Deep slate diagnosis — 20260909

• Games: 15 · hits 10 · misses 5 · winner accuracy 0.6667
• Mean total-run bias on losses: 0.4 (negative = under-projected)
• Mean team-run error on losses: 5.6 · mean Brier on losses: 0.29

Loss primary drivers

• Small edge on the losing side: 1
• Several errors of similar size: 1
• Run environment far lower than projected (over-dispersed): 2
• Starting-pitcher projection was wrong: 1

High-level causal issue register

[CRITICAL] Critical canonical artifact integrity failure

• Domain: artifact_integrity · affected: 20260909_WAS@SD
• Evidence: structural audit totals: {'critical': 1, 'error': 0, 'warning': 0, 'info': 114}
• Action: Exclude this slate from parameter learning until the artifact-generation step that wrote the broken file is fixed.
• Do not infer: A malformed artifact is a file bug, not proof the prediction model was wrong; do not blame accuracy changes on it.

[CRITICAL] Published probability differs materially without a recorded transformation

• Domain: probability_lineage · affected: PIT@CHW, TB@ATL
• Evidence: 2 game(s) have an unexplained ≥2pp calibrated-to-published delta
• Action: Block any promotion until the missing probability transformation is recorded and reconciled.
• Do not infer: Do not assume the probability governor was bypassed until you have traced the pick-side math.

[HIGH] Validated run models explain very little holdout variance

• Domain: model_validation · affected: ARI@KC, CHC@MIL, CIN@LAD, CLE@BAL, COL@NYY, HOU@PHI, LAA@BOS, MIN@DET, NYM@MIA, PIT@CHW, STL@SF, TB@ATL, TEX@SEA, TOR@OAK, WAS@SD
• Evidence: home/away/total R² = 0.0118/0.0210/0.0207
• Action: Treat exact run scores as low-confidence estimates, and look for stronger features or a better target before tuning the score tails.
• Do not infer: A small positive R² only means the model beats guessing the average; it does not make the exact scores reliable.

[INFO] Every settled game received structural QA

• Domain: coverage · affected: ARI@KC, CHC@MIL, CIN@LAD, CLE@BAL, COL@NYY, HOU@PHI, LAA@BOS, MIN@DET, NYM@MIA, PIT@CHW, STL@SF, TB@ATL, TEX@SEA, TOR@OAK, WAS@SD
• Evidence: 15/15 games structurally audited
• Action: Keep the per-game audit receipt in every report.
• Do not infer: Full audit coverage does not cancel the findings raised inside the audited games.

[INFO] Simulation produced no winner-direction intervention on this slate

• Domain: simulation · affected: ARI@KC, CHC@MIL, CIN@LAD, CLE@BAL, COL@NYY, HOU@PHI, LAA@BOS, MIN@DET, NYM@MIA, PIT@CHW, STL@SF, TB@ATL, TEX@SEA, TOR@OAK, WAS@SD
• Evidence: flip candidates 0; published flips 0
• Action: Measure the simulation's contribution across rolling slates, separating score realization from winner flipping.
• Do not infer: Zero winner flips is not a defect and does not mean the simulation did no score-reconciliation work.

Cross-slate trends (persistent-pattern memory)

• Window: 20260830, 20260831, 20260901, 20260902, 20260903, 20260904, 20260905, 20260906, 20260907, 20260908, 20260909 (10 prior slates loaded)
• Highest-confidence pick loss streak: 0 slate(s)
• Actionable pattern detected: True

Escalated persistent patterns

• simulation_inert_on_slate — 11 consecutive slates (11 in window)
• Target lever(s): simulation.allow_score
• Direction: confirm the simulation module is running, not silently no-oping
• Guardrail: simulation_inert_on_slate has persisted 11 consecutive slates; this is no longer single-slate noise. Treat it as an actionable pattern and raise the code/diagnostic action named; do not force a parameter move.
• weak_run_model_signal — 11 consecutive slates (11 in window)
• Target lever(s): training.xgb.n_estimators, training.xgb.max_depth, training.xgb.learning_rate
• Direction: investigate features/target via isolated training backtest (manual)
• Guardrail: weak_run_model_signal has persisted 11 consecutive slates; this is no longer single-slate noise. Treat it as an actionable pattern and raise the code/diagnostic action named; do not force a parameter move.

