Deep slate report — 20260829
Generated 2026-08-30T02:02:21.407927-04:00 · schema deep_slate_diagnosis_v3 · fuses prediction review + forensic deep audit + official starter lines, with a Luna post-mortem.
Process, end to end
1. Settle review
complete
17 games vs official finals
2. Forensic deep audit
complete
per-game probability + pitcher impact
3. Deep diagnosis
built
1 losses attributed to a root cause
4. Luna post-mortem
error
no validated Luna coverage receipt
5. Adjustment
quarantined_invalid_artifact
slate quarantined from new parameter proposals until critical/error artifact findings are corrected
6. Gating
shadow, manual + gated
production champion unchanged
Slate summary
Mean total-run bias (losses)
3
What drove the losses
Backed the wrong side with real conviction: 1
Evidence coverage
This is the source receipt: a source shown as missing was not silently assumed.
| Source | Status | Evidence supplied | Path | Bytes |
|---|
| prediction review | loaded | official outcomes and scoring errors | /app/output/reports/prediction_review_20260829.json | 424215 |
| forensic deep audit | loaded | probability, provenance, starter and decision contracts | /app/output/reports/PREDICTION_SYSTEM_DEEP_AUDIT_20260829.json | 270564 |
| continuous learning | loaded | pipeline-stage quality, calibration, probability lineage, score distribution, and official starter comparisons | /app/output/learning/continuous_learning_daily_20260829.json | 2632202 |
| detailed report | loaded | full game and simulation rendering | /app/output/reports/detailed_20260829.html | 401110 |
| ultra detailed player audit | loaded | player-level prediction trace | /app/output/reports/20260829_ultra_detailed_player_audit.html | 6281152 |
| structural audit | loaded | all invariant findings | /app/output/reports/audit_report_20260829.html | 380826 |
| integrity revision audit | missing | post-Luna remediation proof for immutable revision artifacts | /app/output/reports/audit_report_20260829_integrity_revision.html | n/a |
High-level causal issue register
Deterministic findings across data integrity, model validation, calibration, score distribution, pipeline stages, simulation, and coverage. Critical is reserved for broken evidence or an unexplained contract violation; high marks material model/process risk.
| Severity | Domain | Finding and evidence | Affected | Action and guardrail |
|---|
| HIGH | artifact_integrity | Blocking canonical artifact error
structural audit totals: {'critical': 0, 'error': 2, 'warning': 1, 'info': 125} | 20260829_ARI@SF_2 | quarantine the slate from parameter learning and repair the generation boundary
Guardrail: do not attribute prediction accuracy changes to a malformed artifact |
| HIGH | calibration | Published confidence buckets were poorly aligned on this slate
weighted absolute calibration error 0.3961 across 17 games | ARI@SF, ARI@SF, BAL@OAK, BOS@NYY, BOS@NYY, CHW@MIN, CIN@CHC, COL@ATL, HOU@NYM, KC@CLE, LAD@DET, MIA@WAS, PHI@LAA, PIT@STL, SD@TB, SEA@TOR, TEX@MIL | monitor the same buckets over the registered rolling window before changing calibration weights
Guardrail: seven games cannot establish a stable calibration curve |
| HIGH | model_validation | Validated run models explain very little holdout variance
home/away/total R² = 0.0452/0.0220/0.0224 | ARI@SF, ARI@SF, BAL@OAK, BOS@NYY, BOS@NYY, CHW@MIN, CIN@CHC, COL@ATL, HOU@NYM, KC@CLE, LAD@DET, MIA@WAS, PHI@LAA, PIT@STL, SD@TB, SEA@TOR, TEX@MIL | treat exact scores as low-confidence estimates and investigate feature/target signal before tuning tails
Guardrail: positive R² means better than the mean baseline, not that exact scores are reliable |
| INFO | coverage | Every settled game received structural QA
17/17 games structurally audited | ARI@SF, ARI@SF, BAL@OAK, BOS@NYY, BOS@NYY, CHW@MIN, CIN@CHC, COL@ATL, HOU@NYM, KC@CLE, LAD@DET, MIA@WAS, PHI@LAA, PIT@STL, SD@TB, SEA@TOR, TEX@MIL | retain the per-game receipt in every report
Guardrail: complete coverage does not erase findings inside the audited games |
| INFO | probability_lineage | Material calibrated-to-published movements have recorded lineage
9 material movement(s), all explained by named score-coherence and/or recency transformations | BOS@NYY, BOS@NYY, CIN@CHC, COL@ATL, HOU@NYM, MIA@WAS, PIT@STL, SEA@TOR, TEX@MIL | keep displaying each probability layer and transformation
Guardrail: a higher published value is not automatically a governor bypass |
| INFO | simulation | Simulation produced no winner-direction intervention on this slate
flip candidates 0; published flips 0 | ARI@SF, ARI@SF, BAL@OAK, BOS@NYY, BOS@NYY, CHW@MIN, CIN@CHC, COL@ATL, HOU@NYM, KC@CLE, LAD@DET, MIA@WAS, PHI@LAA, PIT@STL, SD@TB, SEA@TOR, TEX@MIL | measure simulation contribution over rolling slates and distinguish score realization from winner flipping
Guardrail: zero flips is not itself a defect and does not mean simulation did no score-reconciliation work |
Pipeline stage regression
Every stored stage is scored against the same official finals. A later recovery remains visible; it does not erase an unstable intermediate flip.
| Stage | Winner hits | Accuracy | Team-run MAE | Total-run MAE | Margin MAE |
|---|
| historical_ml | 15/17 | 88.2% | 2.112 | 3.029 | 3.008 |
| variance_applied | 13/17 | 76.5% | 2.176 | 3.176 | 2.941 |
| enhanced_prediction | 16/17 | 94.1% | 2.176 | 3.529 | 2.941 |
| embedded_savant_run_environment | 15/17 | 88.2% | 2.265 | 3.706 | 3.118 |
| fundamental_edge_target_review | 16/17 | 94.1% | 2.235 | 3.765 | 3.059 |
| capped_target | 16/17 | 94.1% | 2.235 | 3.765 | 3.059 |
| simulation_result | 16/17 | 94.1% | 2.235 | 3.765 | 3.059 |
| published_final | 16/17 | 94.1% | 2.235 | 3.765 | 3.059 |
| Transition | Δ accuracy | Δ hits | Δ team-run MAE | Δ margin MAE | Games whose winner/tie changed |
|---|
| historical_ml → variance_applied | -0.1177 | -2 | 0.0646 | -0.067 | BOS@NYY: NYY → TIE; SD@TB: TB → TIE |
| variance_applied → enhanced_prediction | 0.1765 | 3 | 0 | 0 | BOS@NYY: NYY → BOS; BOS@NYY: TIE → NYY; SD@TB: TIE → TB |
| enhanced_prediction → embedded_savant_run_environment | -0.0588 | -1 | 0.0882 | 0.1764 | ARI@SF: ARI → TIE |
| embedded_savant_run_environment → fundamental_edge_target_review | 0.0588 | 1 | -0.0294 | -0.0588 | ARI@SF: TIE → ARI |
| fundamental_edge_target_review → capped_target | 0 | 0 | 0 | 0 | |
| capped_target → simulation_result | 0 | 0 | 0 | 0 | |
| simulation_result → published_final | 0 | 0 | 0 | 0 | |
Probability lineage and calibration
All columns are pick-side probabilities, so away favorites are not accidentally compared using home-side semantics.
| Game | Outcome | Raw model | Calibrated | Governor | Score coherence | Published | Explanation |
|---|
| ARI@SF | hit | 0.5085 | 0.5082 | 0.5082 | 0.5466 | 0.5269 | score coherence followed by matchup-recency guardrail |
| ARI@SF | hit | 0.5085 | 0.5082 | 0.5082 | 0.5466 | 0.5185 | score coherence followed by matchup-recency guardrail |
| BAL@OAK | hit | 0.568 | 0.5661 | 0.5661 | 0.5756 | 0.5546 | score coherence followed by matchup-recency guardrail |
| BOS@NYY | hit | 0.5085 | 0.5082 | 0.5082 | 0.5466 | 0.53 | score coherence followed by matchup-recency guardrail |
| BOS@NYY | hit | 0.5085 | 0.5082 | 0.5082 | 0.5466 | 0.5337 | score coherence followed by matchup-recency guardrail |
| CHW@MIN | hit | 0.585 | 0.5814 | 0.5814 | 0.5832 | 0.5707 | score coherence followed by matchup-recency guardrail |
| CIN@CHC | hit | 0.6815 | 0.6705 | 0.6705 | 0.6702 | 0.56 | score coherence followed by matchup-recency guardrail |
| COL@ATL | hit | 0.8613 | 0.6551 | 0.6551 | 0.7345 | 0.7345 | score_coherence aligned probability to the published score margin |
| HOU@NYM | hit | 0.5085 | 0.5082 | 0.5082 | 0.5466 | 0.53 | score coherence followed by matchup-recency guardrail |
| KC@CLE | hit | 0.56 | 0.5584 | 0.5584 | 0.5717 | 0.5518 | score coherence followed by matchup-recency guardrail |
| LAD@DET | miss | 0.67 | 0.6608 | 0.6608 | 0.6654 | 0.6654 | score_coherence aligned probability to the published score margin |
| MIA@WAS | hit | 0.585 | 0.5814 | 0.5814 | 0.5832 | 0.5601 | score coherence followed by matchup-recency guardrail |
| PHI@LAA | hit | 0.67 | 0.6608 | 0.6608 | 0.6654 | 0.6654 | score_coherence aligned probability to the published score margin |
| PIT@STL | hit | 0.5085 | 0.5082 | 0.5082 | 0.5466 | 0.53 | score coherence followed by matchup-recency guardrail |
| SD@TB | hit | 0.6199 | 0.6157 | 0.6157 | 0.6428 | 0.6214 | no recorded transformation explains the publication delta |
| SEA@TOR | hit | 0.624 | 0.6193 | 0.6193 | 0.6022 | 0.54 | score coherence followed by matchup-recency guardrail |
| TEX@MIL | hit | 0.8613 | 0.6551 | 0.6551 | 0.7345 | 0.7345 | score_coherence aligned probability to the published score margin |
Score distribution and margin compression
Predicted margins ≤2: 14 · Actual margins ≥4: 8
| Game | Pred margin | Actual margin | Pred − actual |
|---|
| ARI@SF | 1 | 6 | -5 |
| ARI@SF | 1 | 5 | -4 |
| BAL@OAK | 1 | 2 | -1 |
| BOS@NYY | 1 | 6 | -5 |
| BOS@NYY | 1 | 7 | -6 |
| CHW@MIN | 2 | 1 | 1 |
| CIN@CHC | 2 | 12 | -10 |
| COL@ATL | 5 | 1 | 4 |
| HOU@NYM | 1 | 4 | -3 |
| KC@CLE | 1 | 5 | -4 |
| LAD@DET | 2 | 1 | 1 |
| MIA@WAS | 1 | 1 | 0 |
| PHI@LAA | 3 | 2 | 1 |
| PIT@STL | 1 | 4 | -3 |
| SD@TB | 1 | 1 | 0 |
| SEA@TOR | 1 | 1 | 0 |
| TEX@MIL | 4 | 2 | 2 |
Per-game evidence coverage
Every game, in depth (17)
ARI@SFHIT
Predicted: ARI 7-6 (conf -0.05, lean)Actual: ARI 7-1Brier 0.224total bias +5
Root cause: Right winner, wrong score
Evidence-backed diagnosis- published system edge 0.05 (lean); this is not the half-scale distance from 50%
- team-run absolute error 5; total bias +5
- pregame home-minus-away evidence: season -0.547, recent -0.234, pitcher -0.027, composite -0.419
- confidence pipeline reached 0.508 before later publication/recency controls; published home probability was 0.473
- score-volatility actions: no_change
- structural audit exposed 8 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs over-projected by 5 (13 vs 8)
- SF offense over-projected by 5
- Matt Wilkinson (SF) allowed 5 fewer ER than projected (1 vs 6)
| Team | Starter | Projected | Actual | ER err |
|---|
| ARI | Merrill Kelly | 4 ER / 15 outs | 3 ER / 13 outs | -1 |
| SF | Matt Wilkinson | 6 ER / 9 outs | 1 ER / 14 outs | -5 |
Decision internals: run factors home×0.975 / away×1.002; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | -0.547 | positive favors home; negative favors away |
| Recent form | -0.234 | positive favors home; negative favors away |
| Starting-pitcher matchup | -0.027 | positive favors home; negative favors away |
| Composite | -0.419 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.78-4.4 | n/a | Base ML historical model prediction |
| variance_applied | 5-4 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 6-5 | 0.527 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 6-6 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 7-6 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 7-6 | n/a | Capped at max 12 realistic runs |
| simulation_result | 7-6 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.78-4.4 | 5-4 | 0.615 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-4 | 6-5 | -0 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 6-5 | 6-6 | -1 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 6-6 | 7-6 | 1 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 7-6 | 7-6 | -0 | Capped at max 12 realistic runs |
| simulation_result | 7-6 | 7-6 | -0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| ARI | Geraldo Perdomo | batter / SS | boosted | predicted_RBI 2.59 |
| SF | Rafael Devers | batter / 1B | better | predicted_RBI 2.34 |
| SF | Drew Gilbert | batter / CF | worse | predicted_H 1.73 |
| SF | Turner Hill | batter / LF | worse | predicted_SO 1.5 |
| SF | Jung Hoo Lee | batter / RF | worse | predicted_AB -1.49 |
| ARI | Ildemaro Vargas | batter / 2B | worse | predicted_AB 1.482 |
| SF | Drew Cavanaugh | batter / C | worse | predicted_AB 1.378 |
| SF | Bryce Eldridge | batter / DH | better | predicted_RBI 1.23 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| warning | PITCH_COUNT_PITCHER_MATCH | ARI pitch_count row for Merrill Kelly differs from pitching box score: batters_faced/BF: summary=20, box=18. | ARI Merrill Kelly |
| info | CURRENT_PITCHER_DUPLICATION | ARI pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | ARI |
| info | CURRENT_PITCHER_DUPLICATION | SF pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | SF |
| info | CURRENT_FINALIZED_MATCH_BOX | ARI model-prediction batting totals differ from final simulation box score (AB: current=33, sim=36; H: current=8, sim=9; 2B: current=2, sim=3; RBI: current=5, sim=7; R: current=4, sim=7; BB: current=3, sim=2). | ARI |
