Deep slate report 20260915

Which reports actually fed the diagnosis. A source shown as missing was not silently assumed.

High-level causal issue register

Source	Status	Evidence supplied	Path	Bytes
prediction review	loaded	official outcomes and scoring errors	/app/output/reports/prediction_review_20260915.json	379319
forensic deep audit	loaded	probability, provenance, starter and decision contracts	/app/output/reports/PREDICTION_SYSTEM_DEEP_AUDIT_20260915.json	241007
continuous learning	loaded	pipeline-stage quality, calibration, probability lineage, score distribution, and official starter comparisons	/app/output/learning/continuous_learning_daily_20260915.json	2583429
detailed report	loaded	full game and simulation rendering	/app/output/reports/detailed_20260915.html	354134
ultra detailed player audit	loaded	player-level prediction trace	/app/output/reports/20260915_ultra_detailed_player_audit.html	33583352
structural audit	loaded	all invariant findings	/app/output/reports/audit_report_20260915.html	1881324
integrity revision audit	missing	post-Luna remediation proof for immutable revision artifacts	/app/output/reports/audit_report_20260915_integrity_revision.html	n/a

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. Each row pairs the action to take (Do this) with the wrong conclusion to avoid (Don't conclude).

Cross-slate trends (persistent-pattern memory)

Escalated persistent patterns

A finding that has repeated at or beyond the escalation streak is no longer single-slate noise. It carries its target registered lever and an action-oriented guardrail.

Finding streaks across the window

Metric direction across the window

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.

Probability lineage and calibration

All columns are pick-side probabilities, so away favorites are not accidentally compared using home-side semantics.

Score distribution and margin compression

Narrow predicted margins next to wide actual margins is the signature the score-variance tail is too tight.

Per-game evidence coverage

Every game, in depth (15)

Severity	Domain	Finding and evidence	Affected	Action & guardrail
HIGH	artifact_integrity	Canonical artifact warning requires reconciliation structural audit totals: {'critical': 0, 'error': 0, 'warning': 1, 'info': 114}	slate-wide	Do thisFix the flagged stat or presentation contract before treating this slate as clean evidence. Don't concludeA warning flags a display or contract issue; it does not prove the model picked the wrong side.
HIGH	calibration	Published confidence buckets were poorly aligned on this slate weighted absolute calibration error 0.2795 across 15 games	ATL@CHC, BAL@NYM, BOS@TEX, CHW@CLE, DET@TOR, KC@HOU, LAD@CIN, MIA@ARI, MIL@PIT, NYY@MIN, OAK@TB, PHI@WAS, SD@COL, SEA@LAA, SF@STL	Do thisWatch the same confidence buckets over the full rolling window before changing any calibration weight. Don't concludeA single slate has far too few games to establish a stable calibration curve.
HIGH	model_validation	Validated run models explain very little holdout variance home/away/total R² = 0.0045/0.0275/0.0075	ATL@CHC, BAL@NYM, BOS@TEX, CHW@CLE, DET@TOR, KC@HOU, LAD@CIN, MIA@ARI, MIL@PIT, NYY@MIN, OAK@TB, PHI@WAS, SD@COL, SEA@LAA, SF@STL	Do thisTreat exact run scores as low-confidence estimates, and look for stronger features or a better target before tuning the score tails. Don't concludeA small positive R² only means the model beats guessing the average; it does not make the exact scores reliable.
INFO	coverage	Every settled game received structural QA 15/15 games structurally audited	ATL@CHC, BAL@NYM, BOS@TEX, CHW@CLE, DET@TOR, KC@HOU, LAD@CIN, MIA@ARI, MIL@PIT, NYY@MIN, OAK@TB, PHI@WAS, SD@COL, SEA@LAA, SF@STL	Do thisKeep the per-game audit receipt in every report. Don't concludeFull audit coverage does not cancel the findings raised inside the audited games.
INFO	probability_lineage	Material calibrated-to-published movements have recorded lineage 10 material movement(s), all explained by named score-coherence and/or recency transformations	BAL@NYM, BOS@TEX, DET@TOR, LAD@CIN, MIL@PIT, NYY@MIN, PHI@WAS, SD@COL, SEA@LAA, SF@STL	Do thisKeep showing every probability layer and the named transformation between them. Don't concludeA published probability above the calibrated one is not a governor bypass when a recorded transformation explains it.
INFO	simulation	Simulation produced no winner-direction intervention on this slate flip candidates 0; published flips 0	ATL@CHC, BAL@NYM, BOS@TEX, CHW@CLE, DET@TOR, KC@HOU, LAD@CIN, MIA@ARI, MIL@PIT, NYY@MIN, OAK@TB, PHI@WAS, SD@COL, SEA@LAA, SF@STL	Do thisMeasure the simulation's contribution across rolling slates, separating score realization from winner flipping. Don't concludeZero winner flips is not a defect and does not mean the simulation did no score-reconciliation work.

Finding	Persistence	Target lever(s)	Direction	Guardrail
simulation_inert_on_slate	11 consecutive · 11 in window	simulation.allow_score	confirm the simulation module is running, not silently no-oping	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 · 11 in window	training.xgb.n_estimators, training.xgb.max_depth, training.xgb.learning_rate	investigate features/target via isolated training backtest (manual)	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.

Finding	Consecutive slates	Incidence	Present now	Fix type
high_raw_probability_cluster	0	1/11	no	param
pipeline_winner_instability	0	2/11	no	code/diagnostic
score_margin_compression	0	1/11	no	param
simulation_inert_on_slate	11	11/11	yes	code/diagnostic
single_slate_calibration_instability	2	6/11	yes	param
weak_run_model_signal	11	11/11	yes	code/diagnostic

Metric	Direction	Series (old → new)
winner_accuracy	regressing	[0.5333, 0.6, 0.5455, 0.7333, 0.6667, 0.6, 0.6667, 0.5333, 0.6, 0.7, 0.4]
weighted_calibration_error	regressing	[0.1, 0.0937, 0.2611, 0.1825, 0.1069, 0.1171, 0.2091, 0.1711, 0.0827, 0.2812, 0.2795]
margin_bias	regressing	[-1.467, -1.467, -0.182, -0.867, -2.133, -1.0, -2.933, -2.667, -1.8, -0.4, -1.667]
team_run_error_on_losses	flat	[5.286, 5.333, 3.8, 6.75, 5.6, 6.5, 5.0, 6.143, 5.0, 6.333, 5.556]

Stage	Winner hits	Accuracy	Team-run MAE	Total-run MAE	Margin MAE
historical_ml	7/15	46.7%	2.379	2.958	3.61
variance_applied	7/15	46.7%	2.4	2.933	3.6
enhanced_prediction	7/15	46.7%	2.667	3.467	3.733
embedded_savant_run_environment	7/15	46.7%	2.7	3.533	3.667
fundamental_edge_target_review	6/15	40%	2.767	3.667	3.8
capped_target	6/15	40%	2.767	3.667	3.8
simulation_result	6/15	40%	2.767	3.667	3.8
published_final	6/15	40%	2.767	3.667	3.8

Transition	Δ accuracy	Δ hits	Δ team-run MAE	Δ margin MAE	Games whose winner/tie changed
historical_ml → variance_applied	0	0	0.0211	-0.0102	BAL@NYM: NYM → TIE; SD@COL: SD → TIE
variance_applied → enhanced_prediction	0	0	0.2667	0.1333	BAL@NYM: TIE → NYM; SD@COL: TIE → SD
enhanced_prediction → embedded_savant_run_environment	0	0	0.0333	-0.0666
embedded_savant_run_environment → fundamental_edge_target_review	-0.0667	-1	0.0667	0.1333	BOS@TEX: TEX → BOS
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

Game	Outcome	Raw model	Calibrated	Governor	Score coherence	Published	Explanation
ATL@CHC	miss	0.5467	0.5417	0.5417	0.5417	0.5355	no recorded transformation explains the publication delta
BAL@NYM	miss	0.6158	0.6001	0.6001	0.5889	0.53	score coherence followed by matchup-recency guardrail
BOS@TEX	miss	0.5621	0.537	0.537	0.537	0.5152	score coherence followed by matchup-recency guardrail
CHW@CLE	miss	0.5858	0.5617	0.5617	0.5617	0.5446	score coherence followed by matchup-recency guardrail
DET@TOR	hit	0.6269	0.5945	0.5945	0.5861	0.5622	score coherence followed by matchup-recency guardrail
KC@HOU	hit	0.5855	0.549	0.549	0.549	0.53	score coherence followed by matchup-recency guardrail
LAD@CIN	hit	0.7657	0.7434	0.7434	0.7326	0.58	score coherence followed by matchup-recency guardrail
MIA@ARI	miss	0.5966	0.555	0.555	0.555	0.5467	score coherence followed by matchup-recency guardrail
MIL@PIT	hit	0.6158	0.6425	0.6425	0.6425	0.56	score coherence followed by matchup-recency guardrail
NYY@MIN	hit	0.6729	0.6532	0.6532	0.6528	0.57	score coherence followed by matchup-recency guardrail
OAK@TB	hit	0.804	0.6674	0.6674	0.6674	0.6674	published probability matches the calibrated/governed value
PHI@WAS	miss	0.6037	0.568	0.568	0.568	0.5331	score coherence followed by matchup-recency guardrail
SD@COL	miss	0.6411	0.6975	0.6975	0.6975	0.58	score coherence followed by matchup-recency guardrail
SEA@LAA	miss	0.61	0.5686	0.5686	0.5686	0.53	score coherence followed by matchup-recency guardrail
SF@STL	miss	0.5996	0.5735	0.5735	0.5735	0.5531	score coherence followed by matchup-recency guardrail

