne Lammens).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Man Utd: 17 projected shots inside the realistic envelope.
good	provenance	`STANDING_AVAILABLE`	League-standing context available for both sides.
good	provenance	`RANKING_PROVENANCE`	22/22 starters carry a ranking source.
good	provenance	`DATA_DEPTH`	22/22 starters have adequate recent history.
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.73 ~ Poisson total 2.73.

Málaga vs Real Sociedad

20261018_RSO@MCF · LaLiga

CLEAN risk 0 pred 0-1 (Real Sociedad)λ 0.65 / 1.50conf 0.76 good 40

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-mismatch	0.701	1.452
Pre-competitive-floor	0.683	1.587
Pre-calibration	0.696	1.462
Final	0.650	1.505

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

Top scorelines	P
0-1	17.4%
0-2	13.1%
0-0	11.7%
1-1	11.4%
1-2	8.5%

Outcome — home 16.0% · draw 26.2% · away 57.9%

Markets — O2.5 36.5% · BTTS 37.2%

Passed checks (40)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Málaga: 11 starters.
good	structure	`FORMATION_VALID`	Málaga: formation 4-1-4-1 valid.
good	structure	`LINEUP_ELEVEN`	Real Sociedad: 11 starters.
good	structure	`FORMATION_VALID`	Real Sociedad: formation 4-4-2 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0000.
good	probability	`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.
good	probability	`SCORELINE_PROBS_NONNEG`	All scoreline probabilities within [0,1].
good	probability	`GRID_SUMS_ONE`	scoreline grid sums to 0.9970.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0007).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 0-1 is the modal scoreline.
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0018).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0032).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Real Sociedad' is a participating side.
good	model_behavior	`WINNER_MATCHES_SCORE`	Winner agrees with headline score (Real Sociedad).
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	predicted_winner 'Real Sociedad' follows the published modal scoreline 0-1.
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 0.65, away 1.505).
good	provenance	`LAMBDA_FACTORS_PRESENT`	lambda_factors audit trail present.
good	model_behavior	`LAMBDA_CONSISTENCY`	Lambda identical across prediction / antagonist / factors.
good	model_behavior	`CALIBRATION_SCALE_BOUNDS`	Calibration scale 1.0 in range.
good	provenance	`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning present (XGBoost).
good	model_behavior	`LAMBDA_REASONING_CONSISTENCY`	Lambdas in xgboost_reasoning match final Poisson lambdas (0.65 / 1.505).
good	provenance	`FBREF_PRIORS_ACTIVE`	FBref team priors applied (blend weight 0.228, home E[lambda]=0.2, away E[lambda]=0.2).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Málaga: player goals reconcile to team total (0).
good	integrity	`TEAM_GOALS_SUM`	Real Sociedad: player goals reconcile to team total (1).
good	structure	`POSSESSION_SUMS_100`	Possession sums to 100.0%.
good	integrity	`PLAYER_STAT_REALISM`	All 22 projected player lines pass realism checks.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Málaga: exactly one goalkeeper (Alfonso Herrero).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Málaga: 10 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Real Sociedad: exactly one goalkeeper (Álex Remiro).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Real Sociedad: 16 projected shots inside the realistic envelope.
good	provenance	`STANDING_AVAILABLE`	League-standing context available for both sides.
good	provenance	`RANKING_PROVENANCE`	22/22 starters carry a ranking source.
good	provenance	`DATA_DEPTH`	22/22 starters have adequate recent history.
good	provenance	`PREGAME_DIRECTION_AGREES`	Pregame favorite agrees with the model winner (Real Sociedad).
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.15 ~ Poisson total 2.15.

Real Madrid vs Sevilla

20261018_SEV@RMA · LaLiga

CLEAN risk 0 pred 3-1 (Real Madrid)λ 3.57 / 1.30conf 0.82 good 40

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-mismatch	3.333	1.536
Pre-calibration	3.272	1.470
Final	3.569	1.300

