Deep slate diagnosis — 20260911

• Games: 15 · hits 10 · misses 5 · winner accuracy 0.6667
• Mean total-run bias on losses: 1.4 (negative = under-projected)
• Mean team-run error on losses: 5.0 · mean Brier on losses: 0.303

Loss primary drivers

• Run environment far lower than projected (over-dispersed): 1
• Starting-pitcher projection was wrong: 1
• Genuine toss-up that flipped: 3

High-level causal issue register

[CRITICAL] Published probability differs materially without a recorded transformation

• Domain: probability_lineage · affected: CHW@STL, COL@DET, LAA@WAS
• Evidence: 3 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] Published confidence buckets were poorly aligned on this slate

• Domain: calibration · affected: BAL@TOR, CHW@STL, CIN@MIL, CLE@MIN, COL@DET, HOU@TB, KC@BOS, LAA@WAS, LAD@MIA, NYM@NYY, PHI@ATL, PIT@CHC, SD@SF, SEA@OAK, TEX@ARI
• Evidence: weighted absolute calibration error 0.2091 across 15 games
• Action: Watch the same confidence buckets over the full rolling window before changing any calibration weight.
• Do not infer: A single slate has far too few games to establish a stable calibration curve.

[HIGH] Validated run models explain very little holdout variance

• Domain: model_validation · affected: BAL@TOR, CHW@STL, CIN@MIL, CLE@MIN, COL@DET, HOU@TB, KC@BOS, LAA@WAS, LAD@MIA, NYM@NYY, PHI@ATL, PIT@CHC, SD@SF, SEA@OAK, TEX@ARI
• 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: BAL@TOR, CHW@STL, CIN@MIL, CLE@MIN, COL@DET, HOU@TB, KC@BOS, LAA@WAS, LAD@MIA, NYM@NYY, PHI@ATL, PIT@CHC, SD@SF, SEA@OAK, TEX@ARI
• 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: BAL@TOR, CHW@STL, CIN@MIL, CLE@MIN, COL@DET, HOU@TB, KC@BOS, LAA@WAS, LAD@MIA, NYM@NYY, PHI@ATL, PIT@CHC, SD@SF, SEA@OAK, TEX@ARI
• 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: 20260901, 20260902, 20260903, 20260904, 20260905, 20260906, 20260907, 20260908, 20260909, 20260910, 20260911 (10 prior slates loaded)
• Highest-confidence pick loss streak: 1 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 [0.4667, 0.3333, 0.5556, 0.625, 0.5333, 0.6, 0.5455, 0.7333, 0.6667, 0.6, 0.6667]
• weighted_calibration_error: regressing · series [0.2476, 0.21, 0.1474, 0.1316, 0.1, 0.0937, 0.2611, 0.1825, 0.1069, 0.1171, 0.2091]
• margin_bias: regressing · series [-2.533, -3.067, -1.778, -0.688, -1.467, -1.467, -0.182, -0.867, -2.133, -1.0, -2.933]
• mean_team_run_error_on_losses: improving · series [7.125, 6.4, 6.75, 6.0, 5.286, 5.333, 3.8, 6.75, 5.6, 6.5, 5.0]

Pipeline stage quality

• historical_ml: winner 9/15 (0.6), team-run MAE 2.4049, margin MAE 4.1176
• variance_applied: winner 8/15 (0.5333), team-run MAE 2.4, margin MAE 4.1333
• enhanced_prediction: winner 10/15 (0.6667), team-run MAE 2.4333, margin MAE 3.9333
• embedded_savant_run_environment: winner 10/15 (0.6667), team-run MAE 2.4, margin MAE 3.8667
• fundamental_edge_target_review: winner 10/15 (0.6667), team-run MAE 2.4, margin MAE 3.8667
• capped_target: winner 10/15 (0.6667), team-run MAE 2.4, margin MAE 3.8667
• simulation_result: winner 10/15 (0.6667), team-run MAE 2.4, margin MAE 3.8667
• published_final: winner 10/15 (0.6667), team-run MAE 2.4, margin MAE 3.8667

