Soccer prediction system deep audit

Slate 20260913 · immutable forensic review of 6 match prediction artifact(s)

Executive conclusion

In 3 match(es) the headline score is not the distribution's modal scoreline (decisive-score substitution or a selection issue) — see the MOST_LIKELY_IS_MODE findings.

At a glance

6matches audited

0critical

0errors

2warnings

0grid non-reproducible

3headline != modal

0winner disagreements

0lambda inconsistencies

0with real result

-winner hit-rate

-decisive hit-rate

0%predicted draws

-actual draws

0confident misses

-mean Brier

Matches

Coventry vs Brighton

20260913_BHA@COC · Premier League

WARNINGS risk 10 pred 0-1 (Brighton)λ 1.05 / 1.66conf 0.66 warning 1 info 1 good 38

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	0.892	1.657
Pre-calibration	0.892	1.657
Final	1.050	1.657

competition_calibration_scale: 1.0competition_observed_total_goals: 0.0quality_gap_applied: Truemismatch_score_pressure_applied: Falsecompetitive_underdog_floor_applied: Truestage_discount: 1.0

Scoreline distribution

Top scorelines	P
0-1	11.0%
1-1	11.7%
1-2	9.6%
0-2	9.2%
1-0	6.9%

Outcome — home 23.9% · draw 24.6% · away 51.5%

Markets — O2.5 50.8% · BTTS 52.7%

Passed checks (38)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Coventry: 11 starters.
good	structure	`FORMATION_VALID`	Coventry: formation 3-4-3 valid.
good	structure	`LINEUP_ELEVEN`	Brighton: 11 starters.
good	structure	`FORMATION_VALID`	Brighton: formation 4-2-3-1 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0001.
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 0.0009).
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0026).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0035).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Brighton' is a participating side.
good	model_behavior	`WINNER_MATCHES_SCORE`	Winner agrees with headline score (Brighton).
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	Winner matches the win-probability favorite (Brighton).
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.05, away 1.657).
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 (1.05 / 1.657).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Coventry: player goals reconcile to team total (0).
good	integrity	`TEAM_GOALS_SUM`	Brighton: 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`	Coventry: exactly one goalkeeper (Carl Rushworth).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Coventry: 15 projected shots inside the realistic envelope.
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: 25 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`	17/22 starters have adequate recent history.
good	provenance	`PREGAME_DIRECTION_AGREES`	Pregame favorite agrees with the model winner (Brighton).
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.71 ~ Poisson total 2.71.

Notes (1)

Sev	Category	Rule	Message	Value	Expected
info	model_behavior	`MOST_LIKELY_IS_MODE`	Decisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.1170); the headline shows the most likely decisive score 0-1 (p=0.1097) for the probability favorite Brighton.	0-1	1-1

Getafe vs Deportivo

20260913_COR@GET · LaLiga

WARNINGS risk 10 pred 0-1 (Deportivo)λ 1.02 / 1.13conf 0.63 warning 1 info 1 good 38

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	0.857	1.134
Pre-calibration	0.857	1.134
Final	1.020	1.134

competition_calibration_scale: 1.0competition_observed_total_goals: 0.0quality_gap_applied: Falsemismatch_score_pressure_applied: Falsecompetitive_underdog_floor_applied: Truestage_discount: 1.0

Scoreline distribution

Top scorelines	P
0-1	13.0%
1-1	13.6%
0-0	11.7%
1-0	11.7%
1-2	7.6%

Outcome — home 32.3% · draw 29.7% · away 38.0%

Markets — O2.5 36.5% · BTTS 43.5%

Passed checks (38)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Getafe: 11 starters.
good	structure	`FORMATION_VALID`	Getafe: formation 5-3-2 valid.
good	structure	`LINEUP_ELEVEN`	Deportivo: 11 starters.
good	structure	`FORMATION_VALID`	Deportivo: 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 0.9973.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0013).
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 0.0028).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Deportivo' is a participating side.
good	model_behavior	`WINNER_MATCHES_SCORE`	Winner agrees with headline score (Deportivo).
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	Winner matches the win-probability favorite (Deportivo).
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.02, away 1.134).
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 (1.02 / 1.134).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Getafe: player goals reconcile to team total (0).
good	integrity	`TEAM_GOALS_SUM`	Deportivo: 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`	Getafe: exactly one goalkeeper (David Soria).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Getafe: 14 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Deportivo: exactly one goalkeeper (Leo Román).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Deportivo: 11 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`	18/22 starters have adequate recent history.
good	provenance	`PREGAME_DIRECTION_AGREES`	Pregame favorite agrees with the model winner (Deportivo).
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.15 ~ Poisson total 2.15.