| info | CURRENT_PLAYER_MATCH_BOX | ARI current finalized hitter lines are model predictions and differ from the simulation box score: Ildemaro Vargas: AB current=4 box=5, RBI current=1 box=2, R current=0 box=1, SO current=0 box=1; Nolan Arenado: SO current=1 box=0; Tim Tawa: H current=1 box=0, RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Gabriel Moreno: AB current=5 box=4, HR current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Jose Fernandez: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, R current=0 box=1; Ryan Waldschmidt: AB current=3 box=4, H current=0 box=1, 2B current=0 box=1, RBI current=0 box=1; Geraldo Perdomo: H current=1 box=3, HR current=0 box=1, RBI current=0 box=3, R current=0 box=1; Corbin Carroll: AB current=4 box=5, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0 | ARI |
| info | CURRENT_FINALIZED_MATCH_BOX | SF model-prediction batting totals differ from final simulation box score (2B: current=2, sim=1; HR: current=1, sim=3; RBI: current=4, sim=6; R: current=3, sim=6; BB: current=3, sim=1; SO: current=8, sim=7; HBP: current=0, sim=1). | SF |
| info | CURRENT_PLAYER_MATCH_BOX | SF current finalized hitter lines are model predictions and differ from the simulation box score: Rafael Devers: H current=1 box=2, HR current=1 box=2, RBI current=1 box=3; Turner Hill: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=0 box=1; Drew Gilbert: AB current=3 box=4, H current=1 box=2, RBI current=0 box=1, R current=0 box=1; Bryce Eldridge: AB current=5 box=4, H current=1 box=2, 2B current=1 box=0, HR current=0 box=1; Jung Hoo Lee: AB current=4 box=3, H current=1 box=0, HBP current=0 box=1; Christian Koss: 2B current=0 box=1, R current=0 box=1; Drew Cavanaugh: AB current=3 box=4, H current=1 box=0, BB current=1 box=0; Shay Whitcomb: RBI current=1 box=0, SO current=1 box=0 | SF |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
ARI@SFHIT
Predicted: SF 5-6 (conf +0.04, toss_up)Actual: SF 2-7Brier 0.232total bias +2
Root cause: Toss-up that went our way
Evidence-backed diagnosis- published system edge 0.04 (toss_up); this is not the half-scale distance from 50%
- team-run absolute error 4; total bias +2
- pregame home-minus-away evidence: season -0.547, recent -0.234, pitcher -0.012, composite -0.206
- confidence pipeline reached 0.508 before later publication/recency controls; published home probability was 0.519
- score-volatility actions: no_change
- structural audit exposed 9 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- ARI offense over-projected by 3
Decision internals: run factors home×0.988 / away×1; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | -0.547 | positive favors home; negative favors away |
| Recent form | -0.234 | positive favors home; negative favors away |
| Starting-pitcher matchup | -0.012 | positive favors home; negative favors away |
| Composite | -0.206 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.08-6.19 | n/a | Base ML historical model prediction |
| variance_applied | 4-6 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-6 | 0.519 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_has_no_reliable_baseline, projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 5-6 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 5-6 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 5-6 | n/a | Capped at max 12 realistic runs |
| simulation_result | 5-6 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.08-6.19 | 4-6 | -0.104 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-6 | 5-6 | -1 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_has_no_reliable_baseline, projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 5-6 | 5-6 | 0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 5-6 | 5-6 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 5-6 | 5-6 | 0 | Capped at max 12 realistic runs |
| simulation_result | 5-6 | 5-6 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| ARI | Gerardo Carrillo | pitcher / Reliever | worse | predicted_ER 3 |
| ARI | Tim Tawa | batter / 1B | better | predicted_RBI 2.36 |
| SF | Drew Cavanaugh | batter / C | worse | predicted_SO 2.216 |
| SF | Turner Hill | batter / LF | better | predicted_AB 2.2 |
| ARI | Gabriel Moreno | batter / C | worse | predicted_AB -2 |
| SF | Yunior Marte | pitcher / Starter | neutral | predicted_SO -2 |
| SF | Sam Hentges | pitcher / Setup Man | neutral | predicted_SO 2 |
| SF | Rafael Devers | batter / 1B | better | predicted_2B 1.769 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| error | PITCHING_REALISM | current:SF has implausible starter workload/result line(s): Yunior Marte: 0.2 IP starter outing | SF |
| error | PITCHING_REALISM | simulation:SF has implausible starter workload/result line(s): Yunior Marte: 0.2 IP starter outing | SF |
| info | CURRENT_PITCHER_DUPLICATION | ARI pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | ARI |
| info | CURRENT_PITCHER_DUPLICATION | SF pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | SF |
| info | CURRENT_FINALIZED_MATCH_BOX | ARI model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=8, sim=5; 2B: current=2, sim=3; HR: current=1, sim=0; R: current=4, sim=5; BB: current=3, sim=5; SO: current=7, sim=10; HBP: current=0, sim=1). | ARI |
| info | CURRENT_PLAYER_MATCH_BOX | ARI current finalized hitter lines are model predictions and differ from the simulation box score: Ildemaro Vargas: H current=1 box=0, RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Nolan Arenado: AB current=4 box=5, H current=1 box=2, 2B current=1 box=0, RBI current=1 box=0; Geraldo Perdomo: H current=1 box=0, SO current=0 box=1; Jose Fernandez: AB current=3 box=4, H current=1 box=0, SO current=1 box=2; Corbin Carroll: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Ryan Waldschmidt: H current=0 box=1, 2B current=0 box=1, R current=0 box=1, BB current=0 box=1; Tim Tawa: AB current=3 box=4, 2B current=0 box=1, RBI current=1 box=3, R current=0 box=1; Gabriel Moreno: AB current=5 box=3, 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=2 | ARI |
| info | CURRENT_FINALIZED_MATCH_BOX | SF model-prediction batting totals differ from final simulation box score (AB: current=31, sim=39; H: current=7, sim=9; 2B: current=2, sim=5; HR: current=1, sim=2; RBI: current=4, sim=6; R: current=3, sim=6; BB: current=3, sim=2; SO: current=9, sim=10). | SF |
| info | CURRENT_PLAYER_MATCH_BOX | SF current finalized hitter lines are model predictions and differ from the simulation box score: Rafael Devers: AB current=4 box=5, H current=1 box=2, 2B current=0 box=2, HR current=1 box=0; Turner Hill: AB current=3 box=5, H current=1 box=2, 2B current=1 box=0, HR current=0 box=1; Drew Gilbert: AB current=3 box=4, H current=1 box=0, SO current=1 box=0, HBP current=0 box=1; Jung Hoo Lee: AB current=4 box=5, 2B current=0 box=1, SO current=0 box=1; Bryce Eldridge: AB current=5 box=4, 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=2; Christian Koss: AB current=3 box=4, 2B current=0 box=1; Drew Cavanaugh: AB current=3 box=4, R current=0 box=1, SO current=1 box=3; Shay Whitcomb: AB current=3 box=4, RBI current=1 box=0 | SF |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
BAL@OAKHIT
Predicted: BAL 7-6 (conf -0.11, moderate)Actual: BAL 5-3Brier 0.198total bias +5
Root cause: Right winner, wrong score
Evidence-backed diagnosis- published system edge 0.11 (moderate); this is not the half-scale distance from 50%
- team-run absolute error 5; total bias +5
- pregame home-minus-away evidence: season -0.966, recent -0.248, pitcher -0.027, composite -0.832
- confidence pipeline reached 0.566 before later publication/recency controls; published home probability was 0.445
- score-volatility actions: no_change
- structural audit exposed 8 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs over-projected by 5 (13 vs 8)
- OAK offense over-projected by 3
- Jack Perkins (OAK) allowed 5 fewer ER than projected (1 vs 6)
| Team | Starter | Projected | Actual | ER err |
|---|
| BAL | Shane Baz | 4 ER / 18 outs | 2 ER / 17 outs | -2 |
| OAK | Jack Perkins | 6 ER / 12 outs | 1 ER / 15 outs | -5 |
Decision internals: run factors home×1.054 / away×1.081; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | -0.966 | positive favors home; negative favors away |
| Recent form | -0.248 | positive favors home; negative favors away |
| Starting-pitcher matchup | -0.027 | positive favors home; negative favors away |
| Composite | -0.832 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 5.61-4.38 | n/a | Base ML historical model prediction |
| variance_applied | 6-4 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 7-6 | 0.555 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 7-6 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 7-6 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 7-6 | n/a | Capped at max 12 realistic runs |
| simulation_result | 7-6 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 5.61-4.38 | 6-4 | 0.772 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 6-4 | 7-6 | -1 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 7-6 | 7-6 | -0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 7-6 | 7-6 | -0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 7-6 | 7-6 | -0 | Capped at max 12 realistic runs |
| simulation_result | 7-6 | 7-6 | -0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| BAL | Colton Cowser | batter / CF | worse | predicted_AB 2.429 |
| BAL | Christian Encarnacion-Strand | batter / 3B | better | predicted_RBI 2.2 |
| OAK | Carlos Cortes | batter / LF | worse | predicted_AB 2.162 |
| BAL | Leody Taveras | batter / RF | worse | predicted_SO 2.089 |
| OAK | Chris Roycroft | pitcher / Reliever | boosted | predicted_H 2 |
| BAL | Samuel Basallo | batter / C | better | predicted_SO 1.978 |
| BAL | Coby Mayo | batter / 1B | better | predicted_AB 1.832 |
| BAL | Gunnar Henderson | batter / SS | boosted | predicted_RBI 1.74 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | BAL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | BAL |
| info | CURRENT_PITCHER_DUPLICATION | OAK pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | OAK |
| info | CURRENT_FINALIZED_MATCH_BOX | BAL model-prediction batting totals differ from final simulation box score (AB: current=32, sim=40; H: current=8, sim=10; HR: current=1, sim=4; RBI: current=5, sim=7; R: current=5, sim=7; BB: current=3, sim=4; SO: current=9, sim=11). | BAL |
| info | CURRENT_PLAYER_MATCH_BOX | BAL current finalized hitter lines are model predictions and differ from the simulation box score: Pete Alonso: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Jackson Holliday: AB current=4 box=5, H current=1 box=0, SO current=1 box=0; Dylan Beavers: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, R current=0 box=1; Coby Mayo: AB current=3 box=5, H current=1 box=2, SO current=1 box=0; Samuel Basallo: AB current=4 box=5, HR current=0 box=1, R current=0 box=1, SO current=1 box=3; Leody Taveras: AB current=3 box=4, H current=1 box=0, R current=1 box=0, SO current=1 box=3; Gunnar Henderson: H current=0 box=2, HR current=0 box=1, RBI current=0 box=2, SO current=1 box=2; Christian Encarnacion-Strand: AB current=4 box=5, H current=1 box=3, 2B current=1 box=0, HR current=0 box=2 | BAL |
| info | CURRENT_FINALIZED_MATCH_BOX | OAK model pitcher projections differ from final simulation pitching box score (H: current=9, sim=10; BB: current=3, sim=4; HR: current=3, sim=4). | OAK |
| info | CURRENT_FINALIZED_MATCH_BOX | OAK model-prediction batting totals differ from final simulation box score (AB: current=32, sim=37; H: current=9, sim=7; 2B: current=1, sim=2; 3B: current=0, sim=1; HR: current=1, sim=3; RBI: current=4, sim=6; R: current=5, sim=6; BB: current=3, sim=7). | OAK |
| info | CURRENT_PLAYER_MATCH_BOX | OAK current finalized hitter lines are model predictions and differ from the simulation box score: Zack Gelof: AB current=4 box=3, H current=1 box=2, 2B current=0 box=1, HR current=0 box=1; Henry Bolte: 2B current=0 box=1, SO current=1 box=0; Jeff McNeil: AB current=3 box=4, H current=1 box=0, BB current=0 box=1, SO current=0 box=1; Donovan Walton: AB current=3 box=4, H current=1 box=0, 2B current=1 box=0, R current=1 box=0; Max Muncy: H current=1 box=0, BB current=0 box=2; Carlos Cortes: AB current=3 box=5, HR current=0 box=1, RBI current=0 box=1, R current=0 box=1; Tommy White: AB current=4 box=5, RBI current=1 box=0, R current=0 box=1, SO current=1 box=2; Lawrence Butler: 3B current=0 box=1, SO current=1 box=0 | OAK |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
BOS@NYYHIT
Predicted: BOS 5-4 (conf -0.06, lean)Actual: BOS 6-0Brier 0.221total bias +3