Sides analyzed	Broke on pitching	Broke on offense	Both	Variance	Mean starter ERA error	Mean lineup runs vs proj
30	1	12	2	15	-1.58	-0.7

Game	Trained p(home)	Blended p(home)	Blend favoured	Published p(home)	Published side	Artifact says winner changed	Outcome
BOS@TEX	0.614	0.562	home	0.485	away	no	miss

Slate	Games	Winner acc	Mean starter ERA error	Mean lineup runs vs proj	Sides broke on pitching	Sides broke on offense
20260914	10	0.700	-0.48	-0.20	2	6
20260913	15	0.600	-0.07	0.03	2	8
20260912	15	0.533	1.35	0.57	6	7
20260911	15	0.667	0.99	0.00	4	9
20260910	5	0.600	-3.19	-1.70		8
20260909	15	0.667	3.36	1.07	8	7
20260908	15	0.733	-0.15	-0.23	7	15
20260907	11	0.545	-0.86	-0.36	5	4
20260906	15	0.600	0.74	-0.30	6	10
20260905	15	0.533	-1.17	-0.07	2	6
20260904	16	0.625	-0.23	-0.44	5	12
20260903	9	0.556	0.24	-0.83	4	6

Game	Pred margin	Actual margin	Pred − actual
ATL@CHC	1	3	-2
BAL@NYM	1	2	-1
BOS@TEX	1	2	-1
CHW@CLE	1	1	0
DET@TOR	1	9	-8
KC@HOU	1	2	-1
LAD@CIN	4	4	0
MIA@ARI	2	2	0
MIL@PIT	3	4	-1
NYY@MIN	3	7	-4
OAK@TB	4	1	3
PHI@WAS	1	3	-2
SD@COL	4	6	-2
SEA@LAA	1	1	0
SF@STL	1	7	-6

ATL@CHCMISS

Predicted: CHC 4-5 (conf +0.07, lean)Actual: ATL 6-3Brier 0.287total bias +0

Root cause: Small edge on the losing side

Evidence-backed diagnosis

published system edge 0.07 (lean); this is not the half-scale distance from 50%
team-run absolute error 4; total bias +0
pregame home-minus-away evidence: season +0.235, recent +0.376, pitcher -0.143, composite +0.088
confidence pipeline reached 0.542 before later publication/recency controls; published home probability was 0.535
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

No large single-factor errors.

Team	Starter	Projected	Actual	ER err
ATL	Martín Pérez	3 ER / 17 outs	2 ER / 18 outs	-1
CHC	Kevin Gausman	3 ER / 16 outs	2 ER / 17 outs	-1

Decision internals: run factors home×0.954 / away×1.097; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Pete Crow-Armstrong (CHC CF): 0-for-5, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Ozzie Albies (ATL 2B): 2-for-4, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Matt Shaw (CHC DH): 2-for-2, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.8]
Matt Olson (ATL 1B): 2-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.79]
Michael Harris II (ATL CF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]

Pitcher versus batter: what actually broke this game

Per-pitch matchup fields are not populated in the current feeds, so this is reconstructed from the settled box: each starter's workload-adjusted ERA, the bullpen share behind him, and the opposing lineup's real production against its projection, with the hitters who beat that projection named.

AWAY pitching

Martín Pérez (ATL) threw 18 outs, 2 ER = 3 ERA against a projected 4.76 ERA (-1.76 ERA error)

Bullpen behind him: 0 ER over 9 outs.

The HOME lineup he faced produced 3 runs on 5 hits and 0 HR, against a projected 5 runs / 8 hits (-2 runs vs projection).

Who did the damage to him:

Matt Shaw: 2-for-2, 0 HR, 0 RBI [+0.8 runs vs projection] season xwOBA 0.301, below average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Kevin Gausman (CHC) threw 17 outs, 2 ER = 3.18 ERA against a projected 5.06 ERA (-1.88 ERA error)

Bullpen behind him: 2 ER over 3 outs.

The AWAY lineup he faced produced 6 runs on 9 hits and 3 HR, against a projected 5 runs / 9 hits (1 runs vs projection).

Who did the damage to him:

Ozzie Albies: 2-for-4, 1 HR, 2 RBI [+1.4 runs vs projection] season xwOBA 0.28, weak - a good day, not a model error
Brewer Hicklen: 1-for-3, 0 HR, 0 RBI [+0.64 runs vs projection] season xwOBA 0.346, elite - real projection error
Ronald Acuña Jr.: 1-for-5, 1 HR, 2 RBI [+0.6 runs vs projection] season xwOBA 0.361, elite - real projection error

Verdict: offense: the lineup scored well below its projection

Starting pitchers & relievers: actual / projected + context (ERA is workload-aware: ER over 9 innings, so a short blow-up is not read like a quality start; ERA error in parentheses when material. exp BB and exp SO are the continuous rate expectations. The integer beside each is a simulation draw reconciled to the opposing simulated batting box, so it is not the line that ranks pitchers: on the 2026-09-07 slate the integers published 39 walks and 121 strikeouts against 51 and 130 real, while these expectations landed on 50.4 and 132.6.)

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
ATL	Martín Pérez (Starter)	18 / 17	3 / 4.76	2 / 3	5 / 7	0 / 0	4 / 1	2.34	2 / 4	3.42	venue away: 4.7 vs 1.97 ERA (worse here) · opp run factor 0.95 (strong) · 28 prior appearances this season
ATL	Didier Fuentes (Middle Reliever (High Leverage))	6 / 4	0 / 6.75 (-6.75)	0 / 1	0 / 1	0 / 0	0 / 0	0.31	2 / 1	1.59
ATL	Raisel Iglesias (Closer)	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 1	0 / 0	0 / 0	0.27	0 / 1	0.99
CHC	Kevin Gausman (Starter)	17 / 16	3.18 / 5.06	2 / 3	5 / 6	1 / 2	0 / 1	1.7	6 / 6	5.48	venue home: 3.91 vs 5.4 ERA (ok) · opp run factor 1.1 (strong) · 30 prior appearances this season
CHC	Daniel Palencia (Middle Reliever (High Leverage))	1 / 5	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.46	0 / 2	0.81
CHC	Jacob Webb (Closer)	2 / 3	27 / 0 (+27.00)	2 / 0	2 / 0	1 / 0	1 / 0	0.43	0 / 1	0.95

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Pete Crow-Armstrong (CF)	CHC	5 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	1 / 1	-1.73	0.375 elite	bad day
Ozzie Albies (2B)	ATL	4 / 4	2 / 1	0 / 0	0 / 0	1 / 0	2 / 0	2 / 1	0 / 0	0 / 1	1.4	0.28 weak	good day
Matt Shaw (DH)	CHC	2 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 1	0.8	0.301 below average	good day
Matt Olson (1B)	ATL	4 / 4	2 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	1 / 1	-0.79	0.354 elite	bad day
Michael Harris II (CF)	ATL	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.78	0.339 above average	bad day
Brewer Hicklen (RF)	ATL	3 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	2 / 1	0.64	0.346 elite	projection too low
Seiya Suzuki (RF)	CHC	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.64	0.34 elite	bad day
Ronald Acuña Jr. (DH)	ATL	5 / 4	1 / 1	0 / 0	0 / 0	1 / 0	2 / 1	1 / 1	0 / 1	1 / 1	0.6	0.361 elite	projection too low
Mike Yastrzemski (LF)	ATL	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.312 average	—
Nico Hoerner (SS)	CHC	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	-0.47	0.324 above average	—
Alex Bregman (3B)	CHC	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 0	-0.47	0.32 average	—
Mauricio Dubón (SS)	ATL	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0 / 0	0.33	0.305 average	—
Drake Baldwin (C)	ATL	5 / 4	2 / 1	0 / 0	0 / 0	1 / 1	2 / 1	2 / 1	0 / 1	1 / 1	0.14	0.363 elite	—
Austin Riley (3B)	ATL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	3 / 1	-0.14	0.3 below average	—
Tyrone Taylor (LF)	CHC	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0 / 0	-0.14	0.312 average	—
Miguel Amaya (C)	CHC	1 / 3	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	0 / 1	0	0.296 below average	—
Pedro Ramirez (2B)	CHC	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0 / 1	0	0.29 below average	—
Michael Busch (1B)	CHC	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	1 / 1	0 / 1	0	0.33 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.235	positive favors home; negative favors away
Recent form	0.376	positive favors home; negative favors away
Starting-pitcher matchup	-0.143	positive favors home; negative favors away
Composite	0.088	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.45-4.87	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.535	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	4.45-4.87	4-5	0.576	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)
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
ATL	Martín Pérez	pitcher / Starter	better	predicted_SO 3
CHC	Seiya Suzuki	batter / RF	worse	predicted_AB -2.37
ATL	Austin Riley	batter / 3B	better	predicted_H 2.14
CHC	Alex Bregman	batter / 3B	better	predicted_H 1.92
ATL	Brewer Hicklen	batter / RF	worse	predicted_SO 1.75
ATL	Michael Harris II	batter / CF	worse	predicted_H -1.67
CHC	Tyrone Taylor	batter / LF	better	predicted_RBI 1.55
CHC	Pete Crow-Armstrong	batter / CF	worse	predicted_AB -1.5

