Soccer prediction system deep audit

Slate 20260918 · immutable forensic review of 2 match prediction artifact(s)

Executive conclusion

In 2 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

2matches audited
0critical
0errors
1warnings
0grid non-reproducible
2headline != modal
0winner disagreements
0lambda inconsistencies
0with real result
-winner hit-rate
-decisive hit-rate
0%predicted draws
-actual draws
0confident misses
-mean Brier
 

Matches

Espanyol vs Elche

20260918_ELC@ESP · LaLiga
WARNINGS risk 10 pred 1-0 (Espanyol)λ 1.37 / 1.02conf 0.58 warning 1 info 1 good 38
Problems (1)
SevCategoryRuleMessageValueExpected
warningprovenanceDATA_DEPTH9/22 starters have <3 recent games; their stats lean on market-value / prior synthesis rather than form.9
Lambda evolution (audit trail)
StageHome λAway λ
Pre-competitive-floor1.3730.935
Pre-calibration1.3730.935
Final1.3731.020
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 scorelinesP
1-012.4%
1-112.9%
0-09.2%
0-19.2%
2-18.8%
Outcome — home 44.8% · draw 27.4% · away 27.8%
Markets — O2.5 42.9% · BTTS 47.9%
Passed checks (38)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENEspanyol: 11 starters.
goodstructureFORMATION_VALIDEspanyol: formation 3-5-2 valid.
goodstructureLINEUP_ELEVENElche: 11 starters.
goodstructureFORMATION_VALIDElche: formation 4-2-3-1 valid.
goodprobabilityOUTCOME_SUMS_ONEoutcome sums to 1.0000.
goodprobabilityMARKETS_COMPLEMENTARYOver/Under and BTTS pairs each sum to 1.
goodprobabilitySCORELINE_PROBS_NONNEGAll scoreline probabilities within [0,1].
goodprobabilityGRID_SUMS_ONEscoreline grid sums to 0.9974.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0011).
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0014).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0025).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Espanyol' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Espanyol).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Espanyol).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.373, away 1.02).
goodprovenanceLAMBDA_FACTORS_PRESENTlambda_factors audit trail present.
goodmodel_behaviorLAMBDA_CONSISTENCYLambda identical across prediction / antagonist / factors.
goodmodel_behaviorCALIBRATION_SCALE_BOUNDSCalibration scale 1.0 in range.
goodprovenanceXGBOOST_REASONING_PRESENTMatch-level xgboost_reasoning present (XGBoost).
goodmodel_behaviorLAMBDA_REASONING_CONSISTENCYLambdas in xgboost_reasoning match final Poisson lambdas (1.373 / 1.02).
goodprovenanceFBREF_PRIORS_ACTIVEFBref team priors applied (blend weight 0.228, home E[lambda]=1.235, away E[lambda]=1.237).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMEspanyol: player goals reconcile to team total (1).
goodintegrityTEAM_GOALS_SUMElche: player goals reconcile to team total (0).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodintegrityPLAYER_STAT_REALISMAll 22 projected player lines pass realism checks.
goodstructureEXACTLY_ONE_GOALKEEPEREspanyol: exactly one goalkeeper (Marko Dmitrović).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPEspanyol: 10 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERElche: exactly one goalkeeper (Alejandro Iturbe).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPElche: 10 projected shots inside the realistic envelope.
goodprovenanceSTANDING_AVAILABLELeague-standing context available for both sides.
goodprovenanceRANKING_PROVENANCE22/22 starters carry a ranking source.
goodprovenancePREGAME_DIRECTION_AGREESPregame favorite agrees with the model winner (Espanyol).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 2.39 ~ Poisson total 2.39.
Notes (1)
SevCategoryRuleMessageValueExpected
infomodel_behaviorMOST_LIKELY_IS_MODEDecisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.1290); the headline shows the most likely decisive score 1-0 (p=0.1243) for the probability favorite Espanyol.1-01-1

