Soccer prediction system deep audit

Slate 20260919 · immutable forensic review of 9 match prediction artifact(s)

Executive conclusion

In 4 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

9matches audited
0critical
0errors
11warnings
0grid non-reproducible
4headline != modal
0winner disagreements
0lambda inconsistencies
0with real result
-winner hit-rate
-decisive hit-rate
11%predicted draws
-actual draws
0confident misses
-mean Brier
 

Matches

Tottenham vs Aston Villa

20260919_AVL@TOT · Premier League
WARNINGS risk 20 pred 0-1 (Aston Villa)λ 1.19 / 1.46conf 0.56 warning 2 info 1 good 37
Problems (2)
SevCategoryRuleMessageValueExpected
warningprovenanceFBREF_PRIORS_ACTIVEFBref dataset files not found; team priors inactive.
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-floor1.1931.464
Pre-calibration1.1931.464
Final1.1931.464
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 scorelinesP
0-110.2%
1-112.4%
1-29.0%
1-08.3%
0-27.5%
Outcome — home 30.7% · draw 26.0% · away 43.3%
Markets — O2.5 49.6% · BTTS 53.7%
Passed checks (37)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENTottenham: 11 starters.
goodstructureFORMATION_VALIDTottenham: formation 3-5-2 valid.
goodstructureLINEUP_ELEVENAston Villa: 11 starters.
goodstructureFORMATION_VALIDAston Villa: formation 3-4-2-1 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 0.9967.
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 0.0017).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0033).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Aston Villa' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Aston Villa).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Aston Villa).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.193, away 1.464).
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 (1.193 / 1.464).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMTottenham: player goals reconcile to team total (0).
goodintegrityTEAM_GOALS_SUMAston Villa: player goals reconcile to team total (1).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodintegrityPLAYER_STAT_REALISMAll 22 projected player lines pass realism checks.
goodstructureEXACTLY_ONE_GOALKEEPERTottenham: exactly one goalkeeper (Antonín Kinský).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPTottenham: 18 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERAston Villa: exactly one goalkeeper (Zion Suzuki).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPAston Villa: 15 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 (Aston Villa).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 2.66 ~ Poisson total 2.66.
Notes (1)
SevCategoryRuleMessageValueExpected
infomodel_behaviorMOST_LIKELY_IS_MODEDecisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.1236); the headline shows the most likely decisive score 0-1 (p=0.1016) for the probability favorite Aston Villa.0-11-1

Forest vs Coventry

20260919_COC@NFO · Premier League
WARNINGS risk 20 pred 1-0 (Forest)λ 1.44 / 1.02conf 0.58 warning 2 info 1 good 37
Problems (2)
SevCategoryRuleMessageValueExpected
warningprovenanceFBREF_PRIORS_ACTIVEFBref dataset files not found; team priors inactive.
warningintegrityPLAYER_STAT_REALISMOusmane Diomande (Forest): accurate passes 35 exceed total passes 1.35
Lambda evolution (audit trail)
StageHome λAway λ
Pre-competitive-floor1.4440.875
Pre-calibration1.4440.875
Final1.4441.020
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
1-012.2%
1-112.6%
2-19.0%
2-08.9%
0-08.6%
Outcome — home 46.7% · draw 26.7% · away 26.6%
Markets — O2.5 44.7% · BTTS 49.0%
Passed checks (37)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENForest: 11 starters.
goodstructureFORMATION_VALIDForest: formation 4-3-3 valid.
goodstructureLINEUP_ELEVENCoventry: 11 starters.
goodstructureFORMATION_VALIDCoventry: formation 3-4-2-1 valid.
goodprobabilityOUTCOME_SUMS_ONEoutcome sums to 0.9999.
goodprobabilityMARKETS_COMPLEMENTARYOver/Under and BTTS pairs each sum to 1.
goodprobabilitySCORELINE_PROBS_NONNEGAll scoreline probabilities within [0,1].
goodprobabilityGRID_SUMS_ONEscoreline grid sums to 0.9976.
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 0.0012).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0024).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Forest' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Forest).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Forest).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.444, 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 (1.444 / 1.02).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMForest: player goals reconcile to team total (1).
goodintegrityTEAM_GOALS_SUMCoventry: player goals reconcile to team total (0).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodstructureEXACTLY_ONE_GOALKEEPERForest: exactly one goalkeeper (John Victor).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPForest: 12 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERCoventry: exactly one goalkeeper (Carl Rushworth).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPCoventry: 14 projected shots inside the realistic envelope.
goodprovenanceSTANDING_AVAILABLELeague-standing context available for both sides.
goodprovenanceRANKING_PROVENANCE22/22 starters carry a ranking source.
goodprovenanceDATA_DEPTH15/22 starters have adequate recent history.
goodprovenancePREGAME_DIRECTION_AGREESPregame favorite agrees with the model winner (Forest).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 2.46 ~ Poisson total 2.46.
Notes (1)
SevCategoryRuleMessageValueExpected
infomodel_behaviorMOST_LIKELY_IS_MODEDecisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.1264); the headline shows the most likely decisive score 1-0 (p=0.1218) for the probability favorite Forest.1-01-1