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	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_FINALIZED_MATCH_BOX	ATL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=9, sim=8; RBI: current=5, sim=4; R: current=5, sim=4; BB: current=3, sim=2; SO: current=8, sim=9).	ATL
info	CURRENT_PLAYER_MATCH_BOX	ATL current finalized hitter lines are model predictions and differ from the simulation box score: Ronald Acuña Jr.: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, BB current=1 box=0; Mauricio Dubón: H current=1 box=2, HR current=0 box=1, R current=0 box=1; Matt Olson: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Drake Baldwin: RBI current=1 box=2, SO current=1 box=0; Austin Riley: AB current=3 box=4, H current=1 box=3, RBI current=0 box=1, R current=0 box=1; Michael Harris II: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Ozzie Albies: H current=1 box=0, R current=1 box=0; Mike Yastrzemski: AB current=3 box=4, H current=1 box=0, SO current=1 box=0	ATL
info	CURRENT_FINALIZED_MATCH_BOX	CHC model-prediction batting totals differ from final simulation box score (H: current=8, sim=9; 2B: current=1, sim=5; HR: current=1, sim=0; BB: current=4, sim=1; SF: current=0, sim=1; HBP: current=0, sim=1).	CHC
info	CURRENT_PLAYER_MATCH_BOX	CHC current finalized hitter lines are model predictions and differ from the simulation box score: Tyrone Taylor: 2B current=0 box=1, RBI current=1 box=2, R current=0 box=1, SF current=0 box=1; Nico Hoerner: R current=0 box=1; Alex Bregman: H current=1 box=3, 2B current=0 box=1, BB current=1 box=0; Matt Shaw: AB current=3 box=4, 2B current=0 box=1, RBI current=0 box=1, R current=0 box=1; Pete Crow-Armstrong: 2B current=0 box=1, HR current=1 box=0, BB current=1 box=0, SO current=1 box=2; Seiya Suzuki: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Miguel Amaya: H current=0 box=1, 2B current=0 box=1, R current=1 box=0; Pedro Ramirez: AB current=3 box=4, H current=1 box=0	CHC
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

BAL@NYMMISS

Predicted: NYM 4-5 (conf +0.06, lean)Actual: BAL 7-5Brier 0.281total bias -3

Root cause: Several errors of similar size

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 -3
pregame home-minus-away evidence: season +0.206, recent +0.318, pitcher -0.093, composite +0.168
confidence pipeline reached 0.600 before later publication/recency controls; published home probability was 0.530
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs under-projected by 3 (9 vs 12)
BAL offense under-projected by 3

Team	Starter	Projected	Actual	ER err
BAL	Shane Baz	4 ER / 18 outs	3 ER / 12 outs	-1
NYM	Sean Manaea	1 ER / 17 outs	2 ER / 16 outs	1

Decision internals: run factors home×0.944 / away×1.037; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Coby Mayo (BAL DH): 2-for-5, 1 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Carson Benge (NYM RF): 1-for-5, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.26]
Christian Encarnacion-Strand (BAL 3B): 2-for-5, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]
Bo Bichette (NYM 3B): 3-for-6, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.94]
Jeremiah Jackson (BAL 2B): 1-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]

Pitcher versus batter: what actually broke this game

AWAY pitching

Shane Baz (BAL) threw 12 outs, 3 ER = 6.75 ERA against a projected 6 ERA (0.75 ERA error)

Bullpen behind him: 0 ER over 5 outs.

The HOME lineup he faced produced 5 runs on 12 hits and 1 HR, against a projected 5 runs / 8 hits (0 runs vs projection).

Who did the damage to him:

Carson Benge: 1-for-5, 1 HR, 1 RBI [+1.26 runs vs projection] season xwOBA 0.335, above average - real projection error
Bo Bichette: 3-for-6, 0 HR, 1 RBI [+0.94 runs vs projection] season xwOBA 0.327, above average - real projection error
Marcus Semien: 1-for-4, 0 HR, 1 RBI [+0.64 runs vs projection] season xwOBA 0.304, average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Sean Manaea (NYM) threw 16 outs, 2 ER = 3.38 ERA against a projected 1.59 ERA (1.79 ERA error)

Bullpen behind him: 3 ER over 14 outs.

The AWAY lineup he faced produced 6 runs on 8 hits and 4 HR, against a projected 3 runs / 8 hits (3 runs vs projection).

Who did the damage to him:

Coby Mayo: 2-for-5, 1 HR, 3 RBI [+1.4 runs vs projection] season xwOBA 0.326, above average - real projection error
Christian Encarnacion-Strand: 2-for-5, 1 HR, 1 RBI [+1.09 runs vs projection] season xwOBA 0.341, elite - real projection error
Jeremiah Jackson: 1-for-4, 1 HR, 1 RBI [+0.93 runs vs projection] season xwOBA 0.284, weak - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
BAL	Shane Baz (Starter)	12 / 18	6.75 / 6	3 / 4	7 / 2	1 / 1	2 / 3	2.09	2 / 5	6.49	venue away: 4.18 vs 3.92 ERA (ok) · opp run factor 0.94 (average) · 29 prior appearances this season
BAL	Rico Garcia (Setup Man)	2 / 3	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0.24	0 / 1	1.19
BAL	Andrew Kittredge (Closer)	3 / 0	0 / n/a	0 / 0	1 / 0	0 / 0	0 / 1	0.19	0 / 0	0.94
NYM	Sean Manaea (Starter)	16 / 17	3.38 / 1.59	2 / 1	4 / 4	2 / 1	0 / 1	1.36	6 / 5	4.24	venue home: 4.13 vs 4.83 ERA (ok) · opp run factor 1.04 (average) · 30 prior appearances this season
NYM	Daniel Duarte (Setup Man)	3 / 4	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	0.37	2 / 1	1.23
NYM	Kodai Senga (Closer)	3 / 5	9 / 5.4 (+3.60)	1 / 1	1 / 3	1 / 1	0 / 1	0.51	2 / 3	1.21
NYM	Nathan Lavender (Middle Reliever (High Leverage))	5 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.31	4 / 0	1.17
NYM	Jonathan Pintaro (Middle Reliever (High Leverage))	3 / 0	18 / n/a	2 / 2	2 / 1	1 / 1	1 / 0	0.6	2 / 1	1.17

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Coby Mayo (DH)	BAL	5 / 4	2 / 1	0 / 0	0 / 0	1 / 0	3 / 1	1 / 1	0 / 0	1 / 1	1.4	0.326 above average	projection too low
Carson Benge (RF)	NYM	5 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	2 / 1	1 / 0	2 / 1	1.26	0.335 above average	projection too low
Christian Encarnacion-Strand (3B)	BAL	5 / 4	2 / 1	0 / 1	0 / 0	1 / 0	1 / 1	2 / 0	0 / 0	1 / 1	1.09	0.341 elite	projection too low
Bo Bichette (3B)	NYM	6 / 4	3 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	0 / 1	0.94	0.327 above average	projection too low
Jeremiah Jackson (2B)	BAL	4 / 3	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	0 / 0	2 / 1	0.93	0.284 weak	good day
Leody Taveras (RF)	BAL	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	3 / 1	-0.8	0.27 weak	projection too high
Marcus Semien (2B)	NYM	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0 / 1	0.64	0.304 average	no signal
Rece Hinds (LF)	BAL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.188 weak	—
A.J. Ewing (CF)	NYM	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0.47	0.323 above average	—
Brett Baty (DH)	NYM	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.47	0.311 average	—
Francisco Alvarez (C)	NYM	5 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	2 / 1	-0.47	0.323 above average	—
Juan Soto (LF)	NYM	5 / 4	2 / 1	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	1 / 1	1 / 0	-0.46	0.413 elite	—
Pete Alonso (1B)	BAL	5 / 4	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	2 / 1	-0.33	0.393 elite	—
Gunnar Henderson (SS)	BAL	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	4 / 1	-0.33	0.32 average	—
Francisco Lindor (SS)	NYM	5 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	0.31	0.331 above average	—
Yohel Pozo (C)	BAL	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.25 weak	—
Colton Cowser (CF)	BAL	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.317 average	—
Jared Young (1B)	NYM	4 / 4	1 / 1	1 / 1	0 / 0	0 / 0	1 / 1	0 / 1	0 / 0	0 / 1	0	0.314 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.206	positive favors home; negative favors away
Recent form	0.318	positive favors home; negative favors away
Starting-pitcher matchup	-0.093	positive favors home; negative favors away
Composite	0.168	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.61-4.42	n/a	Base ML historical model prediction
variance_applied	4-4	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	3.61-4.42	4-4	-0.806	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-4	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
BAL	Rece Hinds	batter / LF	better	predicted_H 2.602
NYM	Juan Soto	batter / LF	worse	predicted_RBI 2.48
BAL	Gunnar Henderson	batter / SS	worse	predicted_SO 2.041
BAL	Jeremiah Jackson	batter / 2B	worse	predicted_AB 1.962
NYM	Jared Young	batter / 1B	worse	predicted_SO 1.34
NYM	Francisco Alvarez	batter / C	better	predicted_SO -1.16
BAL	Christian Encarnacion-Strand	batter / 3B	better	predicted_AB 1.12
NYM	Bo Bichette	batter / 3B	worse	predicted_H -1.1