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

Top scorelines	P
3-1	7.6%
4-1	6.8%
2-1	6.4%
3-0	5.8%
4-0	5.2%

Outcome — home 79.0% · draw 11.4% · away 9.6%

Markets — O2.5 86.3% · BTTS 70.7%

Passed checks (40)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Real Madrid: 11 starters.
good	structure	`FORMATION_VALID`	Real Madrid: formation 4-2-3-1 valid.
good	structure	`LINEUP_ELEVEN`	Sevilla: 11 starters.
good	structure	`FORMATION_VALID`	Sevilla: formation 4-2-3-1 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0000.
good	probability	`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.
good	probability	`SCORELINE_PROBS_NONNEG`	All scoreline probabilities within [0,1].
good	probability	`GRID_SUMS_ONE`	scoreline grid sums to 0.9966.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0003).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 3-1 is the modal scoreline.
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0021).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0035).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Real Madrid' is a participating side.
good	model_behavior	`WINNER_MATCHES_SCORE`	Winner agrees with headline score (Real Madrid).
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	predicted_winner 'Real Madrid' follows the published modal scoreline 3-1.
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 3.569, away 1.3).
good	provenance	`LAMBDA_FACTORS_PRESENT`	lambda_factors audit trail present.
good	model_behavior	`LAMBDA_CONSISTENCY`	Lambda identical across prediction / antagonist / factors.
good	model_behavior	`CALIBRATION_SCALE_BOUNDS`	Calibration scale 1.0 in range.
good	provenance	`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning present (XGBoost).
good	model_behavior	`LAMBDA_REASONING_CONSISTENCY`	Lambdas in xgboost_reasoning match final Poisson lambdas (3.569 / 1.3).
good	provenance	`FBREF_PRIORS_ACTIVE`	FBref team priors applied (blend weight 0.228, home E[lambda]=3.5, away E[lambda]=3.5).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Real Madrid: player goals reconcile to team total (3).
good	integrity	`TEAM_GOALS_SUM`	Sevilla: player goals reconcile to team total (1).
good	structure	`POSSESSION_SUMS_100`	Possession sums to 100.0%.
good	integrity	`PLAYER_STAT_REALISM`	All 22 projected player lines pass realism checks.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Real Madrid: exactly one goalkeeper (Thibaut Courtois).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Real Madrid: 19 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Sevilla: exactly one goalkeeper (Odysseas Vlachodimos).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Sevilla: 13 projected shots inside the realistic envelope.
good	provenance	`STANDING_AVAILABLE`	League-standing context available for both sides.
good	provenance	`RANKING_PROVENANCE`	22/22 starters carry a ranking source.
good	provenance	`DATA_DEPTH`	22/22 starters have adequate recent history.
good	provenance	`PREGAME_DIRECTION_AGREES`	Pregame favorite agrees with the model winner (Real Madrid).
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 4.87 ~ Poisson total 4.87.

Bournemouth vs Sunderland

20261018_SUN@BOU · Premier League

CLEAN risk 0 pred 1-1 (Draw)λ 1.28 / 0.99conf 0.71 good 38

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-competitive-floor	1.279	0.986
Pre-calibration	1.279	0.986
Final	1.279	0.986

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

Top scorelines	P
1-1	13.2%
1-0	13.2%
0-0	10.5%
0-1	10.1%
2-0	8.5%

Outcome — home 43.0% · draw 28.5% · away 28.5%

Markets — O2.5 39.5% · BTTS 45.4%

Passed checks (38)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Bournemouth: 11 starters.
good	structure	`FORMATION_VALID`	Bournemouth: formation 4-2-3-1 valid.
good	structure	`LINEUP_ELEVEN`	Sunderland: 11 starters.
good	structure	`FORMATION_VALID`	Sunderland: formation 4-2-3-1 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0000.
good	probability	`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.
good	probability	`SCORELINE_PROBS_NONNEG`	All scoreline probabilities within [0,1].
good	probability	`GRID_SUMS_ONE`	scoreline grid sums to 0.9971.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0012).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 1-1 is the modal scoreline.
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0016).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0030).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Draw' is a participating side.
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	predicted_winner 'Draw' follows the published modal scoreline 1-1. It differs from the calibrated 1X2 favorite 'Bournemouth', which is published separately and does not decide the call.
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.279, away 0.986).
good	provenance	`LAMBDA_FACTORS_PRESENT`	lambda_factors audit trail present.
good	model_behavior	`LAMBDA_CONSISTENCY`	Lambda identical across prediction / antagonist / factors.
good	model_behavior	`CALIBRATION_SCALE_BOUNDS`	Calibration scale 1.0 in range.
good	provenance	`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning present (XGBoost).
good	model_behavior	`LAMBDA_REASONING_CONSISTENCY`	Lambdas in xgboost_reasoning match final Poisson lambdas (1.279 / 0.986).
good	provenance	`FBREF_PRIORS_ACTIVE`	FBref team priors applied (blend weight 0.192, home E[lambda]=0.69, away E[lambda]=0.69).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Bournemouth: player goals reconcile to team total (1).
good	integrity	`TEAM_GOALS_SUM`	Sunderland: player goals reconcile to team total (1).
good	structure	`POSSESSION_SUMS_100`	Possession sums to 100.0%.
good	integrity	`PLAYER_STAT_REALISM`	All 22 projected player lines pass realism checks.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Bournemouth: exactly one goalkeeper (Đorđe Petrović).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Bournemouth: 14 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Sunderland: exactly one goalkeeper (Robin Roefs).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Sunderland: 14 projected shots inside the realistic envelope.
good	provenance	`STANDING_AVAILABLE`	League-standing context available for both sides.
good	provenance	`RANKING_PROVENANCE`	21/22 starters carry a ranking source.
good	provenance	`DATA_DEPTH`	21/22 starters have adequate recent history.
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.27 ~ Poisson total 2.26.