Newcastle vs Hull

20260919_HUL@NEW · Premier League
WARNINGS risk 20 pred 2-1 (Newcastle)λ 2.59 / 1.00conf 0.65 warning 2 good 38
Problems (2)
SevCategoryRuleMessageValueExpected
warningintegrityPLAYER_STAT_REALISMBazoumana Touré (Newcastle): accurate passes 39 exceed total passes 1.39
warningprovenanceDATA_DEPTH11/22 starters have <3 recent games; their stats lean on market-value / prior synthesis rather than form.11
Lambda evolution (audit trail)
StageHome λAway λ
Pre-mismatch2.3811.026
Pre-calibration2.3811.026
Final2.5941.002
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
2-19.2%
2-09.2%
3-18.0%
3-08.0%
1-17.2%
Outcome — home 71.5% · draw 16.2% · away 12.3%
Markets — O2.5 69.6% · BTTS 58.6%
Passed checks (38)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENNewcastle: 11 starters.
goodstructureFORMATION_VALIDNewcastle: formation 4-3-3 valid.
goodstructureLINEUP_ELEVENHull: 11 starters.
goodstructureFORMATION_VALIDHull: formation 3-5-2 valid.
goodprobabilityOUTCOME_SUMS_ONEoutcome sums to 1.0001.
goodprobabilityMARKETS_COMPLEMENTARYOver/Under and BTTS pairs each sum to 1.
goodprobabilitySCORELINE_PROBS_NONNEGAll scoreline probabilities within [0,1].
goodprobabilityGRID_SUMS_ONEscoreline grid sums to 0.9965.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0004).
goodmodel_behaviorMOST_LIKELY_IS_MODEHeadline score 2-1 is the modal scoreline.
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0023).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0035).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Newcastle' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Newcastle).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Newcastle).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 2.594, away 1.002).
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 (2.594 / 1.002).
goodprovenanceFBREF_PRIORS_ACTIVEFBref team priors applied (blend weight 0.192, home E[lambda]=2.222, away E[lambda]=2.222).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMNewcastle: player goals reconcile to team total (2).
goodintegrityTEAM_GOALS_SUMHull: player goals reconcile to team total (1).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodstructureEXACTLY_ONE_GOALKEEPERNewcastle: exactly one goalkeeper (Lukáš Horníček).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPNewcastle: 10 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERHull: exactly one goalkeeper (Konstantinos Tzolakis).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPHull: 9 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 (Newcastle).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 3.60 ~ Poisson total 3.60.

Celta vs R. Racing Club

20260919_SAN@CEL · LaLiga
WARNINGS risk 20 pred 1-1 ()λ 1.02 / 1.02conf 0.56 warning 2 info 1 good 35
Problems (2)
SevCategoryRuleMessageValueExpected
warningprovenanceFBREF_PRIORS_ACTIVEFBref dataset files not found; team priors inactive.
warningintegrityPLAYER_STAT_REALISMFacundo González (R. Racing Club): accurate passes 31 exceed total passes 16.31
Lambda evolution (audit trail)
StageHome λAway λ
Pre-competitive-floor0.8911.018
Pre-calibration0.8911.018
Final1.0201.018
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
1-113.7%
0-013.2%
1-013.2%
0-113.1%
2-16.9%
Outcome — home 34.7% · draw 30.8% · away 34.6%
Markets — O2.5 33.4% · BTTS 41.0%
Passed checks (35)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENCelta: 11 starters.
goodstructureFORMATION_VALIDCelta: formation 3-5-2 valid.
goodstructureLINEUP_ELEVENR. Racing Club: 11 starters.
goodstructureFORMATION_VALIDR. Racing Club: formation 3-5-2 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 0.9976.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0014).
goodmodel_behaviorMOST_LIKELY_IS_MODEHeadline score 1-1 is the modal scoreline.
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0011).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0023).
goodintegrityWINNER_IS_VALID_TEAMDraw predicted (no single winner named), consistent with the headline score / match_resolution.
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.02, away 1.018).
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 (1.02 / 1.018).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMCelta: player goals reconcile to team total (1).
goodintegrityTEAM_GOALS_SUMR. Racing Club: player goals reconcile to team total (1).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodstructureEXACTLY_ONE_GOALKEEPERCelta: exactly one goalkeeper (Ionuț Radu).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPCelta: 11 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERR. Racing Club: exactly one goalkeeper (Julen Agirrezabala).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPR. Racing Club: 11 projected shots inside the realistic envelope.
goodprovenanceSTANDING_AVAILABLELeague-standing context available for both sides.
goodprovenanceRANKING_PROVENANCE22/22 starters carry a ranking source.
goodprovenanceDATA_DEPTH17/22 starters have adequate recent history.
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 2.04 ~ Poisson total 2.04.
Notes (1)
SevCategoryRuleMessageValueExpected
infomodel_behaviorMODAL_OUTCOME_IS_DRAWHeadline is a draw, but the largest outcome bucket is home_win (0.3467); aggregate probability leans to Celta.0.3467