Root cause: Right winner, wrong score
Evidence-backed diagnosis- published system edge 0.06 (lean); this is not the half-scale distance from 50%
- team-run absolute error 5; total bias +3
- pregame home-minus-away evidence: season +0.007, recent -0.266, pitcher -0.032, composite -0.313
- confidence pipeline reached 0.508 before later publication/recency controls; published home probability was 0.470
- score-volatility actions: no_change
- structural audit exposed 8 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs over-projected by 3 (9 vs 6)
- NYY offense over-projected by 4
- Jake Bennett (BOS) allowed 3 fewer ER than projected (0 vs 3)
| Team | Starter | Projected | Actual | ER err |
|---|
| BOS | Jake Bennett | 3 ER / 17 outs | 0 ER / 4 outs | -3 |
| NYY | Carlos Rodón | 3 ER / 17 outs | 1 ER / 11 outs | -2 |
Decision internals: run factors home×0.962 / away×0.994; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 0.007 | positive favors home; negative favors away |
| Recent form | -0.266 | positive favors home; negative favors away |
| Starting-pitcher matchup | -0.032 | positive favors home; negative favors away |
| Composite | -0.313 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.06-4.73 | n/a | Base ML historical model prediction |
| variance_applied | 4-5 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-4 | 0.53 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 5-4 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 5-4 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 5-4 | n/a | Capped at max 12 realistic runs |
| simulation_result | 5-4 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.06-4.73 | 4-5 | -0.326 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-5 | 5-4 | 2 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 5-4 | 5-4 | -0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 5-4 | 5-4 | -0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 5-4 | 5-4 | -0 | Capped at max 12 realistic runs |
| simulation_result | 5-4 | 5-4 | -0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| NYY | Carlos Rodón | pitcher / Starter | better | predicted_SO 3 |
| BOS | Eli White | batter / LF | better | predicted_RBI 2.677 |
| NYY | Paul Goldschmidt | batter / 1B | better | predicted_RBI 2.57 |
| BOS | Adley Rutschman | batter / C | worse | predicted_AB -2.01 |
| NYY | David Bednar | pitcher / Closer | worse | predicted_H 2 |
| NYY | José Caballero | batter / 3B | worse | predicted_BB 1.787 |
| BOS | Jahmai Jones | batter / DH | worse | predicted_AB 1.5 |
| NYY | Ali Sánchez | batter / C | worse | predicted_AB 1.5 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | BOS pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | BOS |
| info | CURRENT_PITCHER_DUPLICATION | NYY pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | NYY |
| info | CURRENT_FINALIZED_MATCH_BOX | BOS model-prediction batting totals differ from final simulation box score (AB: current=35, sim=33; H: current=7, sim=6; 2B: current=2, sim=1; HR: current=1, sim=2; RBI: current=4, sim=5; R: current=4, sim=5; SO: current=7, sim=6). | BOS |
| info | CURRENT_PLAYER_MATCH_BOX | BOS current finalized hitter lines are model predictions and differ from the simulation box score: Jahmai Jones: AB current=3 box=4, SO current=1 box=2; Ceddanne Rafaela: AB current=5 box=4, H current=1 box=2, SO current=1 box=0; Caleb Durbin: H current=1 box=0, RBI current=1 box=0; Wilyer Abreu: H current=1 box=0, 2B current=1 box=0, R current=1 box=0, BB current=1 box=0; Willson Contreras: BB current=1 box=0; Eli White: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, HR current=0 box=1; Adley Rutschman: AB current=5 box=3, RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Nick Sogard: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, R current=1 box=0 | BOS |
| info | CURRENT_FINALIZED_MATCH_BOX | NYY model pitcher projections differ from final simulation pitching box score (BB: current=2, sim=3). | NYY |
| info | CURRENT_FINALIZED_MATCH_BOX | NYY model-prediction batting totals differ from final simulation box score (AB: current=32, sim=33; H: current=7, sim=6; 2B: current=1, sim=0; 3B: current=0, sim=1; BB: current=3, sim=4; SO: current=9, sim=5). | NYY |
| info | CURRENT_PLAYER_MATCH_BOX | NYY current finalized hitter lines are model predictions and differ from the simulation box score: Heliot Ramos: AB current=4 box=3, 3B current=0 box=1, RBI current=0 box=1, R current=0 box=1; Amed Rosario: AB current=3 box=4, H current=1 box=2, R current=0 box=1; Paul Goldschmidt: AB current=4 box=5, HR current=0 box=1, RBI current=1 box=3, R current=0 box=1; José Caballero: AB current=3 box=2, H current=1 box=0, BB current=0 box=2; Cody Bellinger: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; George Lombard Jr.: RBI current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=1 box=0; Ben Rice: HR current=1 box=0, RBI current=1 box=0, BB current=1 box=0, SO current=1 box=0; Ali Sánchez: AB current=3 box=4, SO current=1 box=0 | NYY |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
BOS@NYYHIT
Predicted: NYY 3-4 (conf +0.07, lean)Actual: NYY 2-9Brier 0.217total bias -4
Root cause: Right winner, wrong score
Evidence-backed diagnosis- published system edge 0.07 (lean); this is not the half-scale distance from 50%
- team-run absolute error 6; total bias -4
- pregame home-minus-away evidence: season +0.007, recent -0.266, pitcher -0.007, composite +0.101
- confidence pipeline reached 0.508 before later publication/recency controls; published home probability was 0.534
- score-volatility actions: no_change
- structural audit exposed 8 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs under-projected by 4 (7 vs 11)
- NYY offense under-projected by 5
Decision internals: run factors home×0.996 / away×1.003; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 0.007 | positive favors home; negative favors away |
| Recent form | -0.266 | positive favors home; negative favors away |
| Starting-pitcher matchup | -0.007 | positive favors home; negative favors away |
| Composite | 0.101 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.08-4.49 | n/a | Base ML historical model prediction |
| variance_applied | 4-4 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 3-4 | 0.534 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 3-4 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 3-4 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 3-4 | n/a | Capped at max 12 realistic runs |
| simulation_result | 3-4 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.08-4.49 | 4-4 | -0.404 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-4 | 3-4 | 1 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 3-4 | 3-4 | 0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 3-4 | 3-4 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 3-4 | 3-4 | 0 | Capped at max 12 realistic runs |
| simulation_result | 3-4 | 3-4 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| BOS | Adley Rutschman | batter / C | worse | predicted_SO 2.316 |
| NYY | Heliot Ramos | batter / RF | worse | predicted_SO 1.879 |
| BOS | Eli White | batter / LF | better | predicted_RBI 1.677 |
| NYY | George Lombard Jr. | batter / SS | better | predicted_RBI 1.66 |
| BOS | Jahmai Jones | batter / DH | worse | predicted_AB 1.5 |
| NYY | Ali Sánchez | batter / C | worse | predicted_AB 1.5 |
| BOS | Caleb Durbin | batter / 3B | worse | predicted_SO 1.484 |
| BOS | Ceddanne Rafaela | batter / CF | worse | predicted_H -1.4 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | BOS pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | BOS |
| info | CURRENT_PITCHER_DUPLICATION | NYY pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | NYY |
| info | CURRENT_FINALIZED_MATCH_BOX | BOS model pitcher projections differ from final simulation pitching box score (HR: current=0, sim=1). | BOS |
| info | CURRENT_FINALIZED_MATCH_BOX | BOS model-prediction batting totals differ from final simulation box score (AB: current=35, sim=34; H: current=7, sim=4; 2B: current=2, sim=0; RBI: current=4, sim=3; R: current=4, sim=3; BB: current=3, sim=2; SO: current=7, sim=14). | BOS |
| info | CURRENT_PLAYER_MATCH_BOX | BOS current finalized hitter lines are model predictions and differ from the simulation box score: Jahmai Jones: AB current=3 box=4; Wilyer Abreu: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, R current=1 box=0; Caleb Durbin: R current=0 box=1, SO current=0 box=2; Adley Rutschman: AB current=5 box=4, H current=1 box=0, RBI current=1 box=0, SO current=1 box=3; Ceddanne Rafaela: AB current=5 box=4, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Willson Contreras: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Eli White: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Andruw Monasterio: AB current=3 box=4, SO current=0 box=2 | BOS |
| info | CURRENT_FINALIZED_MATCH_BOX | NYY model-prediction batting totals differ from final simulation box score (AB: current=32, sim=34; H: current=7, sim=5; SO: current=9, sim=6; HBP: current=0, sim=1). | NYY |
| info | CURRENT_PLAYER_MATCH_BOX | NYY current finalized hitter lines are model predictions and differ from the simulation box score: Amed Rosario: AB current=3 box=4, SO current=1 box=2; Heliot Ramos: H current=1 box=0, SO current=1 box=3; Paul Goldschmidt: H current=1 box=0, RBI current=1 box=0, BB current=0 box=1, SO current=1 box=0; José Caballero: AB current=3 box=4, H current=1 box=0, SO current=1 box=0; George Lombard Jr.: 2B current=0 box=1, RBI current=1 box=2, BB current=1 box=0, SO current=1 box=0; Ben Rice: RBI current=1 box=2, BB current=1 box=0, SO current=1 box=0, HBP current=0 box=1; Cody Bellinger: AB current=4 box=3, 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Ali Sánchez: AB current=3 box=4 | NYY |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
CHW@MINHIT
Predicted: CHW 7-5 (conf -0.14, moderate)Actual: CHW 3-2Brier 0.184total bias +7
Root cause: Right winner, wrong score
Evidence-backed diagnosis- published system edge 0.14 (moderate); this is not the half-scale distance from 50%
- team-run absolute error 7; total bias +7
- pregame home-minus-away evidence: season -0.702, recent -0.458, pitcher -0.113, composite -1.109
- confidence pipeline reached 0.581 before later publication/recency controls; published home probability was 0.429
- score-volatility actions: no_change
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs over-projected by 7 (12 vs 5)
- CHW offense over-projected by 4
- MIN offense over-projected by 3
- Bailey Ober (MIN) allowed 4 fewer ER than projected (1 vs 5)
| Team | Starter | Projected | Actual | ER err |
|---|
| CHW | Erick Fedde | 4 ER / 15 outs | 2 ER / 15 outs | -2 |
| MIN | Bailey Ober | 5 ER / 16 outs | 1 ER / 15 outs | -4 |
Decision internals: run factors home×0.984 / away×1.097; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | -0.702 | positive favors home; negative favors away |
| Recent form | -0.458 | positive favors home; negative favors away |
| Starting-pitcher matchup | -0.113 | positive favors home; negative favors away |
| Composite | -1.109 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.53-4.23 | n/a | Base ML historical model prediction |
| variance_applied | 5-4 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 6-5 | 0.571 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail) |
| embedded_savant_run_environment | 7-5 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 7-5 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 7-5 | n/a | Capped at max 12 realistic runs |
| simulation_result | 7-5 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.53-4.23 | 5-4 | 0.704 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-4 | 6-5 | -0 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail) |
| embedded_savant_run_environment | 6-5 | 7-5 | 1 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 7-5 | 7-5 | -0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 7-5 | 7-5 | -0 | Capped at max 12 realistic runs |
| simulation_result | 7-5 | 7-5 | -0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| CHW | Jake Rogers | batter / C | better | predicted_RBI 3.73 |
| CHW | Colson Montgomery | batter / SS | worse | predicted_AB -2.49 |
| CHW | Miguel Vargas | batter / 3B | boosted | predicted_SO 2.229 |
| CHW | Chase Meidroth | batter / 2B | better | predicted_AB -2.1 |
| MIN | Walker Jenkins | batter / CF | worse | predicted_SO 2 |
| MIN | Luke Keaschall | batter / RF | worse | predicted_RBI 1.68 |
| MIN | Brooks Lee | batter / 3B | better | predicted_AB -1.58 |
| CHW | Munetaka Murakami | batter / 1B | worse | predicted_AB 1.404 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | CHW pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | CHW |