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	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	BAL model-prediction batting totals differ from final simulation box score (AB: current=31, sim=39; H: current=8, sim=9; 2B: current=1, sim=0; HR: current=1, sim=4; R: current=3, sim=4; SO: current=9, sim=10).	BAL
info	CURRENT_PLAYER_MATCH_BOX	BAL current finalized hitter lines are model predictions and differ from the simulation box score: Rece Hinds: AB current=3 box=5, H current=1 box=3, HR current=0 box=2, RBI current=0 box=2; Coby Mayo: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=0 box=1; Christian Encarnacion-Strand: AB current=4 box=5, 2B current=1 box=0, HR current=0 box=1, R current=0 box=1; Pete Alonso: AB current=4 box=5, H current=1 box=2, HR current=1 box=0, RBI current=1 box=0; Jeremiah Jackson: AB current=3 box=5, H current=1 box=2, HR current=0 box=1, R current=0 box=1; Leody Taveras: AB current=3 box=4, BB current=1 box=0; Gunnar Henderson: AB current=4 box=5, H current=1 box=0, R current=1 box=0, BB current=1 box=0; Yohel Pozo: H current=1 box=0, BB current=0 box=1, SO current=1 box=0	BAL
info	CURRENT_FINALIZED_MATCH_BOX	NYM model pitcher projections differ from final simulation pitching box score (HR: current=3, sim=4).	NYM
info	CURRENT_FINALIZED_MATCH_BOX	NYM model-prediction batting totals differ from final simulation box score (AB: current=34, sim=33; H: current=8, sim=3; 2B: current=1, sim=0; RBI: current=4, sim=5; BB: current=3, sim=5).	NYM
info	CURRENT_PLAYER_MATCH_BOX	NYM current finalized hitter lines are model predictions and differ from the simulation box score: Carson Benge: AB current=4 box=3, H current=1 box=0, BB current=0 box=1; Francisco Lindor: SO current=1 box=2; Bo Bichette: H current=1 box=0, RBI current=1 box=0, SO current=1 box=0; A.J. Ewing: AB current=4 box=3, H current=1 box=0, SO current=1 box=0; Brett Baty: BB current=0 box=1, SO current=1 box=0; Jared Young: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Juan Soto: AB current=4 box=5, RBI current=1 box=3, BB current=1 box=0, SO current=0 box=2; Marcus Semien: H current=1 box=0	NYM
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

BOS@TEXMISS

Predicted: BOS 7-6 (conf -0.03, toss_up)Actual: TEX 2-4Brier 0.265total bias +7

Root cause: Run environment far lower than projected (over-dispersed)

Evidence-backed diagnosis

published system edge 0.03 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 7; total bias +7
pregame home-minus-away evidence: season -0.701, recent -0.337, pitcher +0.025, composite -0.235
confidence pipeline reached 0.537 before later publication/recency controls; published home probability was 0.485
score-volatility actions: no_change
forecast-file integrity: 1 structural defect(s) on the saved artifact; see the per-game file-integrity table (this is a file check, not why the pick missed)

Quantified contributing factors

total runs over-projected by 7 (13 vs 6)
BOS offense over-projected by 5

Team	Starter	Projected	Actual	ER err
BOS	Patrick Sandoval	6 ER / 15 outs	4 ER / 8 outs	-2
TEX	Jacob deGrom	3 ER / 17 outs	2 ER / 15 outs	-1

Decision internals: run factors home×1.089 / away×1.064; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Wilyer Abreu (BOS RF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Josh Jung (TEX 3B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Willson Contreras (BOS 1B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.07]
Justin Foscue (TEX 2B): 0-for-1, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.07]
Wyatt Langford (TEX LF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Pitcher versus batter: what actually broke this game

AWAY pitching

Patrick Sandoval (BOS) threw 8 outs, 4 ER = 13.5 ERA against a projected 10.8 ERA (2.7 ERA error) only 11 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 4 runs on 7 hits and 0 HR, against a projected 5 runs / 9 hits (-1 runs vs projection).

Who did the damage to him:

Danny Jansen: 1-for-2, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.261, weak - a good day, not a model error
Elias Díaz: 1-for-2, 0 HR, 2 RBI [+0.31 runs vs projection] season xwOBA 0.272, weak - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

HOME pitching

Jacob deGrom (TEX) threw 15 outs, 2 ER = 3.6 ERA against a projected 4.76 ERA (-1.16 ERA error)

Bullpen behind him: 0 ER over 8 outs.

The AWAY lineup he faced produced 2 runs on 5 hits and 1 HR, against a projected 4 runs / 7 hits (-2 runs vs projection).

Who did the damage to him:

Nick Sogard: 1-for-3, 1 HR, 2 RBI [+0.6 runs vs projection] season xwOBA 0.316, average - a good day, not a model error
Caleb Durbin: 1-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.292, below average - a good day, not a model error
Mickey Gasper: 1-for-2, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.371, elite - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
BOS	Patrick Sandoval (Starter)	8 / 15	13.5 / 10.8	4 / 6	4 / 6	0 / 0	2 / 0	2.77	5 / 6	5.43	venue away: 4.74 vs 4.45 ERA (ok) · opp run factor 1.09 (average) · 11 prior appearances this season
TEX	Jacob deGrom (Starter)	15 / 17	3.6 / 4.76	2 / 3	3 / 3	1 / 1	1 / 2	1.37	8 / 8	7.82	venue home: 2.46 vs 5.35 ERA (ok) · opp run factor 1.06 (average) · 28 prior appearances this season
TEX	Jacob Latz (Closer)	5 / 5	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0.47	3 / 1	1.62
TEX	Chase Silseth (Middle Reliever (High Leverage))	3 / 2	0 / 13.5 (-13.50)	0 / 1	1 / 1	0 / 1	0 / 0	0.45	3 / 1	0.81

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Wilyer Abreu (RF)	BOS	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	3 / 1	-1.11	0.352 elite	bad day
Josh Jung (3B)	TEX	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	-1.11	0.346 elite	bad day
Willson Contreras (1B)	BOS	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	2 / 1	-1.07	0.375 elite	bad day
Justin Foscue (2B)	TEX	1 / 3	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	1 / 1	-1.07	0.329 above average	bad day
Wyatt Langford (LF)	TEX	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	-0.8	0.316 average	no signal
Adley Rutschman (C)	BOS	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.78	0.335 above average	bad day
Nick Sogard (2B)	BOS	3 / 3	1 / 1	0 / 0	0 / 0	1 / 0	2 / 0	1 / 0	0 / 1	1 / 1	0.6	0.316 average	no signal
Caleb Durbin (3B)	BOS	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0.47	0.292 below average	—
Roman Anthony (LF)	BOS	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	-0.33	0.349 elite	—
Mickey Gasper (DH)	BOS	2 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	0.33	0.371 elite	—
Trevor Story (SS)	BOS	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 2	0.33	0.251 weak	—
Danny Jansen (C)	TEX	2 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 1	0.33	0.261 weak	—
Elias Díaz (DH)	TEX	2 / 3	1 / 1	1 / 0	0 / 0	0 / 0	2 / 1	1 / 0	0 / 0	1 / 1	0.31	0.272 weak	—
Ezequiel Duran (CF)	TEX	3 / 4	2 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	0 / 1	0.16	0.306 average	—
Jake Burger (1B)	TEX	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 0	1 / 1	-0.14	0.315 average	—
Corey Seager (SS)	TEX	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	2 / 0	0 / 1	0 / 1	1 / 1	0.14	0.353 elite	—
Jarren Duran (CF)	BOS	2 / 4	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	0	0.298 below average	—
Brandon Nimmo (RF)	TEX	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	2 / 1	0	0.346 elite	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.701	positive favors home; negative favors away
Recent form	-0.337	positive favors home; negative favors away
Starting-pitcher matchup	0.025	positive favors home; negative favors away
Composite	-0.235	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.75-5	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-6	0.515	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, starter_quality_favours_the_other_side, 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	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	3.75-5	4-5	0.248	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	5-6	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, starter_quality_favours_the_other_side, 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	7-6	2	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
TEX	Corey Seager	batter / SS	worse	predicted_RBI 3.36
BOS	Trevor Story	batter / SS	worse	predicted_RBI 2.49
BOS	Patrick Sandoval	pitcher / Starter	worse	predicted_H 2
BOS	Roman Anthony	batter / LF	worse	predicted_AB -1.76
TEX	Brandon Nimmo	batter / RF	worse	predicted_AB -1.64
BOS	Mickey Gasper	batter / DH	worse	predicted_H 1.59
TEX	Justin Foscue	batter / 2B	worse	predicted_AB 1.5
BOS	Wilyer Abreu	batter / RF	worse	predicted_AB -1.48