Athletic Club vs Alavés

20260919_ALA@ATH · LaLiga
WARNINGS risk 10 pred 1-0 (Athletic Club)λ 1.29 / 1.02conf 0.57 warning 1 info 1 good 38
Problems (1)
SevCategoryRuleMessageValueExpected
warningintegrityPLAYER_STAT_REALISMMiguel Rodríguez (Alavés): accurate passes 35 exceed total passes 1.35
Lambda evolution (audit trail)
StageHome λAway λ
Pre-competitive-floor1.2870.892
Pre-calibration1.2870.892
Final1.2871.020
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
1-012.7%
1-113.2%
0-010.1%
0-110.0%
2-18.4%
Outcome — home 42.4% · draw 28.2% · away 29.3%
Markets — O2.5 40.6% · BTTS 46.4%
Passed checks (38)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENAthletic Club: 11 starters.
goodstructureFORMATION_VALIDAthletic Club: formation 4-2-3-1 valid.
goodstructureLINEUP_ELEVENAlavés: 11 starters.
goodstructureFORMATION_VALIDAlavés: formation 3-5-2 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 0.9981.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0012).
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0010).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0020).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Athletic Club' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Athletic Club).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Athletic Club).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.287, 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 (1.287 / 1.02).
goodprovenanceFBREF_PRIORS_ACTIVEFBref team priors applied (blend weight 0.228, home E[lambda]=1.237, away E[lambda]=1.235).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMAthletic Club: player goals reconcile to team total (1).
goodintegrityTEAM_GOALS_SUMAlavés: player goals reconcile to team total (0).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodstructureEXACTLY_ONE_GOALKEEPERAthletic Club: exactly one goalkeeper (Unai Simón).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPAthletic Club: 12 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERAlavés: exactly one goalkeeper (Antonio Sivera).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPAlavés: 8 projected shots inside the realistic envelope.
goodprovenanceSTANDING_AVAILABLELeague-standing context available for both sides.
goodprovenanceRANKING_PROVENANCE22/22 starters carry a ranking source.
goodprovenanceDATA_DEPTH17/22 starters have adequate recent history.
goodprovenancePREGAME_DIRECTION_AGREESPregame favorite agrees with the model winner (Athletic Club).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 2.31 ~ Poisson total 2.31.
Notes (1)
SevCategoryRuleMessageValueExpected
infomodel_behaviorMOST_LIKELY_IS_MODEDecisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.1319); the headline shows the most likely decisive score 1-0 (p=0.1270) for the probability favorite Athletic Club.1-01-1

Brighton vs Arsenal

20260919_ARS@BHA · Premier League
WARNINGS risk 10 pred 1-2 (Arsenal)λ 1.30 / 2.12conf 0.6 warning 1 good 39
Problems (1)
SevCategoryRuleMessageValueExpected
warningprovenanceDATA_DEPTH10/22 starters have <3 recent games; their stats lean on market-value / prior synthesis rather than form.10
Lambda evolution (audit trail)
StageHome λAway λ
Pre-mismatch1.3311.939
Pre-competitive-floor1.2972.122
Pre-calibration1.3311.939
Final1.2972.122
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
1-29.6%
1-19.1%
0-27.4%
0-16.9%
1-36.8%
Outcome — home 22.9% · draw 20.9% · away 56.3%
Markets — O2.5 66.4% · BTTS 64.0%
Passed checks (39)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENBrighton: 11 starters.
goodstructureFORMATION_VALIDBrighton: formation 4-2-3-1 valid.
goodstructureLINEUP_ELEVENArsenal: 11 starters.
goodstructureFORMATION_VALIDArsenal: formation 3-5-2 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 0.9948.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0006).
goodmodel_behaviorMOST_LIKELY_IS_MODEHeadline score 1-2 is the modal scoreline.
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0026).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0051).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Arsenal' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Arsenal).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Arsenal).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.297, away 2.122).
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 (1.297 / 2.122).
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_SUMBrighton: player goals reconcile to team total (1).
goodintegrityTEAM_GOALS_SUMArsenal: 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_GOALKEEPERBrighton: exactly one goalkeeper (Bart Verbruggen).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPBrighton: 17 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERArsenal: exactly one goalkeeper (David Raya).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPArsenal: 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 (Arsenal).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 3.42 ~ Poisson total 3.42.