| info | CURRENT_PITCHER_DUPLICATION | MIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | MIN |
| info | CURRENT_FINALIZED_MATCH_BOX | CHW model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=8, sim=9; 2B: current=2, sim=3; HR: current=1, sim=0; RBI: current=5, sim=7; R: current=6, sim=7; BB: current=4, sim=5; SO: current=9, sim=8). | CHW |
| info | CURRENT_PLAYER_MATCH_BOX | CHW current finalized hitter lines are model predictions and differ from the simulation box score: Sam Antonacci: AB current=3 box=4, H current=1 box=0, R current=1 box=0, BB current=0 box=1; Braden Montgomery: AB current=4 box=5, 2B current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Munetaka Murakami: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Miguel Vargas: AB current=4 box=5, 2B current=0 box=1, RBI current=0 box=1, BB current=1 box=0; Andrew Benintendi: AB current=3 box=4, H current=1 box=2, RBI current=1 box=0, R current=0 box=1; Chase Meidroth: AB current=4 box=2, 2B current=1 box=0, RBI current=0 box=2, SO current=1 box=0; Colson Montgomery: AB current=4 box=2, H current=1 box=0, RBI current=1 box=0, R current=0 box=1; Tristan Peters: 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0 | CHW |
| info | CURRENT_FINALIZED_MATCH_BOX | MIN model-prediction batting totals differ from final simulation box score (AB: current=34, sim=32; H: current=8, sim=5; 2B: current=1, sim=0; HR: current=1, sim=3; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=1; SO: current=7, sim=12). | MIN |
| info | CURRENT_PLAYER_MATCH_BOX | MIN current finalized hitter lines are model predictions and differ from the simulation box score: Luke Keaschall: 2B current=1 box=0, RBI current=0 box=2, R current=0 box=1, BB current=1 box=0; Kody Clemens: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Brooks Lee: AB current=5 box=3, H current=1 box=2, SO current=1 box=0, HBP current=0 box=1; Royce Lewis: HR current=0 box=1, RBI current=0 box=1, R current=0 box=1; Josh Bell: H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Trevor Larnach: AB current=3 box=4, H current=1 box=0, SO current=0 box=2; Kaelen Culpepper: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=2; Walker Jenkins: H current=1 box=0, BB current=1 box=0, SO current=1 box=3 | MIN |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
CIN@CHCHIT
Predicted: CHC 5-7 (conf +0.12, moderate)Actual: CHC 5-17Brier 0.194total bias -10
Root cause: Right winner, wrong score
Evidence-backed diagnosis- published system edge 0.12 (moderate); this is not the half-scale distance from 50%
- team-run absolute error 10; total bias -10
- pregame home-minus-away evidence: season +1.256, recent +0.454, pitcher +0.089, composite +1.361
- confidence pipeline reached 0.670 before later publication/recency controls; published home probability was 0.560
- score-volatility actions: supported_blowout_margin_expansion
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs under-projected by 10 (12 vs 22)
- CHC offense under-projected by 10
- Andrew Abbott (CIN) allowed 4 more ER than projected (7 vs 3)
- Kevin Gausman (CHC) allowed 3 more ER than projected (5 vs 2)
| Team | Starter | Projected | Actual | ER err |
|---|
| CIN | Andrew Abbott | 3 ER / 15 outs | 7 ER / 9 outs | 4 |
| CHC | Kevin Gausman | 2 ER / 17 outs | 5 ER / 15 outs | 3 |
Decision internals: run factors home×1.086 / away×0.997; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 1.256 | positive favors home; negative favors away |
| Recent form | 0.454 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.089 | positive favors home; negative favors away |
| Composite | 1.361 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 3.95-5.22 | n/a | Base ML historical model prediction |
| variance_applied | 4-5 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-7 | 0.56 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail) |
| embedded_savant_run_environment | 5-7 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 5-7 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 5-7 | n/a | Capped at max 12 realistic runs |
| simulation_result | 5-7 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 3.95-5.22 | 4-5 | -0.276 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-5 | 5-7 | 1 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail) |
| embedded_savant_run_environment | 5-7 | 5-7 | 0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 5-7 | 5-7 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 5-7 | 5-7 | 0 | Capped at max 12 realistic runs |
| simulation_result | 5-7 | 5-7 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| CHC | Carson Kelly | batter / C | better | predicted_H 3.19 |
| CIN | Jose Trevino | batter / C | better | predicted_RBI 2.57 |
| CHC | Matt Shaw | batter / RF | deboosted | predicted_RBI 1.81 |
| CIN | Elly De La Cruz | batter / SS | worse | predicted_H -1.72 |
| CIN | Héctor Rodríguez | batter / RF | worse | predicted_BB 1.706 |
| CHC | Nico Hoerner | batter / SS | boosted | predicted_RBI 1.546 |
| CHC | Tyrone Taylor | batter / LF | better | predicted_AB 1.5 |
| CHC | Pedro Ramirez | batter / 2B | worse | predicted_SO 1.436 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | CHC pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | CHC |
| info | CURRENT_PITCHER_DUPLICATION | CIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | CIN |
| info | CURRENT_FINALIZED_MATCH_BOX | CHC model-prediction batting totals differ from final simulation box score (AB: current=34, sim=38; H: current=8, sim=12; 2B: current=2, sim=3; HR: current=1, sim=3; RBI: current=5, sim=7; R: current=5, sim=7). | CHC |
| info | CURRENT_PLAYER_MATCH_BOX | CHC current finalized hitter lines are model predictions and differ from the simulation box score: Seiya Suzuki: RBI current=1 box=0; Nico Hoerner: H current=1 box=2, RBI current=0 box=2, R current=0 box=1, BB current=0 box=1; Michael Busch: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, SO current=1 box=0; Alex Bregman: 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Pete Crow-Armstrong: 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Matt Shaw: AB current=3 box=4, RBI current=0 box=2, R current=0 box=1, BB current=0 box=1; Pedro Ramirez: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, SO current=0 box=2; Tyrone Taylor: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=1, R current=0 box=1 | CHC |
| info | CURRENT_FINALIZED_MATCH_BOX | CIN model-prediction batting totals differ from final simulation box score (H: current=8, sim=5; 2B: current=1, sim=0; HR: current=1, sim=4; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=4; SO: current=9, sim=7). | CIN |
| info | CURRENT_PLAYER_MATCH_BOX | CIN current finalized hitter lines are model predictions and differ from the simulation box score: Héctor Rodríguez: AB current=3 box=2, H current=1 box=0, BB current=0 box=2; Tyler Stephenson: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, SO current=1 box=2; Eugenio Suárez: AB current=4 box=3, HR current=0 box=1, R current=0 box=1, SO current=2 box=1; Jose Trevino: AB current=3 box=4, H current=1 box=2, HR current=0 box=2, RBI current=0 box=3; Sal Stewart: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; JJ Bleday: HR current=0 box=1, RBI current=0 box=1, BB current=1 box=0; Elly De La Cruz: AB current=5 box=4, H current=2 box=0, 2B current=1 box=0, RBI current=1 box=0; TJ Friedl: AB current=3 box=4 | CIN |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
COL@ATLHIT
Predicted: ATL 2-7 (conf +0.47, strong)Actual: ATL 1-2Brier 0.071total bias +6
Root cause: Right winner, wrong score
Evidence-backed diagnosis- published system edge 0.47 (strong); this is not the half-scale distance from 50%
- team-run absolute error 6; total bias +6
- pregame home-minus-away evidence: season +1.644, recent +0.962, pitcher -0.003, composite +2.038
- confidence pipeline reached 0.655 before later publication/recency controls; published home probability was 0.734
- score-volatility actions: no_change
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs over-projected by 6 (9 vs 3)
- ATL offense over-projected by 5
- Zach Agnos (COL) allowed 4 fewer ER than projected (0 vs 4)
| Team | Starter | Projected | Actual | ER err |
|---|
| COL | Zach Agnos | 4 ER / 13 outs | 0 ER / 8 outs | -4 |
| ATL | Martín Pérez | 0 ER / 17 outs | 1 ER / 17 outs | 1 |
Decision internals: run factors home×1.013 / away×1.016; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 1.644 | positive favors home; negative favors away |
| Recent form | 0.962 | positive favors home; negative favors away |
| Starting-pitcher matchup | -0.003 | positive favors home; negative favors away |
| Composite | 2.038 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 3.62-5.96 | n/a | Base ML historical model prediction |
| variance_applied | 4-6 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 2-6 | 0.734 | Enhanced ML with recent form, H2H, low-score analysis |
| embedded_savant_run_environment | 2-7 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 2-7 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 2-7 | n/a | Capped at max 12 realistic runs |
| simulation_result | 2-7 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 3.62-5.96 | 4-6 | -0.34 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-6 | 2-6 | 2 | Enhanced ML with recent form, H2H, low-score analysis |
| embedded_savant_run_environment | 2-6 | 2-7 | 1 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 2-7 | 2-7 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 2-7 | 2-7 | 0 | Capped at max 12 realistic runs |
| simulation_result | 2-7 | 2-7 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| ATL | Michael Harris II | batter / CF | worse | predicted_AB -2.5 |
| ATL | Ronald Acuña Jr. | batter / RF | worse | predicted_AB -1.48 |
| ATL | Mike Yastrzemski | batter / LF | worse | predicted_AB 1.369 |
| COL | Hunter Goodman | batter / C | worse | predicted_AB -1.33 |
| ATL | Austin Riley | batter / 3B | better | predicted_SO -1.258 |
| ATL | Drake Baldwin | batter / C | worse | predicted_AB -1.23 |
| ATL | Dominic Smith | batter / DH | better | predicted_AB 1.209 |
| COL | Jake McCarthy | batter / DH | worse | predicted_H -1.114 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | ATL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | ATL |
| info | CURRENT_PITCHER_DUPLICATION | COL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | COL |
| info | CURRENT_FINALIZED_MATCH_BOX | ATL model-prediction batting totals differ from final simulation box score (AB: current=35, sim=32; 2B: current=1, sim=3; HR: current=1, sim=2; RBI: current=4, sim=7; R: current=5, sim=7; SO: current=7, sim=6; SF: current=0, sim=1; HBP: current=0, sim=1). | ATL |
| info | CURRENT_PLAYER_MATCH_BOX | ATL current finalized hitter lines are model predictions and differ from the simulation box score: Mauricio Dubón: AB current=4 box=3, H current=1 box=0, R current=0 box=1, BB current=0 box=1; Drake Baldwin: AB current=5 box=4, RBI current=1 box=2; Ronald Acuña Jr.: AB current=4 box=3, H current=1 box=0, R current=1 box=0, SO current=1 box=2; Matt Olson: H current=1 box=2, 2B current=1 box=0, HR current=0 box=1, RBI current=0 box=1; Michael Harris II: AB current=4 box=2, RBI current=1 box=2, SF current=0 box=1, HBP current=0 box=1; Mike Yastrzemski: AB current=3 box=4, 2B current=0 box=1, R current=0 box=1; Ozzie Albies: H current=1 box=2, RBI current=0 box=1, SO current=1 box=0; Austin Riley: 2B current=0 box=1, RBI current=1 box=0, R current=0 box=1, SO current=1 box=0 | ATL |
| info | CURRENT_FINALIZED_MATCH_BOX | COL model-prediction batting totals differ from final simulation box score (AB: current=34, sim=30; H: current=8, sim=3; 2B: current=2, sim=1; HR: current=1, sim=2; RBI: current=4, sim=2; R: current=4, sim=2; SO: current=9, sim=7). | COL |
| info | CURRENT_PLAYER_MATCH_BOX | COL current finalized hitter lines are model predictions and differ from the simulation box score: Jake McCarthy: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Cole Carrigg: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Willi Castro: H current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=1 box=2; Mickey Moniak: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Hunter Goodman: AB current=5 box=4, 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0; Connor Norby: HR current=0 box=1, RBI current=0 box=1, R current=0 box=1, BB current=0 box=1; Jordan Beck: AB current=3 box=2, BB current=0 box=1; TJ Rumfield: HR current=0 box=1, RBI current=0 box=1, R current=0 box=1, BB current=1 box=0 | COL |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