Brentford vs Chelsea

20260918_CHE@BRE · Premier League
CLEAN risk 0 pred 1-2 (Chelsea)λ 1.74 / 1.82conf 0.56 info 1 good 39
Problems (0)
none
Lambda evolution (audit trail)
StageHome λAway λ
Pre-mismatch1.7781.680
Pre-competitive-floor1.7421.816
Pre-calibration1.7781.680
Final1.7421.816
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 scorelinesP
1-28.2%
1-19.1%
2-17.8%
2-27.1%
0-15.1%
Outcome — home 37.4% · draw 22.2% · away 40.4%
Markets — O2.5 69.0% · BTTS 69.1%
Passed checks (39)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENBrentford: 11 starters.
goodstructureFORMATION_VALIDBrentford: formation 4-2-3-1 valid.
goodstructureLINEUP_ELEVENChelsea: 11 starters.
goodstructureFORMATION_VALIDChelsea: formation 4-2-3-1 valid.
goodprobabilityOUTCOME_SUMS_ONEoutcome sums to 1.0000.
goodprobabilityMARKETS_COMPLEMENTARYOver/Under and BTTS pairs each sum to 1.
goodprobabilitySCORELINE_PROBS_NONNEGAll scoreline probabilities within [0,1].
goodprobabilityGRID_SUMS_ONEscoreline grid sums to 0.9959.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0008).
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0025).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0040).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Chelsea' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Chelsea).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Chelsea).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.742, away 1.816).
goodprovenanceLAMBDA_FACTORS_PRESENTlambda_factors audit trail present.
goodmodel_behaviorLAMBDA_CONSISTENCYLambda identical across prediction / antagonist / factors.
goodmodel_behaviorCALIBRATION_SCALE_BOUNDSCalibration scale 1.0 in range.
goodprovenanceXGBOOST_REASONING_PRESENTMatch-level xgboost_reasoning present (XGBoost).
goodmodel_behaviorLAMBDA_REASONING_CONSISTENCYLambdas in xgboost_reasoning match final Poisson lambdas (1.742 / 1.816).
goodprovenanceFBREF_PRIORS_ACTIVEFBref team priors applied (blend weight 0.192, home E[lambda]=3.5, away E[lambda]=3.5).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMBrentford: player goals reconcile to team total (1).
goodintegrityTEAM_GOALS_SUMChelsea: player goals reconcile to team total (2).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodintegrityPLAYER_STAT_REALISMAll 22 projected player lines pass realism checks.
goodstructureEXACTLY_ONE_GOALKEEPERBrentford: exactly one goalkeeper (Caoimhin Kelleher).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPBrentford: 18 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERChelsea: exactly one goalkeeper (Mike Penders).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPChelsea: 14 projected shots inside the realistic envelope.
goodprovenanceSTANDING_AVAILABLELeague-standing context available for both sides.
goodprovenanceRANKING_PROVENANCE22/22 starters carry a ranking source.
goodprovenanceDATA_DEPTH16/22 starters have adequate recent history.
goodprovenancePREGAME_DIRECTION_AGREESPregame favorite agrees with the model winner (Chelsea).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 3.56 ~ Poisson total 3.56.
Notes (1)
SevCategoryRuleMessageValueExpected
infomodel_behaviorMOST_LIKELY_IS_MODEDecisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.0909); the headline shows the most likely decisive score 1-2 (p=0.0819) for the probability favorite Chelsea.1-21-1