Probability lineage

• BAL@TOR: raw 0.5071 → calibrated 0.508 → published 0.505; published probability matches the calibrated/governed value
• CHW@STL: raw 0.6331 → calibrated 0.623 → published 0.5984; no recorded transformation explains the publication delta
• CIN@MIL: raw 0.609 → calibrated 0.6213 → published 0.54; score coherence followed by matchup-recency guardrail
• CLE@MIN: raw 0.5 → calibrated 0.5073 → published 0.5053; published probability matches the calibrated/governed value
• COL@DET: raw 0.6252 → calibrated 0.6071 → published 0.5412; no recorded transformation explains the publication delta
• HOU@TB: raw 0.5 → calibrated 0.508 → published 0.505; published probability matches the calibrated/governed value
• KC@BOS: raw 0.655 → calibrated 0.6529 → published 0.6529; published probability matches the calibrated/governed value
• LAA@WAS: raw 0.6202 → calibrated 0.6177 → published 0.5757; no recorded transformation explains the publication delta
• LAD@MIA: raw 0.6625 → calibrated 0.6905 → published 0.58; score coherence followed by matchup-recency guardrail
• NYM@NYY: raw 0.5664 → calibrated 0.6162 → published 0.53; score coherence followed by matchup-recency guardrail
• PHI@ATL: raw 0.563 → calibrated 0.5376 → published 0.5272; score coherence followed by matchup-recency guardrail
• PIT@CHC: raw 0.5901 → calibrated 0.5517 → published 0.53; score coherence followed by matchup-recency guardrail
• SD@SF: raw 0.5 → calibrated 0.508 → published 0.505; published probability matches the calibrated/governed value
• SEA@OAK: raw 0.5 → calibrated 0.508 → published 0.5058; published probability matches the calibrated/governed value
• TEX@ARI: raw 0.5798 → calibrated 0.58 → published 0.568; score coherence followed by matchup-recency guardrail

Every game, in detail

BAL@TOR — ❌ miss

• Predicted TOR 4-5 (conf +0.01, toss_up) · actual BAL 7-4
• Root cause: Genuine toss-up that flipped
• Secondary cause: Small edge on the losing side
• Errors: winner WRONG · Brier 0.255025 · team-run err 4.0 · total bias -2
• Contributing factors:
• BAL offense under-projected by 3
• Max Scherzer (TOR) allowed 3 more ER than projected (6 vs 3)

CHW@STL — ✅ hit

• Predicted STL 4-6 (conf +0.20, moderate) · actual STL 3-7
• Root cause: strong_correct_call
• Errors: winner correct · Brier 0.161312 · team-run err 2.0 · total bias +0

CIN@MIL — ✅ hit

• Predicted MIL 6-8 (conf +0.08, lean) · actual MIL 0-20
• Root cause: right_winner_wrong_score
• Errors: winner correct · Brier 0.2116 · team-run err 18.0 · total bias -6
• Contributing factors:
• total runs under-projected by 6 (14 vs 20)
• CIN offense over-projected by 6
• MIL offense under-projected by 12
• Andrew Abbott (CIN) allowed 3 more ER than projected (8 vs 5)
• Dustin May (MIL) allowed 4 fewer ER than projected (0 vs 4)

CLE@MIN — ❌ miss

• Predicted MIN 4-5 (conf +0.01, toss_up) · actual CLE 5-2
• Root cause: Genuine toss-up that flipped
• Secondary cause: Small edge on the losing side
• Errors: winner WRONG · Brier 0.255291 · team-run err 4.0 · total bias +2
• Contributing factors:
• MIN offense over-projected by 3
• Taj Bradley (MIN) allowed 3 fewer ER than projected (0 vs 3)

COL@DET — ✅ hit

• Predicted DET 4-5 (conf +0.08, lean) · actual DET 2-6
• Root cause: solid_correct_call
• Errors: winner correct · Brier 0.210526 · team-run err 3.0 · total bias +1

HOU@TB — ✅ hit

• Predicted TB 3-5 (conf +0.01, toss_up) · actual TB 1-3
• Root cause: toss_up_went_our_way
• Errors: winner correct · Brier 0.245025 · team-run err 4.0 · total bias +4
• Contributing factors:
• total runs over-projected by 4 (8 vs 4)

KC@BOS — ❌ miss

• Predicted BOS 4-6 (conf +0.31, strong) · actual KC 3-2
• Root cause: Run environment far lower than projected (over-dispersed)
• Secondary cause: Backed the wrong side with real conviction
• Errors: winner WRONG · Brier 0.42625 · team-run err 5.0 · total bias +5
• Contributing factors:
• total runs over-projected by 5 (10 vs 5)
• BOS offense over-projected by 4
• committed to BOS at 0.31 confidence and it lost

LAA@WAS — ✅ hit

• Predicted WAS 4-5 (conf +0.15, moderate) · actual WAS 3-4
• Root cause: strong_correct_call
• Errors: winner correct · Brier 0.180053 · team-run err 2.0 · total bias +2