Osasuna vs Rayo Vallecano

20260919_RAY@OSA · LaLiga
WARNINGS risk 10 pred 0-1 (Rayo Vallecano)λ 1.12 / 1.21conf 0.56 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-floor1.1251.212
Pre-calibration1.1251.212
Final1.1251.212
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 scorelinesP
0-111.6%
1-113.3%
1-010.7%
0-09.8%
1-28.0%
Outcome — home 33.7% · draw 28.3% · away 38.0%
Markets — O2.5 41.4% · BTTS 47.6%
Passed checks (38)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENOsasuna: 11 starters.
goodstructureFORMATION_VALIDOsasuna: formation 4-2-3-1 valid.
goodstructureLINEUP_ELEVENRayo Vallecano: 11 starters.
goodstructureFORMATION_VALIDRayo Vallecano: formation 4-3-3 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 0.9961.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0013).
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0021).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0038).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Rayo Vallecano' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Rayo Vallecano).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Rayo Vallecano).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.125, away 1.212).
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 (1.125 / 1.212).
goodprovenanceFBREF_PRIORS_ACTIVEFBref team priors applied (blend weight 0.228, home E[lambda]=0.985, away E[lambda]=0.988).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMOsasuna: player goals reconcile to team total (0).
goodintegrityTEAM_GOALS_SUMRayo Vallecano: player goals reconcile to team total (1).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodintegrityPLAYER_STAT_REALISMAll 22 projected player lines pass realism checks.
goodstructureEXACTLY_ONE_GOALKEEPEROsasuna: exactly one goalkeeper (Sergio Herrera).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPOsasuna: 14 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERRayo Vallecano: exactly one goalkeeper (Augusto Batalla).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPRayo Vallecano: 14 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 (Rayo Vallecano).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 2.34 ~ Poisson total 2.34.
Notes (1)
SevCategoryRuleMessageValueExpected
infomodel_behaviorMOST_LIKELY_IS_MODEDecisive-score substitution policy: the modal scoreline is a draw 1-1 (p=0.1330); the headline shows the most likely decisive score 0-1 (p=0.1158) for the probability favorite Rayo Vallecano.0-11-1

Sevilla vs Barcelona

20260919_BAR@SEV · LaLiga
CLEAN risk 0 pred 1-3 (Barcelona)λ 1.27 / 3.69conf 0.69 good 40
Problems (0)
none
Lambda evolution (audit trail)
StageHome λAway λ
Pre-mismatch1.3543.135
Pre-calibration1.3543.135
Final1.2753.689
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
1-37.5%
1-46.9%
1-26.1%
0-35.9%
0-45.4%
Outcome — home 8.6% · draw 10.7% · away 80.7%
Markets — O2.5 87.2% · BTTS 70.3%
Passed checks (40)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENSevilla: 11 starters.
goodstructureFORMATION_VALIDSevilla: formation 5-4-1 valid.
goodstructureLINEUP_ELEVENBarcelona: 11 starters.
goodstructureFORMATION_VALIDBarcelona: formation 3-5-2 valid.
goodprobabilityOUTCOME_SUMS_ONEoutcome sums to 0.9999.
goodprobabilityMARKETS_COMPLEMENTARYOver/Under and BTTS pairs each sum to 1.
goodprobabilitySCORELINE_PROBS_NONNEGAll scoreline probabilities within [0,1].
goodprobabilityGRID_SUMS_ONEscoreline grid sums to 0.9970.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0004).
goodmodel_behaviorMOST_LIKELY_IS_MODEHeadline score 1-3 is the modal scoreline.
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0020).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0031).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Barcelona' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Barcelona).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Barcelona).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 1.275, away 3.689).
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 (1.275 / 3.689).
goodprovenanceFBREF_PRIORS_ACTIVEFBref team priors applied (blend weight 0.228, home E[lambda]=3.5, away E[lambda]=3.5).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMSevilla: player goals reconcile to team total (1).
goodintegrityTEAM_GOALS_SUMBarcelona: player goals reconcile to team total (3).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodintegrityPLAYER_STAT_REALISMAll 22 projected player lines pass realism checks.
goodstructureEXACTLY_ONE_GOALKEEPERSevilla: exactly one goalkeeper (Odysseas Vlachodimos).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPSevilla: 14 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERBarcelona: exactly one goalkeeper (Joan García).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPBarcelona: 15 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 (Barcelona).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 4.96 ~ Poisson total 4.96.