Rule catalogue

Rule	Description	Triggered
`SCHEMA_KEYS_PRESENT`	All required top-level sections exist in the artifact.	ok
`LINEUP_ELEVEN`	Each side fields exactly 11 starters.	ok
`FORMATION_VALID`	Formation string parses and its outfield lines sum to 10.	ok
`POSSESSION_SUMS_100`	Estimated possession of both teams sums to ~100%.	ok
`TEAM_GOALS_SUM`	Players' projected integer goals sum to the team's headline score.	ok
`TEAM_TOTALS_SCORE_MATCH`	team_totals.projected_goals equals the headline score component.	ok
`TEAM_EG_LAMBDA_MATCH`	team_totals.expected_goals matches the team's Poisson lambda.	ok
`PLAYER_STAT_REALISM`	Projected player lines are internally consistent (xG<=shots, accurate<=total passes, minutes/rating in range).	ok
`EXACTLY_ONE_GOALKEEPER`	Each starting XI fields exactly one goalkeeper.	ok
`SHOT_MATH_INTEGRITY`	Shot components are physically consistent (on target <= total, on+off <= total; blocked shots cannot be negative).	ok
`PLAYER_MINUTES_LEAGUE_CAP`	No player projects more minutes than the competition clock allows (90 + stoppage, or extra time in cups).	ok
`TEAM_SHOTS_CAP`	Team projected shots stay inside the realistic historical envelope (~35).	ok
`OUTCOME_SUMS_ONE`	home_win + draw + away_win sums to 1.	ok
`MARKETS_COMPLEMENTARY`	Over/Under and BTTS pairs each sum to 1.	ok
`GRID_SUMS_ONE`	The scoreline probability grid sums to ~1.	ok
`SCORELINE_PROBS_NONNEG`	Every scoreline probability is a valid [0,1] value.	ok
`GRID_REPRODUCIBLE`	The reported scoreline grid is reproducible from the lambdas.	ok
`MOST_LIKELY_IS_MODE`	The headline most-likely score is the distribution's modal scoreline.	ok
`MLS_PROBABILITY_MATCHES`	most_likely_score_probability equals the grid value of that score.	ok
`OUTCOME_MATCHES_GRID`	The reported win/draw/loss split is derivable from the grid.	ok
`MARKETS_MATCH_GRID`	The reported Over/Under and BTTS are derivable from the grid.	ok
`WINNER_MATCHES_SCORE`	predicted_winner agrees with the headline score direction.	ok
`WINNER_PROBABILITY_DIRECTION`	predicted_winner derives from the published modal scoreline; diverging from the 1X2 favorite is expected.	ok
`WINNER_IS_VALID_TEAM`	predicted_winner names a participating team (or Draw).	ok
`MODAL_OUTCOME_IS_DRAW`	Draw is the single largest outcome bucket while a team is named.	ok
`LAMBDA_BOUNDS`	Both lambdas fall inside the engine's [0.55, 5.00] envelope.	ok
`LAMBDA_CONSISTENCY`	The lambda is identical across prediction / antagonist / lambda_factors.	ok
`LAMBDA_FACTORS_PRESENT`	The lambda_factors decomposition (audit trail) is present.	ok
`CALIBRATION_SCALE_BOUNDS`	competition_calibration_scale stays within [0.85, 1.35].	ok
`CALIBRATION_RECORDED`	When calibration ran, the pre-calibration lambdas are recorded.	ok
`STANDING_AVAILABLE`	League-standing context is available for a domestic fixture.	ok
`RANKING_PROVENANCE`	Player ranking/quality signals carry a recognised source.	ok
`DATA_DEPTH`	Starters have enough recent games to model rather than synthesise.	ok
`FBREF_PRIORS_ACTIVE`	FBref season-prior files exist so that branch is not silently inert.	2 problems
`XGBOOST_REASONING_PRESENT`	Match-level xgboost_reasoning payload is present with full tracing pipeline.	ok
`XGBOOST_PLAYER_REASONING`	Starters have xgboost_reasoning metadata tracking feature contributions and FBref priors.	ok
`LAMBDA_REASONING_CONSISTENCY`	Lambdas reported in xgboost_reasoning match the final Poisson lambdas.	ok
`PREGAME_DIRECTION_AGREES`	The pregame fair-line favorite agrees with the model winner.	ok
`PREGAME_TOTAL_CONSISTENT`	The pregame expected total is close to the sum of the lambdas.	ok
`RESULT_WINNER_HIT`	The predicted winner matched the real result.	ok
`RESULT_EXACT_SCORE_HIT`	The headline score matched the real scoreline exactly.	ok
`RESULT_OU_HIT`	The Over/Under 2.5 lean matched reality.	ok
`RESULT_BTTS_HIT`	The BTTS lean matched reality.	ok
`RESULT_CONFIDENT_MISS`	A high-confidence prediction did not reach the real outcome.	ok
`OUTCOME_BRIER`	Brier score of the win/draw/loss probabilities vs the real result.	ok

Generated 2026-10-18 13:37:43 UTC · Soccer prediction reporting