HOU@NYMHIT
Predicted: NYM 4-5 (conf +0.06, lean)Actual: NYM 2-6Brier 0.221total bias +1
Root cause: Solid correct call
Evidence-backed diagnosis- published system edge 0.06 (lean); this is not the half-scale distance from 50%
- team-run absolute error 3; total bias +1
- pregame home-minus-away evidence: season -0.165, recent -0.490, pitcher +0.068, composite +0.116
- confidence pipeline reached 0.508 before later publication/recency controls; published home probability was 0.530
- score-volatility actions: no_change
- structural audit exposed 8 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- Bennett Sousa (HOU) allowed 4 fewer ER than projected (1 vs 5)
| Team | Starter | Projected | Actual | ER err |
|---|
| HOU | Bennett Sousa | 5 ER / 12 outs | 1 ER / 2 outs | -4 |
| NYM | Nolan McLean | 1 ER / 17 outs | 0 ER / 21 outs | -1 |
Decision internals: run factors home×1 / away×0.932; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | -0.165 | positive favors home; negative favors away |
| Recent form | -0.49 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.068 | positive favors home; negative favors away |
| Composite | 0.116 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.02-4.92 | n/a | Base ML historical model prediction |
| variance_applied | 4-5 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-5 | 0.53 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 4-5 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-5 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-5 | n/a | Capped at max 12 realistic runs |
| simulation_result | 4-5 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.02-4.92 | 4-5 | 0.1 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-5 | 4-5 | 0 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 4-5 | 4-5 | 0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-5 | 4-5 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-5 | 4-5 | 0 | Capped at max 12 realistic runs |
| simulation_result | 4-5 | 4-5 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| NYM | A.J. Ewing | batter / CF | better | predicted_RBI 2.699 |
| HOU | Daulton Varsho | batter / CF | worse | predicted_SO 2.186 |
| HOU | Jeremy Peña | batter / DH | boosted | predicted_SO 2.036 |
| NYM | Carson Benge | batter / RF | worse | predicted_AB -1.832 |
| HOU | Christian Walker | batter / 1B | worse | predicted_AB -1.78 |
| NYM | Francisco Alvarez | batter / C | worse | predicted_AB -1.66 |
| NYM | Jared Young | batter / 1B | better | predicted_SO 1.362 |
| HOU | Christian Vázquez | batter / C | worse | predicted_SO 1.36 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | HOU pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | HOU |
| info | CURRENT_PITCHER_DUPLICATION | NYM pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | NYM |
| info | CURRENT_FINALIZED_MATCH_BOX | HOU model-prediction batting totals differ from final simulation box score (AB: current=34, sim=31; H: current=8, sim=4; 2B: current=1, sim=0; HR: current=1, sim=3; BB: current=3, sim=5; SO: current=8, sim=12). | HOU |
| info | CURRENT_PLAYER_MATCH_BOX | HOU current finalized hitter lines are model predictions and differ from the simulation box score: Jeremy Peña: HR current=0 box=1, RBI current=0 box=1, SO current=1 box=3; Yordan Alvarez: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Isaac Paredes: AB current=4 box=3, HR current=0 box=1, R current=0 box=1; Taylor Trammell: R current=0 box=1, BB current=0 box=1; Christian Walker: AB current=4 box=2, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Daulton Varsho: AB current=5 box=4, 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=2; Jose Altuve: H current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=1 box=2; Christian Vázquez: AB current=3 box=4, H current=1 box=0, SO current=1 box=2 | HOU |
| info | CURRENT_FINALIZED_MATCH_BOX | NYM model pitcher projections differ from final simulation pitching box score (H: current=5, sim=4; HR: current=2, sim=3). | NYM |
| info | CURRENT_FINALIZED_MATCH_BOX | NYM model-prediction batting totals differ from final simulation box score (AB: current=35, sim=31; H: current=8, sim=7; HR: current=1, sim=2; RBI: current=3, sim=5; R: current=4, sim=5; BB: current=3, sim=8; SO: current=8, sim=7). | NYM |
| info | CURRENT_PLAYER_MATCH_BOX | NYM current finalized hitter lines are model predictions and differ from the simulation box score: Juan Soto: AB current=4 box=3, BB current=1 box=2, SO current=1 box=0; Carson Benge: AB current=4 box=2, RBI current=1 box=0, BB current=0 box=2; Marcus Semien: H current=1 box=0, SO current=1 box=0; A.J. Ewing: HR current=0 box=1, RBI current=0 box=3, R current=0 box=1; Brett Baty: BB current=0 box=1; Jared Young: AB current=3 box=4, 2B current=1 box=0, RBI current=0 box=1, R current=0 box=1; Bo Bichette: AB current=4 box=5, 2B current=0 box=1, R current=1 box=0; Francisco Lindor: AB current=5 box=4, RBI current=1 box=0, R current=1 box=0 | NYM |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
KC@CLEHIT
Predicted: KC 5-4 (conf -0.10, moderate)Actual: KC 8-3Brier 0.201total bias -2
Root cause: Solid correct call
Evidence-backed diagnosis- published system edge 0.10 (moderate); this is not the half-scale distance from 50%
- team-run absolute error 4; total bias -2
- pregame home-minus-away evidence: season +0.327, recent -0.258, pitcher -0.089, composite -0.261
- confidence pipeline reached 0.558 before later publication/recency controls; published home probability was 0.448
- score-volatility actions: no_change
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- KC offense under-projected by 3
| Team | Starter | Projected | Actual | ER err |
|---|
| KC | Daniel Lynch IV | 3 ER / 15 outs | 2 ER / 10 outs | -1 |
| CLE | Foster Griffin | 2 ER / 17 outs | 2 ER / 15 outs | 0 |
Decision internals: run factors home×0.977 / away×1.066; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 0.327 | positive favors home; negative favors away |
| Recent form | -0.258 | positive favors home; negative favors away |
| Starting-pitcher matchup | -0.089 | positive favors home; negative favors away |
| Composite | -0.261 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.96-3.85 | n/a | Base ML historical model prediction |
| variance_applied | 5-4 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-4 | 0.552 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 5-4 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 5-4 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 5-4 | n/a | Capped at max 12 realistic runs |
| simulation_result | 5-4 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.96-3.85 | 5-4 | -0.102 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-4 | 5-4 | -0 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 5-4 | 5-4 | -0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 5-4 | 5-4 | -0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 5-4 | 5-4 | -0 | Capped at max 12 realistic runs |
| simulation_result | 5-4 | 5-4 | -0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| KC | Luke Maile | batter / C | neutral | predicted_RBI 2.91 |
| KC | Nick Loftin | batter / 3B | worse | predicted_SO 2.415 |
| CLE | Jo Adell | batter / RF | worse | predicted_RBI 2.33 |
| KC | Jac Caglianone | batter / RF | worse | predicted_SO 1.976 |
| KC | Tyler Tolbert | batter / CF | worse | predicted_RBI 1.818 |
| KC | Carter Jensen | batter / DH | better | predicted_AB 1.468 |
| CLE | Brayan Rocchio | batter / SS | worse | predicted_SO 1.421 |
| CLE | Steven Kwan | batter / CF | better | predicted_AB 1.408 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | CLE pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | CLE |
| info | CURRENT_PITCHER_DUPLICATION | KC pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | KC |
| info | CURRENT_FINALIZED_MATCH_BOX | CLE model-prediction batting totals differ from final simulation box score (AB: current=35, sim=33; H: current=8, sim=6; 3B: current=0, sim=1; R: current=5, sim=4; BB: current=3, sim=4; SO: current=7, sim=8). | CLE |
| info | CURRENT_PLAYER_MATCH_BOX | CLE current finalized hitter lines are model predictions and differ from the simulation box score: Chase DeLauter: AB current=4 box=3, RBI current=1 box=0; Steven Kwan: AB current=4 box=5, H current=1 box=2, 3B current=0 box=1, BB current=1 box=0; David Fry: H current=0 box=1, BB current=0 box=1; Angel Genao: AB current=4 box=3, H current=1 box=0, BB current=0 box=1; Angel Martínez: H current=1 box=0, SO current=1 box=2; José Ramírez: AB current=4 box=3, SO current=1 box=0; Austin Hedges: AB current=3 box=4, H current=1 box=0; Jo Adell: AB current=5 box=4, RBI current=1 box=3 | CLE |
| info | CURRENT_FINALIZED_MATCH_BOX | KC model-prediction batting totals differ from final simulation box score (AB: current=34, sim=35; 2B: current=2, sim=0; HR: current=1, sim=2; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=2; SO: current=8, sim=13). | KC |
| info | CURRENT_PLAYER_MATCH_BOX | KC current finalized hitter lines are model predictions and differ from the simulation box score: Carter Jensen: AB current=4 box=5, HR current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Salvador Perez: RBI current=1 box=0; Tyler Tolbert: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=2, SO current=1 box=2; Isaac Collins: H current=1 box=2, R current=0 box=1, SO current=1 box=0; Nick Loftin: RBI current=1 box=0, SO current=0 box=3; Matthew Lugo: BB current=0 box=1; Bobby Witt Jr.: AB current=5 box=4, 2B current=1 box=0, BB current=1 box=0; Jac Caglianone: AB current=5 box=4, H current=2 box=0, 2B current=1 box=0, RBI current=1 box=0 | KC |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
LAD@DETMISS
Predicted: LAD 4-2 (conf -0.33, strong)Actual: DET 1-2Brier 0.443total bias +3
Root cause: Backed the wrong side with real conviction
Evidence-backed diagnosis- published system edge 0.33 (strong); this is not the half-scale distance from 50%
- team-run absolute error 3; total bias +3
- pregame home-minus-away evidence: season -0.832, recent -1.250, pitcher +0.014, composite -1.099
- confidence pipeline reached 0.661 before later publication/recency controls; published home probability was 0.335
- score-volatility actions: no_change
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs over-projected by 3 (6 vs 3)
- LAD offense over-projected by 3
- committed to LAD at 0.33 confidence and it lost
| Team | Starter | Projected | Actual | ER err |
|---|
| LAD | Blake Snell | 2 ER / 16 outs | 1 ER / 15 outs | -1 |
| DET | Keider Montero | 2 ER / 18 outs | 1 ER / 16 outs | -1 |
Decision internals: run factors home×0.989 / away×0.974; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | -0.832 | positive favors home; negative favors away |
| Recent form | -1.25 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.014 | positive favors home; negative favors away |
| Composite | -1.099 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 5.32-3.23 | n/a | Base ML historical model prediction |
| variance_applied | 5-3 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-2 | 0.665 | Enhanced ML with recent form, H2H, low-score analysis |
| embedded_savant_run_environment | 4-2 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-2 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-2 | n/a | Capped at max 12 realistic runs |
| simulation_result | 4-2 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 5.32-3.23 | 5-3 | -0.088 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-3 | 4-2 | -0 | Enhanced ML with recent form, H2H, low-score analysis |
| embedded_savant_run_environment | 4-2 | 4-2 | -0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-2 | 4-2 | -0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-2 | 4-2 | -0 | Capped at max 12 realistic runs |
| simulation_result | 4-2 | 4-2 | -0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| DET | Spencer Torkelson | batter / 1B | worse | predicted_SO 1.758 |
| DET | Max Clark | batter / CF | worse | predicted_AB -1.6 |
| DET | Ben Malgeri | batter / RF | worse | predicted_AB 1.5 |
| LAD | Max Muncy | batter / 3B | better | predicted_RBI 1.49 |
| LAD | Hunter Feduccia | batter / C | worse | predicted_AB 1.486 |
| DET | Javier Báez | batter / LF | worse | predicted_AB -1.432 |
| DET | Kevin McGonigle | batter / SS | worse | predicted_SO 1.38 |
| LAD | Shohei Ohtani | batter / DH | better | predicted_RBI 1.371 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | DET pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | DET |
| info | CURRENT_PITCHER_DUPLICATION | LAD pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | LAD |