Everton vs Ipswich

20260919_IPS@EVE · Premier League
CLEAN risk 0 pred 2-1 (Everton)λ 2.65 / 1.14conf 0.65 good 40
Problems (0)
none
Lambda evolution (audit trail)
StageHome λAway λ
Pre-calibration2.6551.144
Final2.6551.144
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 scorelinesP
2-19.0%
3-18.0%
2-07.9%
3-07.0%
1-16.8%
Outcome — home 69.5% · draw 16.4% · away 14.1%
Markets — O2.5 73.1% · BTTS 63.4%
Passed checks (40)
SevCategoryRuleMessageValueExpected
goodintegritySCHEMA_KEYS_PRESENTAll required sections present.
goodstructureLINEUP_ELEVENEverton: 11 starters.
goodstructureFORMATION_VALIDEverton: formation 3-4-2-1 valid.
goodstructureLINEUP_ELEVENIpswich: 11 starters.
goodstructureFORMATION_VALIDIpswich: formation 3-5-2 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 0.9958.
goodmodel_behaviorGRID_REPRODUCIBLEGrid reproducible from lambdas (max cell diff 0.0004).
goodmodel_behaviorMOST_LIKELY_IS_MODEHeadline score 2-1 is the modal scoreline.
goodprobabilityMLS_PROBABILITY_MATCHESHeadline probability matches the grid.
goodmodel_behaviorOUTCOME_MATCHES_GRIDOutcome split matches the grid (max diff 0.0022).
goodmodel_behaviorMARKETS_MATCH_GRIDMarkets match the grid (max diff 0.0043).
goodintegrityWINNER_IS_VALID_TEAMpredicted_winner 'Everton' is a participating side.
goodmodel_behaviorWINNER_MATCHES_SCOREWinner agrees with headline score (Everton).
goodmodel_behaviorWINNER_PROBABILITY_DIRECTIONWinner matches the win-probability favorite (Everton).
goodmodel_behaviorLAMBDA_BOUNDSLambdas in range (home 2.655, away 1.144).
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 (2.655 / 1.144).
goodprovenanceFBREF_PRIORS_ACTIVEFBref team priors applied (blend weight 0.192, home E[lambda]=2.222, away E[lambda]=2.222).
goodprovenanceXGBOOST_PLAYER_REASONINGAll 22 starters carry full XGBoost / FBref reasoning.
goodintegrityTEAM_GOALS_SUMEverton: player goals reconcile to team total (2).
goodintegrityTEAM_GOALS_SUMIpswich: player goals reconcile to team total (1).
goodstructurePOSSESSION_SUMS_100Possession sums to 100.0%.
goodintegrityPLAYER_STAT_REALISMAll 22 projected player lines pass realism checks.
goodstructureEXACTLY_ONE_GOALKEEPEREverton: exactly one goalkeeper (Jordan Pickford).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPEverton: 16 projected shots inside the realistic envelope.
goodstructureEXACTLY_ONE_GOALKEEPERIpswich: exactly one goalkeeper (Kjell Scherpen).
goodintegritySHOT_MATH_INTEGRITYAll 11 projected lines have consistent shot components.
goodintegrityTEAM_SHOTS_CAPIpswich: 13 projected shots inside the realistic envelope.
goodprovenanceSTANDING_AVAILABLELeague-standing context available for both sides.
goodprovenanceRANKING_PROVENANCE22/22 starters carry a ranking source.
goodprovenanceDATA_DEPTH18/22 starters have adequate recent history.
goodprovenancePREGAME_DIRECTION_AGREESPregame favorite agrees with the model winner (Everton).
goodprovenancePREGAME_TOTAL_CONSISTENTPregame total 3.80 ~ Poisson total 3.80.

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).4 problems
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.4 problems
FBREF_PRIORS_ACTIVEFBref season-prior files exist so that branch is not silently inert.3 problems
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 2026-09-19 13:33:58 UTC · Soccer prediction reporting