Metric direction across the window

• winner_accuracy: improving · series [0.5, 0.6667, 0.4667, 0.3333, 0.5556, 0.625, 0.5333, 0.6, 0.5455, 0.7333, 0.6667]
• weighted_calibration_error: improving · series [0.0693, 0.1661, 0.2476, 0.21, 0.1474, 0.1316, 0.1, 0.0937, 0.2611, 0.1825, 0.1069]
• margin_bias: regressing · series [-2.786, -3.5, -2.533, -3.067, -1.778, -0.688, -1.467, -1.467, -0.182, -0.867, -2.133]
• mean_team_run_error_on_losses: improving · series [7.0, 8.25, 7.125, 6.4, 6.75, 6.0, 5.286, 5.333, 3.8, 6.75, 5.6]

Pipeline stage quality

• historical_ml: winner 10/15 (0.6667), team-run MAE 3.1009, margin MAE 3.1847
• variance_applied: winner 9/15 (0.6), team-run MAE 3.1333, margin MAE 3.2
• enhanced_prediction: winner 10/15 (0.6667), team-run MAE 3.2667, margin MAE 3.3333
• embedded_savant_run_environment: winner 10/15 (0.6667), team-run MAE 3.2667, margin MAE 3.2
• fundamental_edge_target_review: winner 10/15 (0.6667), team-run MAE 3.2667, margin MAE 3.2
• capped_target: winner 10/15 (0.6667), team-run MAE 3.2667, margin MAE 3.2
• simulation_result: winner 10/15 (0.6667), team-run MAE 3.2667, margin MAE 3.2
• published_final: winner 10/15 (0.6667), team-run MAE 3.2667, margin MAE 3.2

Probability lineage

• ARI@KC: raw 0.5823 → calibrated 0.5474 → published 0.53; score coherence followed by matchup-recency guardrail
• CHC@MIL: raw 0.5411 → calibrated 0.5339 → published 0.5245; score coherence followed by matchup-recency guardrail
• CIN@LAD: raw 0.691 → calibrated 0.6629 → published 0.58; score coherence followed by matchup-recency guardrail
• CLE@BAL: raw 0.5282 → calibrated 0.5802 → published 0.558; score coherence followed by matchup-recency guardrail
• COL@NYY: raw 0.7754 → calibrated 0.7839 → published 0.7839; published probability matches the calibrated/governed value
• HOU@PHI: raw 0.5438 → calibrated 0.5273 → published 0.5197; score coherence followed by matchup-recency guardrail
• LAA@BOS: raw 0.5671 → calibrated 0.5397 → published 0.5287; score coherence followed by matchup-recency guardrail
• MIN@DET: raw 0.5546 → calibrated 0.5331 → published 0.5239; score coherence followed by matchup-recency guardrail
• NYM@MIA: raw 0.5464 → calibrated 0.6118 → published 0.5416; score coherence followed by matchup-recency guardrail
• PIT@CHW: raw 0.6887 → calibrated 0.6496 → published 0.5984; no recorded transformation explains the publication delta
• STL@SF: raw 0.5472 → calibrated 0.5291 → published 0.521; score coherence followed by matchup-recency guardrail
• TB@ATL: raw 0.6188 → calibrated 0.5996 → published 0.5772; no recorded transformation explains the publication delta
• TEX@SEA: raw 0.5169 → calibrated 0.6398 → published 0.5487; score coherence followed by matchup-recency guardrail
• TOR@OAK: raw 0.5523 → calibrated 0.55 → published 0.5362; score coherence followed by matchup-recency guardrail
• WAS@SD: raw 0.5641 → calibrated 0.5546 → published 0.5245; score coherence followed by matchup-recency guardrail

Every game, in detail

ARI@KC — ✅ hit

• Predicted KC 5-7 (conf +0.06, lean) · actual KC 2-5
• Root cause: right_winner_wrong_score
• Errors: winner correct · Brier 0.2209 · team-run err 5.0 · total bias +5
• Contributing factors:
• total runs over-projected by 5 (12 vs 7)
• ARI offense over-projected by 3