| info | CURRENT_FINALIZED_MATCH_BOX | DET model-prediction batting totals differ from final simulation box score (AB: current=31, sim=29; H: current=8, sim=3; 2B: current=2, sim=1; RBI: current=4, sim=2; R: current=5, sim=2; BB: current=3, sim=4; SO: current=8, sim=11; SF: current=0, sim=1). | DET |
| info | CURRENT_PLAYER_MATCH_BOX | DET current finalized hitter lines are model predictions and differ from the simulation box score: Hao-Yu Lee: RBI current=0 box=1, R current=0 box=1, SO current=1 box=0, SF current=0 box=1; Dillon Dingler: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Eduardo Valencia: AB current=3 box=4, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Ben Malgeri: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=1, R current=0 box=1; Javier Báez: AB current=3 box=2, R current=1 box=0, BB current=0 box=1, SO current=1 box=0; Kevin McGonigle: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Spencer Torkelson: AB current=3 box=4, SO current=1 box=3; Gleyber Torres: AB current=4 box=3, H current=1 box=0, R current=1 box=0, SO current=1 box=2 | DET |
| info | CURRENT_FINALIZED_MATCH_BOX | LAD model-prediction batting totals differ from final simulation box score (HR: current=1, sim=2; R: current=5, sim=4; BB: current=4, sim=2; SO: current=8, sim=12; HBP: current=0, sim=1). | LAD |
| info | CURRENT_PLAYER_MATCH_BOX | LAD current finalized hitter lines are model predictions and differ from the simulation box score: Shohei Ohtani: AB current=4 box=5, H current=1 box=2, RBI current=1 box=2, BB current=1 box=0; Max Muncy: HR current=0 box=1, RBI current=1 box=2, BB current=1 box=0, SO current=1 box=2; Kyle Tucker: H current=1 box=0, R current=1 box=0, BB current=1 box=0; Teoscar Hernández: AB current=4 box=3, SO current=1 box=2, HBP current=0 box=1; Tommy Edman: AB current=4 box=3, 2B current=0 box=1, RBI current=1 box=0, R current=1 box=0; Freddie Freeman: 2B current=1 box=0, SO current=1 box=2; Mookie Betts: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, R current=0 box=1; Hunter Feduccia: AB current=3 box=4 | LAD |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
MIA@WASHIT
Predicted: WAS 2-3 (conf +0.12, moderate)Actual: WAS 4-5Brier 0.193total bias -4
Root cause: Solid correct call
Evidence-backed diagnosis- published system edge 0.12 (moderate); this is not the half-scale distance from 50%
- team-run absolute error 4; total bias -4
- pregame home-minus-away evidence: season +0.068, recent -0.006, pitcher +0.009, composite +0.236
- confidence pipeline reached 0.581 before later publication/recency controls; published home probability was 0.560
- score-volatility actions: no_change
- structural audit exposed 8 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs under-projected by 4 (5 vs 9)
| Team | Starter | Projected | Actual | ER err |
|---|
| MIA | Sandy Alcantara | 3 ER / 19 outs | 4 ER / 18 outs | 1 |
| WAS | Cade Cavalli | 2 ER / 18 outs | 0 ER / 12 outs | -2 |
Decision internals: run factors home×0.994 / away×0.985; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 0.068 | positive favors home; negative favors away |
| Recent form | -0.006 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.009 | positive favors home; negative favors away |
| Composite | 0.236 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.59-5.52 | n/a | Base ML historical model prediction |
| variance_applied | 5-6 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 2-3 | 0.56 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 2-3 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 2-3 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 2-3 | n/a | Capped at max 12 realistic runs |
| simulation_result | 2-3 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.59-5.52 | 5-6 | 0.071 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-6 | 2-3 | 0 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 2-3 | 2-3 | 0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 2-3 | 2-3 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 2-3 | 2-3 | 0 | Capped at max 12 realistic runs |
| simulation_result | 2-3 | 2-3 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| WAS | Brady House | batter / 3B | better | predicted_RBI 2.38 |
| MIA | Griffin Conine | batter / DH | better | predicted_RBI 1.531 |
| WAS | José Tena | batter / LF | worse | predicted_AB 1.5 |
| MIA | Owen Caissie | batter / RF | worse | predicted_SO -1.177 |
| MIA | Kyle Stowers | batter / 1B | worse | predicted_H -1.06 |
| MIA | Heriberto Hernández | batter / LF | worse | predicted_SO -1.01 |
| WAS | Keibert Ruiz | batter / C | worse | predicted_AB -1.01 |
| WAS | CJ Abrams | batter / DH | worse | predicted_SO -1.008 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | MIA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | MIA |
| info | CURRENT_PITCHER_DUPLICATION | WAS pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | WAS |
| info | CURRENT_FINALIZED_MATCH_BOX | MIA model pitcher projections differ from final simulation pitching box score (H: current=5, sim=6). | MIA |
| info | CURRENT_FINALIZED_MATCH_BOX | MIA model-prediction batting totals differ from final simulation box score (H: current=8, sim=6; 2B: current=1, sim=0; RBI: current=4, sim=2; R: current=5, sim=2; BB: current=3, sim=1; SO: current=8, sim=7). | MIA |
| info | CURRENT_PLAYER_MATCH_BOX | MIA current finalized hitter lines are model predictions and differ from the simulation box score: Griffin Conine: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=2; Heriberto Hernández: HR current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=0; Otto Lopez: AB current=5 box=4, H current=1 box=2, RBI current=1 box=0; Jakob Marsee: R current=1 box=0, BB current=1 box=0, SO current=1 box=0; Kyle Stowers: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Xavier Edwards: RBI current=1 box=0, BB current=1 box=0; Javier Sanoja: H current=1 box=0, 2B current=1 box=0, BB current=0 box=1, SO current=0 box=1; Owen Caissie: SO current=1 box=0 | MIA |
| info | CURRENT_FINALIZED_MATCH_BOX | WAS model-prediction batting totals differ from final simulation box score (AB: current=32, sim=30; H: current=8, sim=6; 2B: current=2, sim=0; RBI: current=5, sim=3; R: current=4, sim=3; BB: current=3, sim=1; SO: current=8, sim=5). | WAS |
| info | CURRENT_PLAYER_MATCH_BOX | WAS current finalized hitter lines are model predictions and differ from the simulation box score: José Tena: AB current=3 box=4, H current=1 box=0; CJ Abrams: 2B current=1 box=0, RBI current=1 box=0, BB current=1 box=0, SO current=1 box=0; Dylan Crews: AB current=4 box=3, RBI current=1 box=0, R current=1 box=0, SO current=1 box=0; Daylen Lile: HR current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=0; Jorbit Vivas: AB current=3 box=2, R current=0 box=1, SO current=1 box=0; Brady House: AB current=4 box=3, HR current=0 box=1, RBI current=1 box=3, R current=0 box=1; Abimelec Ortiz: AB current=3 box=4, H current=1 box=0; Keibert Ruiz: AB current=4 box=3, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0 | WAS |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
PHI@LAAHIT
Predicted: PHI 6-3 (conf -0.33, strong)Actual: PHI 4-2Brier 0.112total bias +3
Root cause: Strong correct call
Evidence-backed diagnosis- published system edge 0.33 (strong); this is not the half-scale distance from 50%
- team-run absolute error 3; total bias +3
- pregame home-minus-away evidence: season -0.877, recent -0.978, pitcher +0.050, composite -1.157
- confidence pipeline reached 0.661 before later publication/recency controls; published home probability was 0.335
- score-volatility actions: no_change
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs over-projected by 3 (9 vs 6)
| Team | Starter | Projected | Actual | ER err |
|---|
| PHI | Cristopher Sánchez | 1 ER / 18 outs | 0 ER / 21 outs | -1 |
| LAA | Ryan Johnson | 3 ER / 16 outs | 1 ER / 15 outs | -2 |
Decision internals: run factors home×0.986 / away×0.936; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | -0.877 | positive favors home; negative favors away |
| Recent form | -0.978 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.05 | positive favors home; negative favors away |
| Composite | -1.157 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.73-2.64 | n/a | Base ML historical model prediction |
| variance_applied | 5-3 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-3 | 0.665 | Enhanced ML with recent form, H2H, low-score analysis |
| embedded_savant_run_environment | 6-3 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 6-3 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 6-3 | n/a | Capped at max 12 realistic runs |
| simulation_result | 6-3 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.73-2.64 | 5-3 | -0.086 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-3 | 5-3 | -0 | Enhanced ML with recent form, H2H, low-score analysis |
| embedded_savant_run_environment | 5-3 | 6-3 | 1 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 6-3 | 6-3 | -0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 6-3 | 6-3 | -0 | Capped at max 12 realistic runs |
| simulation_result | 6-3 | 6-3 | -0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| PHI | Trea Turner | batter / SS | neutral | predicted_RBI 2.64 |
| PHI | Cristopher Sánchez | pitcher / Starter | better | predicted_SO 2 |
| PHI | Orion Kerkering | pitcher / Middle Reliever (High Leverage) | worse | predicted_SO -2 |
| PHI | Kyle Schwarber | batter / DH | better | predicted_AB -1.75 |
| LAA | Vaughn Grissom | batter / 1B | worse | predicted_AB -1.607 |
| LAA | Christian Moore | batter / 2B | better | predicted_SO -1.5 |
| LAA | Travis d'Arnaud | batter / C | worse | predicted_AB 1.5 |
| PHI | Luis Arraez | batter / 2B | worse | predicted_H -1.45 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | LAA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | LAA |
| info | CURRENT_PITCHER_DUPLICATION | PHI pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | PHI |
| info | CURRENT_FINALIZED_MATCH_BOX | LAA model-prediction batting totals differ from final simulation box score (H: current=7, sim=4; 2B: current=1, sim=0; 3B: current=0, sim=1; HR: current=1, sim=0; R: current=4, sim=3; BB: current=3, sim=4; SO: current=10, sim=4; HBP: current=0, sim=1). | LAA |
| info | CURRENT_PLAYER_MATCH_BOX | LAA current finalized hitter lines are model predictions and differ from the simulation box score: Vaughn Grissom: AB current=4 box=2, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Jose Siri: BB current=0 box=1, SO current=1 box=0; Oswald Peraza: AB current=3 box=4, H current=0 box=1, RBI current=0 box=1, R current=0 box=1; Christian Moore: AB current=3 box=4, 3B current=0 box=1, RBI current=0 box=1, R current=0 box=1; Zach Neto: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Wade Meckler: H current=1 box=0, R current=1 box=0; Mike Trout: AB current=4 box=3, RBI current=0 box=1, SO current=1 box=0; Denzer Guzman: H current=1 box=0, RBI current=1 box=0 | LAA |
| info | CURRENT_FINALIZED_MATCH_BOX | PHI model-prediction batting totals differ from final simulation box score (AB: current=38, sim=33; H: current=9, sim=7; 2B: current=2, sim=3; HR: current=1, sim=0; RBI: current=4, sim=6; R: current=5, sim=6; BB: current=3, sim=6; SO: current=8, sim=4). | PHI |
| info | CURRENT_PLAYER_MATCH_BOX | PHI current finalized hitter lines are model predictions and differ from the simulation box score: Bryce Harper: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, SF current=0 box=1; Alec Bohm: RBI current=1 box=0; Bryson Stott: H current=1 box=0, 2B current=1 box=0, BB current=1 box=0, SO current=1 box=0; Brandon Marsh: AB current=4 box=3, BB current=0 box=1, SO current=1 box=0; J.T. Realmuto: RBI current=1 box=0, SO current=1 box=0; Trea Turner: H current=1 box=2, 2B current=0 box=2, RBI current=0 box=3; Luis Arraez: AB current=5 box=4, H current=1 box=0, BB current=0 box=1, SO current=1 box=0; Kyle Schwarber: AB current=5 box=3, HR current=1 box=0, BB current=1 box=2, SO current=1 box=0 | PHI |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
PIT@STLHIT
Predicted: PIT 4-3 (conf -0.06, lean)Actual: PIT 6-2Brier 0.221total bias -1
Root cause: Solid correct call
Evidence-backed diagnosis- published system edge 0.06 (lean); this is not the half-scale distance from 50%
- team-run absolute error 3; total bias -1
- pregame home-minus-away evidence: season -0.127, recent +0.070, pitcher +0.048, composite -0.127
- confidence pipeline reached 0.508 before later publication/recency controls; published home probability was 0.470
- score-volatility actions: no_change
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
No large single-factor errors.