Structural audit findings

Severity	Rule	Finding	Subject
warning	PITCH_COUNT_PITCHER_MATCH	BOS pitch_count row for Patrick Sandoval differs from pitching box score: hits_allowed/H: summary=6, box=5; batters_faced/BF: summary=21, box=20.	BOS Patrick Sandoval
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	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	BOS model pitcher projections differ from final simulation pitching box score (H: current=6, sim=5).	BOS
info	CURRENT_FINALIZED_MATCH_BOX	BOS model-prediction batting totals differ from final simulation box score (H: current=7, sim=6; 2B: current=2, sim=1; 3B: current=0, sim=1; HR: current=1, sim=2; RBI: current=4, sim=7; R: current=4, sim=7; SO: current=10, sim=12; HBP: current=0, sim=1).	BOS
info	CURRENT_PLAYER_MATCH_BOX	BOS current finalized hitter lines are model predictions and differ from the simulation box score: Roman Anthony: AB current=4 box=3, H current=1 box=0, R current=0 box=1, SO current=1 box=2; Willson Contreras: AB current=4 box=3, HR current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Adley Rutschman: 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=2; Jarren Duran: H current=0 box=1, 3B current=0 box=1; Mickey Gasper: AB current=3 box=4, H current=1 box=2, HR current=0 box=1, R current=0 box=1; Wilyer Abreu: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Nick Sogard: AB current=3 box=4, H current=1 box=0, BB current=1 box=0, SO current=1 box=2; Caleb Durbin: AB current=3 box=4, RBI current=0 box=1, R current=0 box=1, SO current=1 box=2	BOS
info	CURRENT_FINALIZED_MATCH_BOX	TEX model pitcher projections differ from final simulation pitching box score (ER: current=6, sim=7).	TEX
info	CURRENT_FINALIZED_MATCH_BOX	TEX model-prediction batting totals differ from final simulation box score (AB: current=31, sim=32; H: current=9, sim=5; 2B: current=2, sim=3; HR: current=1, sim=0; RBI: current=4, sim=6; R: current=5, sim=6; BB: current=3, sim=1; SO: current=9, sim=11).	TEX
info	CURRENT_PLAYER_MATCH_BOX	TEX current finalized hitter lines are model predictions and differ from the simulation box score: Justin Foscue: AB current=3 box=4, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Jake Burger: AB current=3 box=4, RBI current=1 box=0; Corey Seager: 2B current=0 box=1, RBI current=0 box=4, BB current=1 box=0; Ezequiel Duran: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Elias Díaz: RBI current=1 box=0, R current=0 box=1; Josh Jung: BB current=1 box=0; Danny Jansen: H current=1 box=0, SO current=1 box=2; Wyatt Langford: AB current=3 box=4, H current=1 box=0, R current=1 box=0, BB current=1 box=0	TEX
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

CHW@CLEMISS

Predicted: CHW 5-4 (conf -0.09, lean)Actual: CLE 6-7Brier 0.297total bias -4

Root cause: Small edge on the losing side

Evidence-backed diagnosis

published system edge 0.09 (lean); this is not the half-scale distance from 50%
team-run absolute error 4; total bias -4
pregame home-minus-away evidence: season -0.180, recent -0.090, pitcher -0.008, composite -0.319
confidence pipeline reached 0.562 before later publication/recency controls; published home probability was 0.455
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs under-projected by 4 (9 vs 13)
CLE offense under-projected by 3

Team	Starter	Projected	Actual	ER err
CHW	Chris Murphy	1 ER / 14 outs	3 ER / 2 outs	2
CLE	Foster Griffin	4 ER / 16 outs	2 ER / 12 outs	-2

Decision internals: run factors home×1.002 / away×1.01; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Travis Bazzana (CLE DH): 2-for-4, 2 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 2]
Munetaka Murakami (CHW 1B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
José Ramírez (CLE 3B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Brayan Rocchio (CLE SS): 2-for-3, 0 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.11]
Brenton Doyle (CHW CF): 2-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]

Pitcher versus batter: what actually broke this game

AWAY pitching

No settled pitching line for this club. The forecast named Chris Murphy (CHW) at a projected 1.93 ERA, but the per-player feed carries no pitching row for him in this game, so the actual-versus-projected comparison cannot be made on this side. That is a source-coverage gap in the settled box, not a forecasting failure.

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 5 runs on 8 hits and 3 HR, against a projected 4 runs / 8 hits (1 runs vs projection).

Who did the damage to him:

Travis Bazzana: 2-for-4, 2 HR, 2 RBI [+2 runs vs projection] season xwOBA 0.305, average - a good day, not a model error
Brayan Rocchio: 2-for-3, 0 HR, 2 RBI [+1.11 runs vs projection] season xwOBA 0.293, below average - a good day, not a model error
Nathaniel Lowe: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.338, above average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Foster Griffin (CLE) threw 12 outs, 2 ER = 4.5 ERA against a projected 6.75 ERA (-2.25 ERA error)

Bullpen behind him: 1 ER over 14 outs.

The AWAY lineup he faced produced 2 runs on 6 hits and 0 HR, against a projected 3 runs / 6 hits (-1 runs vs projection).

Who did the damage to him:

Brenton Doyle: 2-for-3, 0 HR, 0 RBI [+1.09 runs vs projection] season xwOBA 0.234, weak - a good day, not a model error
Chase Meidroth: 2-for-5, 0 HR, 1 RBI [+0.45 runs vs projection] season xwOBA 0.295, below average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
CLE	Foster Griffin (Starter)	12 / 16	4.5 / 6.75	2 / 4	5 / 6	0 / 2	2 / 2	1.81	5 / 5	4.04	venue home: 3.28 vs 3.92 ERA (ok) · opp run factor 1.01 (average) · 29 prior appearances this season
CLE	Cade Smith (Closer)	5 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.32	1 / 2	1.33
CLE	Erik Sabrowski (Middle Reliever (High Leverage))	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0.43	2 / 1	1.18
CLE	Hunter Gaddis (Setup Man)	3 / 3	9 / 0 (+9.00)	1 / 0	2 / 0	0 / 0	0 / 1	0.34	2 / 1	0.8
CLE	Tim Herrin (Middle Reliever (High Leverage))	3 / 2	0 / 13.5 (-13.50)	0 / 1	0 / 2	0 / 0	1 / 1	0.41	2 / 1	0.87

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Travis Bazzana (DH)	CLE	4 / 4	2 / 1	0 / 0	0 / 0	2 / 0	2 / 1	2 / 1	0 / 1	1 / 1	2	0.305 average	no signal
Munetaka Murakami (1B)	CHW	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	4 / 1	-1.4	0.342 elite	bad day
José Ramírez (3B)	CLE	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	-1.11	0.326 above average	bad day
Brayan Rocchio (SS)	CLE	3 / 3	2 / 1	1 / 0	0 / 0	0 / 0	2 / 0	1 / 0	1 / 0	0 / 0	1.11	0.293 below average	good day
Brenton Doyle (CF)	CHW	3 / 3	2 / 1	0 / 0	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	1.09	0.234 weak	good day
Angel Genao (2B)	CLE	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.294 below average	—
Nathaniel Lowe (1B)	CLE	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	0.47	0.338 above average	—
Chase Meidroth (2B)	CHW	5 / 4	2 / 1	1 / 0	0 / 0	0 / 0	1 / 0	1 / 1	0 / 1	0 / 1	0.45	0.295 below average	—
Randal Grichuk (DH)	CHW	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	1 / 1	-0.31	0.333 above average	—
Miguel Vargas (3B)	CHW	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	3 / 1	0 / 1	1 / 1	0 / 1	-0.31	0.38 elite	—
Luisangel Acuña (SS)	CHW	1 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	-0.14	0.266 weak	—
Patrick Bailey (C)	CLE	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	-0.14	0.308 average	—
Jake Rogers (C)	CHW	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.272 weak	—
Jo Adell (RF)	CLE	4 / 4	1 / 1	0 / 0	0 / 0	1 / 1	2 / 1	1 / 1	0 / 0	0 / 1	0	0.325 above average	—
Steven Kwan (LF)	CLE	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	0 / 0	0	0.323 above average	—
Petey Halpin (CF)	CLE	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.283 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.18	positive favors home; negative favors away
Recent form	-0.09	positive favors home; negative favors away
Starting-pitcher matchup	-0.008	positive favors home; negative favors away
Composite	-0.319	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.95-3.73	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.545	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	4.95-3.73	5-4	-0.223	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
CHW	Randal Grichuk	batter / DH	worse	predicted_AB 2.5
CHW	Brenton Doyle	batter / CF	better	predicted_RBI 1.88
CLE	Brayan Rocchio	batter / SS	worse	predicted_2B 1.85
CLE	Petey Halpin	batter / CF	worse	predicted_H 1.76
CHW	Chase Meidroth	batter / 2B	worse	predicted_AB -1.72
CLE	Steven Kwan	batter / LF	boosted	predicted_RBI 1.64
CHW	Sam Antonacci	batter / LF	worse	predicted_SO 1.375
CHW	Jake Rogers	batter / C	worse	predicted_AB 1.24