LAD@MIA — ✅ hit

• Predicted LAD 6-4 (conf -0.16, moderate) · actual LAD 6-2
• Root cause: strong_correct_call
• Errors: winner correct · Brier 0.1764 · team-run err 2.0 · total bias +2

NYM@NYY — ✅ hit

• Predicted NYY 4-6 (conf +0.06, lean) · actual NYY 4-6
• Root cause: solid_correct_call
• Errors: winner correct · Brier 0.2209 · team-run err 0.0 · total bias +0
• Contributing factors:
• Nolan McLean (NYM) allowed 3 more ER than projected (5 vs 2)

PHI@ATL — ✅ hit

• Predicted ATL 3-4 (conf +0.05, lean) · actual ATL 5-6
• Root cause: solid_correct_call
• Errors: winner correct · Brier 0.223581 · team-run err 4.0 · total bias -4
• Contributing factors:
• total runs under-projected by 4 (7 vs 11)

PIT@CHC — ✅ hit

• Predicted CHC 4-5 (conf +0.06, lean) · actual CHC 2-12
• Root cause: right_winner_wrong_score
• Errors: winner correct · Brier 0.2209 · team-run err 9.0 · total bias -5
• Contributing factors:
• total runs under-projected by 5 (9 vs 14)
• CHC offense under-projected by 7

SD@SF — ✅ hit

• Predicted SD 5-4 (conf -0.01, toss_up) · actual SD 7-5
• Root cause: toss_up_went_our_way
• Errors: winner correct · Brier 0.245025 · team-run err 3.0 · total bias -3
• Contributing factors:
• total runs under-projected by 3 (9 vs 12)
• Anthony Molina (SF) allowed 3 more ER than projected (7 vs 4)

SEA@OAK — ❌ miss

• Predicted SEA 7-6 (conf -0.01, toss_up) · actual OAK 5-6
• Root cause: Genuine toss-up that flipped
• Errors: winner WRONG · Brier 0.255838 · team-run err 2.0 · total bias +2

TEX@ARI — ❌ miss

• Predicted TEX 6-4 (conf -0.14, moderate) · actual ARI 1-9
• Root cause: Starting-pitcher projection was wrong
• Secondary cause: Backed the wrong side with real conviction
• Errors: winner WRONG · Brier 0.322638 · team-run err 10.0 · total bias +0
• Contributing factors:
• TEX offense over-projected by 5
• ARI offense under-projected by 5
• Merrill Kelly (ARI) allowed 4 fewer ER than projected (1 vs 5)
• committed to TEX at 0.14 confidence and it lost

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: 2 helped vs 1 hurt
• Would help:
• KC@BOS (miss): lost a 0.31-confidence pick; wider tails lower overconfidence
• TEX@ARI (miss): lost a 0.14-confidence pick; wider tails lower overconfidence
• Would hurt / risk:
• CHW@STL (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: 6 helped vs 5 hurt
• Would help:
• BAL@TOR (miss): starter ER off by 3
• CIN@MIL (hit): starter ER off by 4
• CLE@MIN (miss): starter ER off by 3
• NYM@NYY (hit): starter ER off by 3
• SD@SF (hit): starter ER off by 3
• TEX@ARI (miss): starter ER off by 4
• Would hurt / risk:
• CHW@STL (hit): starter projection was accurate (ER off 1); reweight risks disturbing it
• LAA@WAS (hit): starter projection was accurate (ER off 1); reweight risks disturbing it
• LAD@MIA (hit): starter projection was accurate (ER off 1); reweight risks disturbing it
• PIT@CHC (hit): starter projection was accurate (ER off 1); reweight risks disturbing it
• SEA@OAK (miss): 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: 8 helped vs 5 hurt
• Would help:
• BAL@TOR (miss): team runs off by 4
• CIN@MIL (hit): team runs off by 18
• CLE@MIN (miss): team runs off by 4
• HOU@TB (hit): team runs off by 4
• KC@BOS (miss): team runs off by 5
• PHI@ATL (hit): team runs off by 4
• PIT@CHC (hit): team runs off by 9
• TEX@ARI (miss): team runs off by 10
• Would hurt / risk:
• CHW@STL (hit): team runs already close (off 2); reweight risks disturbing it
• LAA@WAS (hit): team runs already close (off 2); reweight risks disturbing it
• LAD@MIA (hit): team runs already close (off 2); reweight risks disturbing it
• NYM@NYY (hit): team runs already close (off 0); reweight risks disturbing it
• SEA@OAK (miss): team runs already close (off 2); reweight risks disturbing it