CHC@MIL — ✅ hit

• Predicted MIL 5-6 (conf +0.05, lean) · actual MIL 6-8
• Root cause: solid_correct_call
• Errors: winner correct · Brier 0.226118 · team-run err 3.0 · total bias -3
• Contributing factors:
• total runs under-projected by 3 (11 vs 14)
• Logan Henderson (MIL) allowed 3 fewer ER than projected (0 vs 3)

CIN@LAD — ✅ hit

• Predicted LAD 5-8 (conf +0.16, moderate) · actual LAD 1-14
• Root cause: right_winner_wrong_score
• Errors: winner correct · Brier 0.1764 · team-run err 10.0 · total bias -2
• Contributing factors:
• CIN offense over-projected by 4
• LAD offense under-projected by 6
• Rhett Lowder (CIN) allowed 3 more ER than projected (7 vs 4)

CLE@BAL — ❌ miss

• Predicted CLE 4-3 (conf -0.12, moderate) · actual BAL 5-9
• Root cause: Starting-pitcher projection was wrong
• Secondary cause: Run environment far higher than projected (under-dispersed)
• Errors: winner WRONG · Brier 0.311325 · team-run err 7.0 · total bias -7
• Contributing factors:
• total runs under-projected by 7 (7 vs 14)
• BAL offense under-projected by 6
• Foster Griffin (CLE) allowed 5 more ER than projected (8 vs 3)
• Shane Baz (BAL) allowed 4 more ER than projected (5 vs 1)
• committed to CLE at 0.12 confidence and it lost

COL@NYY — ✅ hit

• Predicted NYY 4-9 (conf +0.57, strong) · actual NYY 1-6
• Root cause: right_winner_wrong_score
• Errors: winner correct · Brier 0.046714 · team-run err 6.0 · total bias +6
• Contributing factors:
• total runs over-projected by 6 (13 vs 7)
• COL offense over-projected by 3
• NYY offense over-projected by 3
• Will Warren (NYY) allowed 3 fewer ER than projected (1 vs 4)

HOU@PHI — ✅ hit

• Predicted PHI 3-4 (conf +0.04, toss_up) · actual PHI 7-11
• Root cause: toss_up_went_our_way
• Errors: winner correct · Brier 0.230691 · team-run err 11.0 · total bias -11
• Contributing factors:
• total runs under-projected by 11 (7 vs 18)
• HOU offense under-projected by 4
• PHI offense under-projected by 7
• Cristopher Sánchez (PHI) allowed 4 more ER than projected (6 vs 2)

LAA@BOS — ❌ miss

• Predicted BOS 4-5 (conf +0.06, lean) · actual LAA 6-4
• Root cause: Several errors of similar size
• Errors: winner WRONG · Brier 0.279527 · team-run err 3.0 · total bias -1
• Contributing factors:
• Jake Bennett (BOS) allowed 3 more ER than projected (6 vs 3)

MIN@DET — ✅ hit

• Predicted DET 4-5 (conf +0.05, lean) · actual DET 2-7
• Root cause: solid_correct_call
• Errors: winner correct · Brier 0.226629 · team-run err 4.0 · total bias +0

NYM@MIA — ✅ hit

• Predicted NYM 7-4 (conf -0.08, lean) · actual NYM 15-14
• Root cause: right_winner_wrong_score
• Errors: winner correct · Brier 0.210168 · team-run err 18.0 · total bias -18
• Contributing factors:
• total runs under-projected by 18 (11 vs 29)
• NYM offense under-projected by 8
• MIA offense under-projected by 10
• Janson Junk (MIA) allowed 5 more ER than projected (8 vs 3)

PIT@CHW — ✅ hit

• Predicted PIT 5-4 (conf -0.20, moderate) · actual PIT 4-2
• Root cause: strong_correct_call
• Errors: winner correct · Brier 0.161277 · team-run err 3.0 · total bias +3
• Contributing factors:
• total runs over-projected by 3 (9 vs 6)
• Davis Martin (CHW) allowed 3 fewer ER than projected (1 vs 4)