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	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_FINALIZED_MATCH_BOX	CHW model-prediction batting totals differ from final simulation box score (AB: current=32, sim=35; 2B: current=2, sim=3; HR: current=1, sim=2; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=4; SO: current=9, sim=10; HBP: current=0, sim=1).	CHW
info	CURRENT_PLAYER_MATCH_BOX	CHW current finalized hitter lines are model predictions and differ from the simulation box score: Munetaka Murakami: 2B current=0 box=1, HR current=1 box=0, SO current=1 box=2; Brenton Doyle: HR current=0 box=1, RBI current=0 box=2, R current=0 box=1, BB current=0 box=1; Sam Antonacci: 2B current=0 box=1, R current=1 box=0, SO current=1 box=2; Randal Grichuk: AB current=4 box=5, 2B current=1 box=0, HR current=0 box=1, R current=0 box=1; Miguel Vargas: AB current=4 box=5, H current=1 box=2, BB current=1 box=0; Tristan Peters: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0; Chase Meidroth: AB current=4 box=3, BB current=1 box=2; Jake Rogers: AB current=3 box=4, SO current=1 box=2	CHW
info	CURRENT_FINALIZED_MATCH_BOX	CLE model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; 2B: current=1, sim=4; HR: current=1, sim=0; BB: current=3, sim=2; 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: Travis Bazzana: AB current=4 box=3, RBI current=1 box=0, R current=1 box=0, SO current=1 box=0; José Ramírez: 2B current=1 box=0, BB current=1 box=0; Angel Genao: H current=1 box=0; Nathaniel Lowe: H current=1 box=0, RBI current=1 box=0, SO current=1 box=2; Jo Adell: 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Steven Kwan: AB current=4 box=5, 2B current=0 box=1, RBI current=0 box=2, BB current=1 box=0; Patrick Bailey: AB current=3 box=4, H current=1 box=0; Petey Halpin: AB current=3 box=4, H current=0 box=2, R current=0 box=1	CLE
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

DET@TORHIT

Predicted: DET 5-4 (conf -0.12, moderate)Actual: DET 10-1Brier 0.192total bias -2

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 8; total bias -2
pregame home-minus-away evidence: season -0.572, recent -0.428, pitcher -0.050, composite -0.711
confidence pipeline reached 0.595 before later publication/recency controls; published home probability was 0.438
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 9 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

DET offense under-projected by 5
TOR offense over-projected by 3
Drew Anderson (DET) allowed 3 fewer ER than projected (0 vs 3)
Braydon Fisher (TOR) allowed 3 fewer ER than projected (0 vs 3)

Team	Starter	Projected	Actual	ER err
DET	Drew Anderson	3 ER / 15 outs	0 ER / 15 outs	-3
TOR	Braydon Fisher	3 ER / 15 outs	0 ER / 4 outs	-3

Decision internals: run factors home×0.948 / away×0.997; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Spencer Torkelson (DET 1B): 2-for-4, 1 HR, 3 RBI (projected 0.0 H, 0.0 HR) [run impact 2.18]
Hao-Yu Lee (DET 3B): 2-for-5, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Kazuma Okamoto (TOR 3B): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Max Clark (DET CF): 2-for-5, 0 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 0.78]

Starter miss: Drew Anderson (DET) posted 0.00 ERA vs projected 5.40 (0 ER in 15 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

What to do: Variance — no model change is warranted for this start.

Pitcher versus batter: what actually broke this game

AWAY pitching

Drew Anderson (DET) threw 15 outs, 0 ER = 0 ERA against a projected 5.4 ERA (-5.4 ERA error)

Bullpen behind him: 1 ER over 1 outs.

The HOME lineup he faced produced 1 runs on 6 hits and 0 HR, against a projected 4 runs / 8 hits (-3 runs vs projection).

No individual hitter beat his projection by a meaningful margin.

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Braydon Fisher (TOR) threw 4 outs, 0 ER = 0 ERA against a projected 5.4 ERA (-5.4 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 9 runs on 9 hits and 2 HR, against a projected 5 runs / 7 hits (4 runs vs projection).

Who did the damage to him:

Spencer Torkelson: 2-for-4, 1 HR, 3 RBI [+2.18 runs vs projection] season xwOBA 0.308, average - a good day, not a model error
Hao-Yu Lee: 2-for-5, 1 HR, 1 RBI [+1.4 runs vs projection] season xwOBA 0.297, below average - a good day, not a model error
Max Clark: 2-for-5, 0 HR, 2 RBI [+0.78 runs vs projection] season xwOBA 0.32, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
DET	Drew Anderson (Starter)	15 / 15	0 / 5.4 (-5.40)	0 / 3	2 / 2	0 / 1	3 / 0	2.71	4 / 5	4.62	venue away: 2.65 vs 2.66 ERA (ok) · opp run factor 0.95 (weak) · 47 prior appearances this season
DET	Kyle Finnegan (Middle Reliever (High Leverage))	1 / 5	27 / 0 (+27.00)	1 / 0	3 / 2	0 / 0	0 / 0	0.54	0 / 0	0.77
TOR	Braydon Fisher (Starter)	4 / 15	0 / 5.4 (-5.40)	0 / 3	0 / 4	0 / 0	2 / 3	2.82	1 / 5	2.63	venue home: 1.69 vs 0 ERA (ok) · opp run factor 1 (average) · 72 prior appearances this season

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Spencer Torkelson (1B)	DET	4 / 3	2 / 0	1 / 0	0 / 0	1 / 0	3 / 0	3 / 0	1 / 1	2 / 1	2.18	0.308 average	no signal
Hao-Yu Lee (3B)	DET	5 / 4	2 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	0 / 0	1 / 1	1.4	0.297 below average	good day
Kazuma Okamoto (3B)	TOR	4 / 3	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-0.93	0.33 above average	bad day
Max Clark (CF)	DET	5 / 3	2 / 1	1 / 0	0 / 0	0 / 0	2 / 0	2 / 1	0 / 0	1 / 1	0.78	0.32 average	no signal
Gleyber Torres (2B)	DET	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	-0.47	0.324 above average	—
Andrés Giménez (SS)	TOR	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.274 weak	—
John Peck (SS)	DET	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	1 / 1	0.33	0.355 elite	—
Alejandro Kirk (C)	TOR	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 1	0 / 1	-0.33	0.312 average	—
Riley Greene (LF)	DET	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	1 / 1	-0.31	0.373 elite	—
George Springer (DH)	TOR	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	-0.31	0.308 average	—
Eduardo Valencia (C)	DET	3 / 4	1 / 1	1 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 0	0 / 1	-0.29	0.329 above average	—
Kevin McGonigle (DH)	DET	1 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	3 / 1	0 / 1	0.19	0.354 elite	—
Brett Bateman (CF)	TOR	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	1 / 1	0.14	0.306 average	—
Vladimir Guerrero Jr. (1B)	TOR	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	-0.14	0.335 above average	—
Ernie Clement (2B)	TOR	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	-0.14	0.262 weak	—
Zach McKinstry (RF)	DET	5 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0	0.286 below average	—
Josh Smith (LF)	TOR	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.285 below average	—
Nathan Lukes (RF)	TOR	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.287 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.572	positive favors home; negative favors away
Recent form	-0.428	positive favors home; negative favors away
Starting-pitcher matchup	-0.05	positive favors home; negative favors away
Composite	-0.711	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.33-3.92	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.562	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	5.33-3.92	5-4	-0.415	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
DET	Max Clark	batter / CF	worse	predicted_RBI 2.36
TOR	Brett Bateman	batter / CF	worse	predicted_H -1.5
DET	Riley Greene	batter / LF	worse	predicted_H -1.49
TOR	Vladimir Guerrero Jr.	batter / 1B	worse	predicted_SO 1.43
DET	Kevin McGonigle	batter / DH	worse	predicted_SO 1.392
TOR	Andrés Giménez	batter / SS	worse	predicted_SO 1.323
TOR	Josh Smith	batter / LF	worse	predicted_AB 1.242
TOR	Kazuma Okamoto	batter / 3B	worse	predicted_SO -1.201