Notes (1)

Sev	Category	Rule	Message	Value	Expected
info	model_behavior	`MOST_LIKELY_IS_MODE`	Decisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.1355); the headline shows the most likely decisive score 0-1 (p=0.1303) for the probability favorite Deportivo.	0-1	1-1

Real Sociedad vs Atl. Madrid

20260913_ATM@RSO · LaLiga

CLEAN risk 0 pred 0-1 (Atl. Madrid)λ 1.02 / 1.59conf 0.67 info 1 good 39

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-mismatch	0.943	1.441
Pre-competitive-floor	0.915	1.593
Pre-calibration	0.943	1.441
Final	1.020	1.593

competition_calibration_scale: 1.0competition_observed_total_goals: 0.0quality_gap_applied: Truemismatch_score_pressure_applied: Truecompetitive_underdog_floor_applied: Truestage_discount: 1.0

Scoreline distribution

Top scorelines	P
0-1	11.6%
1-1	12.0%
1-2	9.5%
0-2	9.3%
0-0	7.4%

Outcome — home 24.2% · draw 25.2% · away 50.6%

Markets — O2.5 48.5% · BTTS 51.0%

Passed checks (39)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Real Sociedad: 11 starters.
good	structure	`FORMATION_VALID`	Real Sociedad: formation 4-4-2 valid.
good	structure	`LINEUP_ELEVEN`	Atl. Madrid: 11 starters.
good	structure	`FORMATION_VALID`	Atl. Madrid: 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 0.9972.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0010).
good	probability	`MLS_PROBABILITY_MATCHES`	Headline probability matches the grid.
good	model_behavior	`OUTCOME_MATCHES_GRID`	Outcome split matches the grid (max diff 0.0017).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0028).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Atl. Madrid' is a participating side.
good	model_behavior	`WINNER_MATCHES_SCORE`	Winner agrees with headline score (Atl. Madrid).
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	Winner matches the win-probability favorite (Atl. Madrid).
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.02, away 1.593).
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 (1.02 / 1.593).
good	provenance	`FBREF_PRIORS_ACTIVE`	FBref team priors applied (blend weight 0.228, home E[lambda]=1.294, away E[lambda]=1.296).
good	provenance	`XGBOOST_PLAYER_REASONING`	All 22 starters carry full XGBoost / FBref reasoning.
good	integrity	`TEAM_GOALS_SUM`	Real Sociedad: player goals reconcile to team total (0).
good	integrity	`TEAM_GOALS_SUM`	Atl. Madrid: 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 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: 10 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Atl. Madrid: exactly one goalkeeper (Jan Oblak).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Atl. Madrid: 18 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`	17/22 starters have adequate recent history.
good	provenance	`PREGAME_DIRECTION_AGREES`	Pregame favorite agrees with the model winner (Atl. Madrid).
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 2.61 ~ Poisson total 2.61.

Notes (1)

Sev	Category	Rule	Message	Value	Expected
info	model_behavior	`MOST_LIKELY_IS_MODE`	Decisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.1201); the headline shows the most likely decisive score 0-1 (p=0.1159) for the probability favorite Atl. Madrid.	0-1	1-1

Levante vs Barcelona

20260913_BAR@LEV · LaLiga

CLEAN risk 0 pred 1-3 (Barcelona)λ 1.27 / 3.84conf 0.77 good 40

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-mismatch	1.353	3.248
Pre-calibration	1.353	3.248
Final	1.271	3.838

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
1-3	7.3%
1-4	7.0%
0-3	5.7%
1-2	5.7%
0-4	5.5%

Outcome — home 7.8% · draw 9.9% · away 82.3%

Markets — O2.5 88.3% · BTTS 70.4%

Passed checks (40)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Levante: 11 starters.
good	structure	`FORMATION_VALID`	Levante: formation 4-1-4-1 valid.
good	structure	`LINEUP_ELEVEN`	Barcelona: 11 starters.
good	structure	`FORMATION_VALID`	Barcelona: formation 4-2-3-1 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0001.
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 0.0005).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 1-3 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 0.0034).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Barcelona' is a participating side.
good	model_behavior	`WINNER_MATCHES_SCORE`	Winner agrees with headline score (Barcelona).
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	Winner matches the win-probability favorite (Barcelona).
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.271, away 3.838).
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 (1.271 / 3.838).
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`	Levante: player goals reconcile to team total (1).
good	integrity	`TEAM_GOALS_SUM`	Barcelona: player goals reconcile to team total (3).
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`	Levante: exactly one goalkeeper (Mathew Ryan).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Levante: 9 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Barcelona: exactly one goalkeeper (Joan García).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Barcelona: 14 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 (Barcelona).
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 5.11 ~ Poisson total 5.11.