Rule catalogue

RuleDescriptionTriggered
SCHEMA_KEYS_PRESENTAll required top-level sections exist in the artifact.ok
LINEUP_ELEVENEach side fields exactly 11 starters.ok
FORMATION_VALIDFormation string parses and its outfield lines sum to 10.ok
POSSESSION_SUMS_100Estimated possession of both teams sums to ~100%.ok
TEAM_GOALS_SUMPlayers' projected integer goals sum to the team's headline score.ok
TEAM_TOTALS_SCORE_MATCHteam_totals.projected_goals equals the headline score component.ok
TEAM_EG_LAMBDA_MATCHteam_totals.expected_goals matches the team's Poisson lambda.ok
PLAYER_STAT_REALISMProjected player lines are internally consistent (xG<=shots, accurate<=total passes, minutes/rating in range).ok
EXACTLY_ONE_GOALKEEPEREach starting XI fields exactly one goalkeeper.ok
SHOT_MATH_INTEGRITYShot components are physically consistent (on target <= total, on+off <= total; blocked shots cannot be negative).ok
PLAYER_MINUTES_LEAGUE_CAPNo player projects more minutes than the competition clock allows (90 + stoppage, or extra time in cups).ok
TEAM_SHOTS_CAPTeam projected shots stay inside the realistic historical envelope (~35).ok
OUTCOME_SUMS_ONEhome_win + draw + away_win sums to 1.ok
MARKETS_COMPLEMENTARYOver/Under and BTTS pairs each sum to 1.ok
GRID_SUMS_ONEThe scoreline probability grid sums to ~1.ok
SCORELINE_PROBS_NONNEGEvery scoreline probability is a valid [0,1] value.ok
GRID_REPRODUCIBLEThe reported scoreline grid is reproducible from the lambdas.ok
MOST_LIKELY_IS_MODEThe headline most-likely score is the distribution's modal scoreline.ok
MLS_PROBABILITY_MATCHESmost_likely_score_probability equals the grid value of that score.ok
OUTCOME_MATCHES_GRIDThe reported win/draw/loss split is derivable from the grid.ok
MARKETS_MATCH_GRIDThe reported Over/Under and BTTS are derivable from the grid.ok
WINNER_MATCHES_SCOREpredicted_winner agrees with the headline score direction.ok
WINNER_PROBABILITY_DIRECTIONpredicted_winner is the team with the higher win probability.ok
WINNER_IS_VALID_TEAMpredicted_winner names a participating team (or Draw).ok
MODAL_OUTCOME_IS_DRAWDraw is the single largest outcome bucket while a team is named.ok
LAMBDA_BOUNDSBoth lambdas fall inside the engine's [0.55, 5.00] envelope.ok
LAMBDA_CONSISTENCYThe lambda is identical across prediction / antagonist / lambda_factors.ok
LAMBDA_FACTORS_PRESENTThe lambda_factors decomposition (audit trail) is present.ok
CALIBRATION_SCALE_BOUNDScompetition_calibration_scale stays within [0.85, 1.35].ok
CALIBRATION_RECORDEDWhen calibration ran, the pre-calibration lambdas are recorded.ok
STANDING_AVAILABLELeague-standing context is available for a domestic fixture.ok
RANKING_PROVENANCEPlayer ranking/quality signals carry a recognised source.ok
DATA_DEPTHStarters have enough recent games to model rather than synthesise.1 problems
FBREF_PRIORS_ACTIVEFBref season-prior files exist so that branch is not silently inert.ok
XGBOOST_REASONING_PRESENTMatch-level xgboost_reasoning payload is present with full tracing pipeline.ok
XGBOOST_PLAYER_REASONINGStarters have xgboost_reasoning metadata tracking feature contributions and FBref priors.ok
LAMBDA_REASONING_CONSISTENCYLambdas reported in xgboost_reasoning match the final Poisson lambdas.ok
PREGAME_DIRECTION_AGREESThe pregame fair-line favorite agrees with the model winner.ok
PREGAME_TOTAL_CONSISTENTThe pregame expected total is close to the sum of the lambdas.ok
RESULT_WINNER_HITThe predicted winner matched the real result.ok
RESULT_EXACT_SCORE_HITThe headline score matched the real scoreline exactly.ok
RESULT_OU_HITThe Over/Under 2.5 lean matched reality.ok
RESULT_BTTS_HITThe BTTS lean matched reality.ok
RESULT_CONFIDENT_MISSA high-confidence prediction did not reach the real outcome.ok
OUTCOME_BRIERBrier score of the win/draw/loss probabilities vs the real result.ok
Generated 2026-09-18 13:35:29 UTC · Soccer prediction reporting