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	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	DET model pitcher projections differ from final simulation pitching box score (H: current=8, sim=10).	DET
info	CURRENT_FINALIZED_MATCH_BOX	DET model-prediction batting totals differ from final simulation box score (AB: current=32, sim=37; 2B: current=1, sim=2; HR: current=1, sim=0; RBI: current=4, sim=5; BB: current=4, sim=5; HBP: 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: Kevin McGonigle: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=2; Hao-Yu Lee: RBI current=1 box=0, R current=0 box=1, BB current=0 box=1, SO current=1 box=0; Eduardo Valencia: AB current=4 box=5, 2B current=0 box=1, HR current=1 box=0, SO current=1 box=0; Max Clark: AB current=3 box=5, H current=1 box=3, RBI current=0 box=3; Riley Greene: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Gleyber Torres: 2B current=0 box=1, RBI current=0 box=1, BB current=1 box=0, HBP current=0 box=1; Spencer Torkelson: AB current=3 box=4, H current=0 box=1, R current=0 box=1; Zach McKinstry: AB current=3 box=4, SO current=0 box=1	DET
info	CURRENT_FINALIZED_MATCH_BOX	TOR model pitcher projections differ from final simulation pitching box score (H: current=6, sim=7).	TOR
info	CURRENT_FINALIZED_MATCH_BOX	TOR model-prediction batting totals differ from final simulation box score (AB: current=32, sim=40; H: current=8, sim=10; BB: current=3, sim=0).	TOR
info	CURRENT_PLAYER_MATCH_BOX	TOR current finalized hitter lines are model predictions and differ from the simulation box score: Brett Bateman: AB current=4 box=5, H current=1 box=0, R current=1 box=0, BB current=1 box=0; Vladimir Guerrero Jr.: AB current=4 box=5, R current=1 box=0, SO current=0 box=2; Alejandro Kirk: AB current=4 box=5, RBI current=1 box=0, BB current=1 box=0; George Springer: AB current=4 box=5, H current=1 box=2, 2B current=1 box=0, BB current=1 box=0; Josh Smith: AB current=3 box=4, SO current=1 box=0; Nathan Lukes: H current=1 box=2, HR current=0 box=1, RBI current=0 box=1, R current=0 box=1; Kazuma Okamoto: AB current=3 box=4, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Ernie Clement: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, RBI current=0 box=1	TOR
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

KC@HOUHIT

Predicted: HOU 3-4 (conf +0.06, lean)Actual: HOU 2-4Brier 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 1; total bias +1
pregame home-minus-away evidence: season +0.212, recent +0.049, pitcher +0.048, composite +0.415
confidence pipeline reached 0.549 before later publication/recency controls; published home probability was 0.530
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

No large single-factor errors.

Team	Starter	Projected	Actual	ER err
KC	Michael Wacha	2 ER / 19 outs	4 ER / 18 outs	2
HOU	Hunter Brown	1 ER / 17 outs	1 ER / 15 outs	0

Decision internals: run factors home×1.004 / away×0.956; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Jeremy Peña (HOU SS): 3-for-4, 2 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 2.8]
Jac Caglianone (KC DH): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Isaac Paredes (HOU 3B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Daulton Varsho (HOU CF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]
Bobby Witt Jr. (KC SS): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.64]

Starter miss: Michael Wacha (KC) posted 6.00 ERA vs projected 2.84 (4 ER in 18 outs); he is materially worse pitching away (3.56 ERA there vs 2.93 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

Michael Wacha (KC) threw 18 outs, 4 ER = 6 ERA against a projected 2.84 ERA (3.16 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 4 runs on 4 hits and 3 HR, against a projected 4 runs / 7 hits (0 runs vs projection).

Who did the damage to him:

Jeremy Peña: 3-for-4, 2 HR, 2 RBI [+2.8 runs vs projection] season xwOBA 0.321, above average - real projection error
Yordan Alvarez: 1-for-2, 1 HR, 2 RBI [+0.33 runs vs projection] season xwOBA 0.443, elite - a good day, not a model error

Verdict: both: the starter was hit hard AND the lineup underproduced

HOME pitching

Hunter Brown (HOU) threw 15 outs, 1 ER = 1.8 ERA against a projected 1.59 ERA (0.21 ERA error)

Bullpen behind him: 0 ER over 9 outs.

The AWAY lineup he faced produced 2 runs on 3 hits and 1 HR, against a projected 4 runs / 7 hits (-2 runs vs projection).

Who did the damage to him:

Vinnie Pasquantino: 1-for-3, 1 HR, 1 RBI [+0.6 runs vs projection] season xwOBA 0.329, above average - real projection error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
KC	Michael Wacha (Starter)	18 / 19	6 / 2.84 (+3.16)	4 / 2	4 / 5	3 / 0	1 / 5	1.54	2 / 9	5.19	venue away: 3.56 vs 2.93 ERA (worse here) · opp run factor 1 (average) · 29 prior appearances this season
HOU	Hunter Brown (Starter)	15 / 17	1.8 / 1.59	1 / 1	2 / 3	0 / 1	3 / 1	2.37	10 / 4	5.99	venue home: 3.71 vs 3.26 ERA (ok) · opp run factor 0.96 (weak) · 17 prior appearances this season
HOU	AJ Blubaugh (Middle Reliever (High Leverage))	3 / 7	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0.61	1 / 2	1.18
HOU	Bryan Abreu (Setup Man)	3 / 3	0 / 18 (-18.00)	0 / 2	0 / 1	0 / 1	0 / 1	0.57	1 / 2	1.25
HOU	Josh Hader (Closer)	3 / 0	0 / n/a	0 / 0	1 / 0	0 / 0	0 / 0	0.2	1 / 1	1.2

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Jeremy Peña (SS)	HOU	4 / 4	3 / 1	0 / 0	0 / 0	2 / 0	2 / 1	3 / 1	0 / 0	1 / 1	2.8	0.321 above average	projection too low
Jac Caglianone (DH)	KC	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	3 / 1	-1.4	0.355 elite	bad day
Isaac Paredes (3B)	HOU	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	-0.8	0.322 above average	bad day
Daulton Varsho (CF)	HOU	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.78	0.287 below average	projection too high
Bobby Witt Jr. (SS)	KC	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	-0.64	0.381 elite	bad day
Vinnie Pasquantino (1B)	KC	3 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	0 / 1	2 / 1	0.6	0.329 above average	projection too low
Michael Massey (2B)	KC	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.308 average	—
Jose Altuve (2B)	HOU	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	-0.47	0.273 weak	—
Christian Walker (1B)	HOU	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.47	0.304 average	—
Yainer Diaz (C)	HOU	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.299 below average	—
Carter Jensen (C)	KC	4 / 4	1 / 1	1 / 1	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	2 / 1	-0.33	0.315 average	—
Yordan Alvarez (DH)	HOU	2 / 4	1 / 1	0 / 0	0 / 0	1 / 1	2 / 1	1 / 1	2 / 1	0 / 1	0.33	0.443 elite	—
LaMonte Wade Jr. (LF)	HOU	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	-0.33	0.276 weak	—
Maikel Garcia (3B)	KC	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	-0.14	0.317 average	—
John Rave (RF)	KC	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 1	-0.14	0.321 above average	—
Kyle Isbel (CF)	KC	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.261 weak	—
Isaac Collins (LF)	KC	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	2 / 1	0	0.297 below average	—
Cam Smith (RF)	HOU	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.32 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.212	positive favors home; negative favors away
Recent form	0.049	positive favors home; negative favors away
Starting-pitcher matchup	0.048	positive favors home; negative favors away
Composite	0.415	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.92-5.05	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-4	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	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	3.92-5.05	4-5	-0.126	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	3-4	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	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
HOU	Christian Walker	batter / 1B	worse	predicted_SO 3.07
KC	Michael Wacha	pitcher / Starter	neutral	predicted_SO 3
HOU	LaMonte Wade Jr.	batter / LF	worse	predicted_AB 2.5
KC	Jac Caglianone	batter / DH	worse	predicted_SO 1.985
HOU	Daulton Varsho	batter / CF	better	predicted_RBI 1.72
HOU	Cam Smith	batter / RF	worse	predicted_AB 1.44
HOU	Isaac Paredes	batter / 3B	better	predicted_SO 1.423
HOU	Yainer Diaz	batter / C	worse	predicted_H -1.23