| Team | Starter | Projected | Actual | ER err |
|---|
| PIT | Lake Bachar | 2 ER / 15 outs | 1 ER / 7 outs | -1 |
| STL | Kyle Leahy | 3 ER / 17 outs | 1 ER / 9 outs | -2 |
Decision internals: run factors home×0.984 / away×0.936; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | -0.127 | positive favors home; negative favors away |
| Recent form | 0.07 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.048 | positive favors home; negative favors away |
| Composite | -0.127 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.83-3.86 | n/a | Base ML historical model prediction |
| variance_applied | 5-4 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-3 | 0.53 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 4-3 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-3 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-3 | n/a | Capped at max 12 realistic runs |
| simulation_result | 4-3 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.83-3.86 | 5-4 | 0.021 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 5-4 | 4-3 | -0 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 4-3 | 4-3 | -0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-3 | 4-3 | -0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-3 | 4-3 | -0 | Capped at max 12 realistic runs |
| simulation_result | 4-3 | 4-3 | -0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| PIT | Jared Triolo | batter / 3B | worse | predicted_SO 2.356 |
| STL | Alec Burleson | batter / 1B | worse | predicted_AB -2.01 |
| STL | Kyle Leahy | pitcher / Starter | better | predicted_SO 2 |
| PIT | Oneil Cruz | batter / CF | worse | predicted_SO -1.556 |
| STL | Joshua Baez | batter / LF | worse | predicted_SO -1.538 |
| STL | Nathan Church | batter / CF | better | predicted_RBI 1.47 |
| STL | Jordan Walker | batter / RF | worse | predicted_H -1.33 |
| PIT | Nick Gonzales | batter / DH | better | predicted_SO 1.309 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | PIT pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | PIT |
| info | CURRENT_PITCHER_DUPLICATION | STL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | STL |
| info | CURRENT_FINALIZED_MATCH_BOX | PIT model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=7, sim=8; 2B: current=1, sim=2; HR: current=1, sim=2; RBI: current=3, sim=4; BB: current=3, sim=1; HBP: current=0, sim=1). | PIT |
| info | CURRENT_PLAYER_MATCH_BOX | PIT current finalized hitter lines are model predictions and differ from the simulation box score: Nick Gonzales: H current=1 box=2, 2B current=0 box=1, R current=0 box=1, SO current=1 box=2; Spencer Horwitz: AB current=3 box=4, H current=1 box=0, SO current=1 box=0, HBP current=0 box=1; Jake Mangum: R current=1 box=0; Brandon Lowe: AB current=4 box=3, H current=0 box=1; Oneil Cruz: RBI current=1 box=2, BB current=1 box=0, SO current=1 box=0; Jared Triolo: AB current=3 box=4, H current=1 box=0, SO current=1 box=3; Bryan Reynolds: R current=1 box=0, BB current=1 box=0; Rafael Flores Jr.: HR current=0 box=1, R current=0 box=1, SO current=1 box=0 | PIT |
| info | CURRENT_FINALIZED_MATCH_BOX | STL model-prediction batting totals differ from final simulation box score (H: current=8, sim=5; RBI: current=4, sim=3; R: current=5, sim=3; BB: current=3, sim=2). | STL |
| info | CURRENT_PLAYER_MATCH_BOX | STL current finalized hitter lines are model predictions and differ from the simulation box score: Iván Herrera: H current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=1 box=2; Jordan Walker: AB current=5 box=4, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Alec Burleson: AB current=4 box=2, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; JJ Wetherholt: H current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=1 box=0; José Fermín: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, HR current=0 box=1; Joshua Baez: HR current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Nolan Gorman: AB current=3 box=4, BB current=1 box=0, SO current=1 box=2; Nathan Church: AB current=3 box=4, RBI current=1 box=2, R current=0 box=1 | STL |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
SD@TBHIT
Predicted: TB 4-5 (conf +0.24, strong)Actual: TB 6-7Brier 0.143total bias -4
Root cause: Solid correct call
Evidence-backed diagnosis- published system edge 0.24 (strong); this is not the half-scale distance from 50%
- team-run absolute error 4; total bias -4
- pregame home-minus-away evidence: season +0.215, recent +0.230, pitcher +0.050, composite +0.465
- confidence pipeline reached 0.616 before later publication/recency controls; published home probability was 0.621
- score-volatility actions: no_change
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
- total runs under-projected by 4 (9 vs 13)
| Team | Starter | Projected | Actual | ER err |
|---|
| SD | Walker Buehler | 2 ER / 15 outs | 3 ER / 15 outs | 1 |
| TB | Nick Martinez | 3 ER / 18 outs | 3 ER / 18 outs | 0 |
Decision internals: run factors home×0.99 / away×0.94; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 0.215 | positive favors home; negative favors away |
| Recent form | 0.23 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.05 | positive favors home; negative favors away |
| Composite | 0.465 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 4.35-4.45 | n/a | Base ML historical model prediction |
| variance_applied | 4-4 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 3-5 | 0.621 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run) |
| embedded_savant_run_environment | 4-5 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-5 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-5 | n/a | Capped at max 12 realistic runs |
| simulation_result | 4-5 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 4.35-4.45 | 4-4 | -0.103 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-4 | 3-5 | 2 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run) |
| embedded_savant_run_environment | 3-5 | 4-5 | -1 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-5 | 4-5 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-5 | 4-5 | 0 | Capped at max 12 realistic runs |
| simulation_result | 4-5 | 4-5 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| TB | Jonny DeLuca | batter / RF | worse | predicted_SO 2.3 |
| SD | Jackson Merrill | batter / CF | worse | predicted_AB -1.69 |
| TB | Richie Palacios | batter / 2B | worse | predicted_BB 1.612 |
| SD | Dustin Harris | batter / LF | worse | predicted_SO 1.529 |
| TB | Liam Hicks | batter / C | better | predicted_RBI 1.51 |
| TB | Yandy Díaz | batter / DH | worse | predicted_SO 1.375 |
| TB | Cedric Mullins | batter / CF | better | predicted_H 1.34 |
| SD | Gavin Sheets | batter / DH | worse | predicted_SO 1.3 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | SD pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | SD |
| info | CURRENT_PITCHER_DUPLICATION | TB pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | TB |
| info | CURRENT_FINALIZED_MATCH_BOX | SD model-prediction batting totals differ from final simulation box score (AB: current=33, sim=36; H: current=7, sim=9; HR: current=1, sim=0; BB: current=3, sim=1). | SD |
| info | CURRENT_PLAYER_MATCH_BOX | SD current finalized hitter lines are model predictions and differ from the simulation box score: Jake Cronenworth: AB current=3 box=4, RBI current=0 box=1, R current=0 box=1, SO current=1 box=0; Ty France: AB current=3 box=4, H current=1 box=2, SO current=1 box=0; Jackson Merrill: AB current=5 box=3, HR current=1 box=0, BB current=0 box=1, SO current=1 box=0; Gavin Sheets: AB current=3 box=4, H current=0 box=1, RBI current=0 box=1, R current=0 box=1; Luis Rengifo: AB current=3 box=4, H current=1 box=0, SO current=0 box=1; Fernando Tatis Jr.: H current=1 box=2, BB current=1 box=0, SO current=1 box=0; Manny Machado: AB current=5 box=4, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Luis Campusano: AB current=3 box=4, RBI current=1 box=0, BB current=1 box=0 | SD |
| info | CURRENT_FINALIZED_MATCH_BOX | TB model-prediction batting totals differ from final simulation box score (AB: current=34, sim=35; H: current=8, sim=9; HR: current=1, sim=0; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=5; SO: current=6, sim=8). | TB |
| info | CURRENT_PLAYER_MATCH_BOX | TB current finalized hitter lines are model predictions and differ from the simulation box score: Jonathan Aranda: H current=1 box=2, RBI current=0 box=1, BB current=0 box=1; Junior Caminero: H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Liam Hicks: RBI current=1 box=2; Yandy Díaz: HR current=1 box=0, SO current=0 box=2; Chandler Simpson: SO current=0 box=1; Richie Palacios: AB current=3 box=2, BB current=0 box=2, SO current=1 box=0; Cedric Mullins: H current=1 box=2, 2B current=0 box=1, R current=0 box=1, BB current=1 box=0; Jonny DeLuca: AB current=3 box=4, H current=1 box=0, 2B current=1 box=0, SO current=1 box=3 | TB |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
SEA@TORHIT
Predicted: TOR 4-5 (conf +0.08, lean)Actual: TOR 3-4Brier 0.212total bias +2
Root cause: Solid correct call
Evidence-backed diagnosis- published system edge 0.08 (lean); this is not the half-scale distance from 50%
- team-run absolute error 2; total bias +2
- pregame home-minus-away evidence: season +0.220, recent +0.508, pitcher +0.067, composite +0.574
- confidence pipeline reached 0.619 before later publication/recency controls; published home probability was 0.540
- score-volatility actions: supported_blowout_margin_expansion
- structural audit exposed 7 non-pass finding(s); see the per-game audit table
Quantified contributing factors
No large single-factor errors.