STL@SF — ❌ miss

• Predicted STL 5-4 (conf -0.04, lean) · actual SF 6-7
• Root cause: Small edge on the losing side
• Errors: winner WRONG · Brier 0.27145 · team-run err 4.0 · total bias -4
• Contributing factors:
• total runs under-projected by 4 (9 vs 13)
• SF offense under-projected by 3
• Blade Tidwell (SF) allowed 3 more ER than projected (4 vs 1)

TB@ATL — ✅ hit

• Predicted TB 4-3 (conf -0.15, moderate) · actual TB 7-2
• Root cause: solid_correct_call
• Errors: winner correct · Brier 0.178782 · team-run err 4.0 · total bias -2
• Contributing factors:
• TB offense under-projected by 3
• Reynaldo López (ATL) allowed 3 more ER than projected (6 vs 3)

TEX@SEA — ❌ miss

• Predicted TEX 6-4 (conf -0.10, lean) · actual SEA 2-3
• Root cause: Run environment far lower than projected (over-dispersed)
• Errors: winner WRONG · Brier 0.301097 · team-run err 5.0 · total bias +5
• Contributing factors:
• total runs over-projected by 5 (10 vs 5)
• TEX offense over-projected by 4

TOR@OAK — ❌ miss

• Predicted TOR 6-5 (conf -0.07, lean) · actual OAK 0-2
• Root cause: Run environment far lower than projected (over-dispersed)
• Errors: winner WRONG · Brier 0.287464 · team-run err 9.0 · total bias +9
• Contributing factors:
• total runs over-projected by 9 (11 vs 2)
• TOR offense over-projected by 6
• OAK offense over-projected by 3
• Brady Basso (OAK) allowed 3 fewer ER than projected (0 vs 3)

WAS@SD — ✅ hit

• Predicted SD 4-5 (conf +0.05, lean) · actual SD 2-9
• Root cause: right_winner_wrong_score
• Errors: winner correct · Brier 0.226103 · team-run err 6.0 · total bias -2
• Contributing factors:
• SD offense under-projected by 4

Cross-game lever tension

Each candidate fix, with the games it would help and the games it would 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
• Net: 1 helped vs 2 hurt
• Would help:
• CLE@BAL (miss): lost a 0.12-confidence pick; wider tails lower overconfidence
• Would hurt / risk:
• COL@NYY (hit): strong correct call at 0.57; widening softens a good edge
• PIT@CHW (hit): strong correct call at 0.20; widening softens a good edge

player.pitcher.v4 form weights / player.pitcher_prior_games

• Targets: starter_er_mae, starter_outs_mae
• Net: 11 helped vs 2 hurt
• Would help:
• CHC@MIL (hit): starter ER off by 3
• CIN@LAD (hit): starter ER off by 3
• CLE@BAL (miss): starter ER off by 5
• COL@NYY (hit): starter ER off by 3
• HOU@PHI (hit): starter ER off by 4
• LAA@BOS (miss): starter ER off by 3
• NYM@MIA (hit): starter ER off by 5
• PIT@CHW (hit): starter ER off by 3
• STL@SF (miss): starter ER off by 3
• TB@ATL (hit): starter ER off by 3
• TOR@OAK (miss): starter ER off by 3
• Would hurt / risk:
• MIN@DET (hit): starter projection was accurate (ER off 1); reweight risks disturbing it
• WAS@SD (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
• Net: 12 helped vs 0 hurt
• Would help:
• ARI@KC (hit): team runs off by 5
• CIN@LAD (hit): team runs off by 10
• CLE@BAL (miss): team runs off by 7
• COL@NYY (hit): team runs off by 6
• HOU@PHI (hit): team runs off by 11
• MIN@DET (hit): team runs off by 4
• NYM@MIA (hit): team runs off by 18
• STL@SF (miss): team runs off by 4
• TB@ATL (hit): team runs off by 4
• TEX@SEA (miss): team runs off by 5
• TOR@OAK (miss): team runs off by 9
• WAS@SD (hit): team runs off by 6