Celta vs Málaga

20260913_MCF@CEL · LaLiga

CLEAN risk 0 pred 0-1 (Málaga)λ 0.82 / 1.02conf 0.64 good 40

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-competitive-floor	0.823	0.650
Pre-calibration	0.823	0.600
Final	0.823	1.020

competition_calibration_scale: 1.0competition_observed_total_goals: 0.0quality_gap_applied: Truemismatch_score_pressure_applied: Falsecompetitive_underdog_floor_applied: Truestage_discount: 1.0

Scoreline distribution

Top scorelines	P
0-1	16.0%
0-0	16.0%
1-1	13.4%
1-0	12.9%
0-2	8.2%

Outcome — home 28.4% · draw 32.5% · away 39.1%

Markets — O2.5 28.1% · BTTS 36.0%

Passed checks (40)

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`	Málaga: 11 starters.
good	structure	`FORMATION_VALID`	Málaga: formation 4-3-3 valid.
good	probability	`OUTCOME_SUMS_ONE`	outcome sums to 1.0001.
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 0.9975.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0015).
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 0.0013).
good	model_behavior	`MARKETS_MATCH_GRID`	Markets match the grid (max diff 0.0025).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Málaga' is a participating side.
good	model_behavior	`WINNER_MATCHES_SCORE`	Winner agrees with headline score (Málaga).
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	Winner matches the win-probability favorite (Málaga).
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 0.823, away 1.02).
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 (0.823 / 1.02).
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`	Celta: player goals reconcile to team total (0).
good	integrity	`TEAM_GOALS_SUM`	Málaga: 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`	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: 17 projected shots inside the realistic envelope.
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: 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`	20/22 starters have adequate recent history.
good	provenance	`PREGAME_DIRECTION_AGREES`	Pregame favorite agrees with the model winner (Málaga).
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 1.84 ~ Poisson total 1.84.

Man Utd vs Man City

20260913_MCI@MUN · Premier League

CLEAN risk 0 pred 1-2 (Man City)λ 1.49 / 2.42conf 0.68 good 40

Problems (0)

none

Lambda evolution (audit trail)

Stage	Home λ	Away λ
Pre-competitive-floor	1.489	2.424
Pre-calibration	1.489	2.424
Final	1.489	2.424

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
1-2	8.7%
1-1	7.3%
1-3	7.1%
2-2	6.5%
0-2	5.9%

Outcome — home 22.8% · draw 19.1% · away 58.1%

Markets — O2.5 74.9% · BTTS 70.6%

Passed checks (40)

Sev	Category	Rule	Message
good	integrity	`SCHEMA_KEYS_PRESENT`	All required sections present.
good	structure	`LINEUP_ELEVEN`	Man Utd: 11 starters.
good	structure	`FORMATION_VALID`	Man Utd: formation 4-2-3-1 valid.
good	structure	`LINEUP_ELEVEN`	Man City: 11 starters.
good	structure	`FORMATION_VALID`	Man City: 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 0.9962.
good	model_behavior	`GRID_REPRODUCIBLE`	Grid reproducible from lambdas (max cell diff 0.0005).
good	model_behavior	`MOST_LIKELY_IS_MODE`	Headline score 1-2 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 0.0038).
good	integrity	`WINNER_IS_VALID_TEAM`	predicted_winner 'Man City' is a participating side.
good	model_behavior	`WINNER_MATCHES_SCORE`	Winner agrees with headline score (Man City).
good	model_behavior	`WINNER_PROBABILITY_DIRECTION`	Winner matches the win-probability favorite (Man City).
good	model_behavior	`LAMBDA_BOUNDS`	Lambdas in range (home 1.489, away 2.424).
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 (1.489 / 2.424).
good	provenance	`FBREF_PRIORS_ACTIVE`	FBref team priors applied (blend weight 0.192, 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`	Man Utd: player goals reconcile to team total (1).
good	integrity	`TEAM_GOALS_SUM`	Man City: player goals reconcile to team total (2).
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`	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: 20 projected shots inside the realistic envelope.
good	structure	`EXACTLY_ONE_GOALKEEPER`	Man City: exactly one goalkeeper (Gianluigi Donnarumma).
good	integrity	`SHOT_MATH_INTEGRITY`	All 11 projected lines have consistent shot components.
good	integrity	`TEAM_SHOTS_CAP`	Man City: 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`	20/22 starters have adequate recent history.
good	provenance	`PREGAME_DIRECTION_AGREES`	Pregame favorite agrees with the model winner (Man City).
good	provenance	`PREGAME_TOTAL_CONSISTENT`	Pregame total 3.91 ~ Poisson total 3.91.

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 is the team with the higher win probability.	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 2026-09-13 13:34:12 UTC · Soccer prediction reporting