| Team | Starter | Projected | Actual | ER err |
|---|
| SEA | Kade Anderson | 2 ER / 15 outs | 2 ER / 13 outs | 0 |
| TOR | José Soriano | 3 ER / 16 outs | 3 ER / 21 outs | 0 |
Decision internals: run factors home×0.997 / away×0.931; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 0.22 | positive favors home; negative favors away |
| Recent form | 0.508 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.067 | positive favors home; negative favors away |
| Composite | 0.574 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 3.28-5.38 | n/a | Base ML historical model prediction |
| variance_applied | 3-5 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-5 | 0.54 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 4-5 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-5 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-5 | n/a | Capped at max 12 realistic runs |
| simulation_result | 4-5 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 3.28-5.38 | 3-5 | -0.093 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 3-5 | 4-5 | -1 | Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail) |
| embedded_savant_run_environment | 4-5 | 4-5 | 0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 4-5 | 4-5 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 4-5 | 4-5 | 0 | Capped at max 12 realistic runs |
| simulation_result | 4-5 | 4-5 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| TOR | Myles Straw | batter / CF | better | predicted_H 2.75 |
| SEA | Julio Rodríguez | batter / CF | worse | predicted_RBI 2.4 |
| TOR | Ernie Clement | batter / SS | worse | predicted_AB -1.826 |
| TOR | Vladimir Guerrero Jr. | batter / 1B | better | predicted_RBI 1.65 |
| SEA | Josh Naylor | batter / 1B | worse | predicted_AB -1.58 |
| TOR | Alejandro Kirk | batter / C | worse | predicted_RBI 1.45 |
| TOR | Brett Bateman | batter / RF | worse | predicted_H -1.412 |
| TOR | Daz Cameron | batter / LF | deboosted | predicted_H 1.286 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | SEA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | SEA |
| info | CURRENT_PITCHER_DUPLICATION | TOR pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | TOR |
| info | CURRENT_FINALIZED_MATCH_BOX | SEA model-prediction batting totals differ from final simulation box score (AB: current=37, sim=34; H: current=8, sim=7; 2B: current=1, sim=2; R: current=5, sim=4; BB: current=4, sim=2; SO: current=9, sim=10; HBP: current=0, sim=1). | SEA |
| info | CURRENT_PLAYER_MATCH_BOX | SEA current finalized hitter lines are model predictions and differ from the simulation box score: Dominic Canzone: RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Cal Raleigh: HR current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Josh Naylor: AB current=5 box=3, H current=1 box=0, HBP current=0 box=1; Cole Young: AB current=5 box=4, H current=1 box=0, R current=1 box=0; Randy Arozarena: AB current=4 box=5, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Brock Rodden: AB current=3 box=4, H current=0 box=1, 2B current=0 box=1; Julio Rodríguez: AB current=5 box=4, HR current=0 box=1, RBI current=1 box=3; Taylor Ward: 2B current=0 box=1, RBI current=0 box=1, BB current=1 box=0, SO current=1 box=2 | SEA |
| info | CURRENT_FINALIZED_MATCH_BOX | TOR model-prediction batting totals differ from final simulation box score (AB: current=34, sim=33; H: current=9, sim=10; 2B: current=2, sim=4; HR: current=1, sim=0; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=2, sim=1; SO: current=6, sim=7). | TOR |
| info | CURRENT_PLAYER_MATCH_BOX | TOR current finalized hitter lines are model predictions and differ from the simulation box score: Brett Bateman: H current=1 box=0, R current=1 box=0; George Springer: AB current=5 box=4, H current=2 box=1, 2B current=1 box=0, HR current=1 box=0; Alejandro Kirk: AB current=4 box=3, 2B current=0 box=1, RBI current=1 box=2, R current=0 box=1; Ernie Clement: AB current=4 box=2, H current=1 box=0, BB current=0 box=1, HBP current=0 box=1; Daz Cameron: AB current=3 box=4, H current=1 box=2, RBI current=1 box=0, R current=0 box=1; Vladimir Guerrero Jr.: H current=1 box=2, 2B current=0 box=1, RBI current=0 box=2; Kazuma Okamoto: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Charles McAdoo: AB current=3 box=4, 2B current=0 box=1 | TOR |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
TEX@MILHIT
Predicted: MIL 3-7 (conf +0.47, strong)Actual: MIL 3-5Brier 0.071total bias +2
Root cause: Strong correct call
Evidence-backed diagnosis- published system edge 0.47 (strong); this is not the half-scale distance from 50%
- team-run absolute error 2; total bias +2
- pregame home-minus-away evidence: season +1.545, recent +1.230, pitcher +0.137, composite +2.179
- confidence pipeline reached 0.655 before later publication/recency controls; published home probability was 0.734
- score-volatility actions: no_change
- structural audit exposed 8 non-pass finding(s); see the per-game audit table
Quantified contributing factors
No large single-factor errors.
| Team | Starter | Projected | Actual | ER err |
|---|
| TEX | Cal Quantrill | 4 ER / 17 outs | 3 ER / 18 outs | -1 |
| MIL | Shane Drohan | 1 ER / 16 outs | 1 ER / 15 outs | 0 |
Decision internals: run factors home×1.068 / away×0.931; savant lineup adjustment applied
Full model trace, player evidence, and audit findings
Fundamental evidence entering the pick
| Signal | Home-away edge | Meaning |
|---|
| Season quality | 1.545 | positive favors home; negative favors away |
| Recent form | 1.23 | positive favors home; negative favors away |
| Starting-pitcher matchup | 0.137 | positive favors home; negative favors away |
| Composite | 2.179 | positive favors home; negative favors away |
Score and confidence evolution
| Stage | Away-home runs | Confidence | What changed |
|---|
| historical_ml | 3.77-5.2 | n/a | Base ML historical model prediction |
| variance_applied | 4-5 | n/a | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 3-7 | 0.734 | Enhanced ML with recent form, H2H, low-score analysis |
| embedded_savant_run_environment | 3-7 | n/a | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 3-7 | n/a | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 3-7 | n/a | Capped at max 12 realistic runs |
| simulation_result | 3-7 | n/a | Final inning-by-inning simulation result |
Official postgame stage attribution
| Stage | Before | After | Toward final pick | What changed |
|---|
| variance_applied | 3.77-5.2 | 4-5 | -0.432 | Deterministic baseline target (variance factor 1.0) |
| enhanced_prediction | 4-5 | 3-7 | 3 | Enhanced ML with recent form, H2H, low-score analysis |
| embedded_savant_run_environment | 3-7 | 3-7 | 0 | Target score adjusted by temporal Savant lineup quality versus opposing starter quality |
| fundamental_edge_target_review | 3-7 | 3-7 | 0 | Refreshed score target and win probability after Savant/pitcher/team-strength adjustments |
| capped_target | 3-7 | 3-7 | 0 | Capped at max 12 realistic runs |
| simulation_result | 3-7 | 3-7 | 0 | Final inning-by-inning simulation result |
Largest player-level adjustments from the ultra-detailed trace
These are pregame model adjustments, not claims about actual postgame player error.
| Team | Player | Role | Net effect | Largest change |
|---|
| MIL | Jake Bauers | batter / 1B | better | predicted_RBI 2.39 |
| MIL | William Contreras | batter / C | better | predicted_AB -1.95 |
| MIL | Cooper Pratt | batter / SS | better | predicted_RBI 1.64 |
| TEX | Danny Jansen | batter / C | better | predicted_RBI 1.58 |
| TEX | Brandon Nimmo | batter / RF | worse | predicted_AB -1.57 |
| TEX | Cody Freeman | batter / 3B | worse | predicted_AB 1.421 |
| TEX | Justin Foscue | batter / 2B | worse | predicted_SO 1.329 |
| MIL | Garrett Mitchell | batter / CF | worse | predicted_H -1.1 |
Structural audit findings
| Severity | Rule | Finding | Subject |
|---|
| info | CURRENT_PITCHER_DUPLICATION | MIL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | MIL |
| info | CURRENT_PITCHER_DUPLICATION | TEX pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit. | TEX |
| info | CURRENT_FINALIZED_MATCH_BOX | MIL model-prediction batting totals differ from final simulation box score (AB: current=36, sim=33; 3B: current=0, sim=1; RBI: current=5, sim=7; R: current=5, sim=7; BB: current=4, sim=5; SO: current=8, sim=10; SF: current=0, sim=1). | MIL |
| info | CURRENT_PLAYER_MATCH_BOX | MIL current finalized hitter lines are model predictions and differ from the simulation box score: Brice Turang: H current=1 box=2, BB current=1 box=0, SO current=1 box=0; Jake Bauers: AB current=3 box=4, H current=1 box=2, HR current=0 box=1, RBI current=1 box=3; William Contreras: AB current=5 box=3, RBI current=1 box=0, BB current=0 box=1, SO current=0 box=1; Luis Lara: H current=1 box=0, BB current=0 box=1, SO current=1 box=0; Jackson Chourio: AB current=5 box=4, H current=1 box=2, HR current=1 box=0, RBI current=1 box=0; Garrett Mitchell: AB current=4 box=3, H current=1 box=0, BB current=0 box=1, SO current=1 box=2; Christian Yelich: AB current=5 box=4, RBI current=1 box=0, BB current=1 box=0, SO current=1 box=2; Cooper Pratt: AB current=3 box=4, 3B current=0 box=1, RBI current=0 box=2, R current=0 box=1 | MIL |
| info | CURRENT_FINALIZED_MATCH_BOX | TEX model pitcher projections differ from final simulation pitching box score (H: current=8, sim=10; HR: current=0, sim=1). | TEX |
| info | CURRENT_FINALIZED_MATCH_BOX | TEX model-prediction batting totals differ from final simulation box score (AB: current=33, sim=32; H: current=7, sim=5; 2B: current=2, sim=1; HR: current=1, sim=2; RBI: current=4, sim=3; R: current=4, sim=3; BB: current=3, sim=4; SO: current=9, sim=7). | TEX |
| info | CURRENT_PLAYER_MATCH_BOX | TEX current finalized hitter lines are model predictions and differ from the simulation box score: Corey Seager: RBI current=1 box=0, BB current=1 box=0, SO current=1 box=2; Ezequiel Duran: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Justin Foscue: H current=1 box=0, BB current=0 box=1, SO current=1 box=2; Brandon Nimmo: AB current=5 box=3, H current=1 box=0, 2B current=1 box=0, BB current=0 box=1; Logan O'Hoppe: BB current=0 box=1; Wyatt Langford: H current=1 box=2, 2B current=0 box=1, HR current=0 box=1, RBI current=0 box=1; Jake Burger: HR current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Cody Freeman: AB current=3 box=4, H current=1 box=0, SO current=1 box=0 | TEX |
| info | RECONCILIATION_PROPAGATION | No score_target_reconciliation step was used for this game. | |
Cross-game lever tension
Each candidate fix, and exactly which games it would help versus hurt. A lever that repairs one game and breaks two is visible here before it is proposed.
score.variance_multiplier / score.variance_floor (widen run distributions)
Targets: brier_score, calibration_error · 1 helped vs 4 hurt
Would help- LAD@DET (miss): lost a 0.33-confidence pick; wider tails lower overconfidence
Would hurt / risk- COL@ATL (hit): strong correct call at 0.47; widening softens a good edge
- PHI@LAA (hit): strong correct call at 0.33; widening softens a good edge
- SD@TB (hit): strong correct call at 0.24; widening softens a good edge
- TEX@MIL (hit): strong correct call at 0.47; widening softens a good edge
player.pitcher.v4 form weights / player.pitcher_prior_games
Targets: starter_er_mae, starter_outs_mae · 7 helped vs 5 hurt
Would help- ARI@SF (hit): starter ER off by 5
- BAL@OAK (hit): starter ER off by 5
- BOS@NYY (hit): starter ER off by 3
- CHW@MIN (hit): starter ER off by 4
- CIN@CHC (hit): starter ER off by 4
- COL@ATL (hit): starter ER off by 4
- HOU@NYM (hit): starter ER off by 4
Would hurt / risk- KC@CLE (hit): starter projection was accurate (ER off 1); reweight risks disturbing it
- LAD@DET (miss): starter projection was accurate (ER off 1); reweight risks disturbing it
- SD@TB (hit): starter projection was accurate (ER off 1); reweight risks disturbing it
- SEA@TOR (hit): starter projection was accurate (ER off 0); reweight risks disturbing it
- TEX@MIL (hit): starter projection was accurate (ER off 1); reweight risks disturbing it
recent.* / season.* run-mean weighting
Targets: team_run_mae, total_run_mae, accuracy · 11 helped vs 2 hurt
Would help- ARI@SF (hit): team runs off by 5
- ARI@SF (hit): team runs off by 4
- BAL@OAK (hit): team runs off by 5
- BOS@NYY (hit): team runs off by 5
- BOS@NYY (hit): team runs off by 6
- CHW@MIN (hit): team runs off by 7
- CIN@CHC (hit): team runs off by 10
- COL@ATL (hit): team runs off by 6
- KC@CLE (hit): team runs off by 4
- MIA@WAS (hit): team runs off by 4
- SD@TB (hit): team runs off by 4
Would hurt / risk- SEA@TOR (hit): team runs already close (off 2); reweight risks disturbing it
- TEX@MIL (hit): team runs already close (off 2); reweight risks disturbing it
Day over day: did it improve or worsen
Compared against packet 20260828. Last recorded change: v4.9.0 (2026-08-29) Make Adaptive Status permanently readable, deepen Daily Learning attribution, and complete the apply-observe lifecycle.
| Metric | Prior | Current | Δ | Result |
|---|
| brier_score | 0.2418 | 0.2338 | -0.007973 | improved |
| accuracy | 0.5833 | 0.6348 | +0.0515 | improved |
| team_run_mae | 2.4306 | 2.4261 | -0.0045 | improved |
| total_run_mae | 3.713 | 3.7391 | +0.0261 | regressed |
| calibration_error | 0.1004 | 0.1236 | +0.0232 | regressed |
| starter_er_mae | 1.7854 | 1.7917 | +0.0063 | regressed |
| starter_outs_mae | 3.1073 | 3.1042 | -0.0031 | improved |
The adjustment the system made in response
Status: quarantined invalid artifact. No candidate was created because a critical/error artifact finding must be corrected first. slate quarantined from new parameter proposals until critical/error artifact findings are corrected
Luna deep post-mortem (full)