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	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	HOU model-prediction batting totals differ from final simulation box score (AB: current=34, sim=39; H: current=7, sim=9; 2B: current=1, sim=3; HR: current=1, sim=0; BB: current=3, sim=6; SO: current=8, sim=10).	HOU
info	CURRENT_PLAYER_MATCH_BOX	HOU current finalized hitter lines are model predictions and differ from the simulation box score: Yordan Alvarez: HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Jose Altuve: RBI current=0 box=1, BB current=0 box=1, SO current=1 box=0; Isaac Paredes: 2B current=0 box=1, SO current=0 box=2; Jeremy Peña: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, BB current=0 box=1; Christian Walker: AB current=4 box=5, RBI current=1 box=0, SO current=1 box=4; Yainer Diaz: H current=1 box=0, BB current=0 box=1; Daulton Varsho: H current=1 box=2, RBI current=0 box=2, R current=0 box=1, BB current=0 box=1; Cam Smith: AB current=3 box=5, SO current=1 box=0	HOU
info	CURRENT_FINALIZED_MATCH_BOX	KC model pitcher projections differ from final simulation pitching box score (H: current=8, sim=9; BB: current=5, sim=6).	KC
info	CURRENT_FINALIZED_MATCH_BOX	KC model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=7, sim=5; 2B: current=2, sim=0; HR: current=1, sim=2; RBI: current=4, sim=3; R: current=4, sim=3; BB: current=4, sim=3).	KC
info	CURRENT_PLAYER_MATCH_BOX	KC current finalized hitter lines are model predictions and differ from the simulation box score: Jac Caglianone: RBI current=1 box=2, SO current=1 box=3; Maikel Garcia: H current=1 box=0, RBI current=1 box=0; Bobby Witt Jr.: AB current=4 box=5, 2B current=1 box=0, HR current=0 box=1, RBI current=0 box=1; Vinnie Pasquantino: AB current=4 box=3, RBI current=1 box=0, R current=0 box=1, SO current=1 box=0; Carter Jensen: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Michael Massey: H current=1 box=0; Kyle Isbel: AB current=3 box=4; Isaac Collins: H current=0 box=1	KC
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

LAD@CINHIT

Predicted: LAD 9-5 (conf -0.16, moderate)Actual: LAD 4-0Brier 0.176total bias +10

Root cause: Right winner, wrong score

Evidence-backed diagnosis

published system edge 0.16 (moderate); this is not the half-scale distance from 50%
team-run absolute error 10; total bias +10
pregame home-minus-away evidence: season -1.999, recent -1.698, pitcher -0.111, composite -2.488
confidence pipeline reached 0.743 before later publication/recency controls; published home probability was 0.420
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 10 (14 vs 4)
LAD offense over-projected by 5
CIN offense over-projected by 5

Team	Starter	Projected	Actual	ER err
LAD	Yoshinobu Yamamoto	2 ER / 21 outs	0 ER / 21 outs	-2
CIN	Rhett Lowder	3 ER / 14 outs	3 ER / 15 outs	0

Decision internals: run factors home×0.979 / away×1.089; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Max Muncy (LAD 3B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Alex Freeland (LAD 2B): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.71]
Teoscar Hernández (LAD LF): 1-for-3, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.26]
Hunter Feduccia (LAD C): 1-for-2, 0 HR, 1 RBI (projected 0.0 H, 0.0 HR) [run impact 1.13]
Josue De Paula (LAD DH): 0-for-5, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Pitcher versus batter: what actually broke this game

AWAY pitching

Yoshinobu Yamamoto (LAD) threw 21 outs, 0 ER = 0 ERA against a projected 2.57 ERA (-2.57 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The HOME lineup he faced produced 0 runs on 3 hits and 0 HR, against a projected 3 runs / 6 hits (-3 runs vs projection).

No individual hitter beat his projection by a meaningful margin.

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Rhett Lowder (CIN) threw 15 outs, 3 ER = 5.4 ERA against a projected 5.79 ERA (-0.39 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 4 runs on 8 hits and 2 HR, against a projected 5 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

Alex Freeland: 2-for-4, 1 HR, 1 RBI [+1.71 runs vs projection] season xwOBA 0.289, below average - a good day, not a model error
Teoscar Hernández: 1-for-3, 1 HR, 2 RBI [+1.26 runs vs projection] season xwOBA 0.333, above average - real projection error
Hunter Feduccia: 1-for-2, 0 HR, 1 RBI [+1.13 runs vs projection] season xwOBA 0.314, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
LAD	Yoshinobu Yamamoto (Starter)	21 / 21	0 / 2.57	0 / 2	1 / 3	0 / 0	2 / 1	1.54	7 / 8	7.71	venue away: 1.94 vs 3.13 ERA (ok) · opp run factor 0.98 (weak) · 26 prior appearances this season
LAD	Brock Stewart (Setup Man)	3 / 3	0 / 27 (-27.00)	0 / 3	1 / 3	0 / 1	0 / 0	0.48	1 / 0	0.76
LAD	Tanner Scott (Closer)	3 / 2	0 / 0	0 / 0	1 / 1	0 / 0	1 / 0	0.16	3 / 0	0.83
CIN	Rhett Lowder (Starter)	15 / 14	5.4 / 5.79	3 / 3	5 / 6	1 / 1	2 / 1	2.09	4 / 5	4.08	venue home: 4.29 vs 7.44 ERA (ok) · opp run factor 1.09 (strong) · 26 prior appearances this season

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Max Muncy (3B)	LAD	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	1 / 1	-1.73	0.348 elite	bad day
Alex Freeland (2B)	LAD	4 / 3	2 / 1	1 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	1 / 1	1.71	0.289 below average	good day
Teoscar Hernández (LF)	LAD	3 / 4	1 / 1	0 / 0	0 / 0	1 / 0	2 / 1	1 / 1	1 / 0	1 / 1	1.26	0.333 above average	projection too low
Hunter Feduccia (C)	LAD	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	2 / 0	0 / 1	1.13	0.314 average	no signal
Josue De Paula (DH)	LAD	5 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 0	-0.8	0.499 elite	bad day
JJ Bleday (LF)	CIN	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	-0.8	0.342 elite	bad day
Mookie Betts (SS)	LAD	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 0	-0.47	0.334 above average	—
Héctor Rodríguez (RF)	CIN	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.263 weak	—
Sal Stewart (1B)	CIN	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	3 / 1	-0.47	0.348 elite	—
Carlos Jorge (CF)	CIN	2 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	-0.45	0.203 weak	—
Eugenio Suárez (DH)	CIN	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	3 / 2	-0.33	0.298 below average	—
Freddie Freeman (1B)	LAD	5 / 4	3 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0.3	0.368 elite	—
Elly De La Cruz (SS)	CIN	3 / 4	2 / 1	1 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	0 / 2	-0.15	0.374 elite	—
Kyle Tucker (RF)	LAD	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	2 / 1	1 / 1	-0.14	0.33 above average	—
Tommy Edman (CF)	LAD	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.307 average	—
Juan Brito (3B)	CIN	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0	0.294 below average	—
Will Banfield (C)	CIN	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.127 weak	—
Matt McLain (2B)	CIN	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	3 / 1	0	0.304 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-1.999	positive favors home; negative favors away
Recent form	-1.698	positive favors home; negative favors away
Starting-pitcher matchup	-0.111	positive favors home; negative favors away
Composite	-2.488	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.18-2.53	n/a	Base ML historical model prediction
variance_applied	5-3	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	8-5	0.58	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	9-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	9-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	9-5	n/a	Capped at max 12 realistic runs
simulation_result	9-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	5.18-2.53	5-3	-0.647	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	8-5	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	8-5	9-5	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	9-5	9-5	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	9-5	9-5	-0	Capped at max 12 realistic runs
simulation_result	9-5	9-5	-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
CIN	JJ Bleday	batter / LF	better	predicted_RBI 3.63
LAD	Yoshinobu Yamamoto	pitcher / Starter	better	predicted_SO 3
LAD	Josue De Paula	batter / DH	neutral	predicted_AB 2.5
LAD	Hunter Feduccia	batter / C	worse	predicted_SO 2.278
LAD	Teoscar Hernández	batter / LF	better	predicted_H 2
LAD	Tanner Scott	pitcher / Closer	worse	predicted_SO -2
LAD	Mookie Betts	batter / SS	better	predicted_HR 1.76
LAD	Alex Freeland	batter / 2B	worse	predicted_AB 1.293

Structural audit findings

Severity	Rule	Finding	Subject
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_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	CIN model pitcher projections differ from final simulation pitching box score (ER: current=8, sim=9).	CIN
info	CURRENT_FINALIZED_MATCH_BOX	CIN model-prediction batting totals differ from final simulation box score (AB: current=32, sim=34; H: current=6, sim=7; RBI: current=4, sim=5; R: current=3, sim=5; BB: current=3, sim=1; SO: current=11, sim=8).	CIN
info	CURRENT_PLAYER_MATCH_BOX	CIN current finalized hitter lines are model predictions and differ from the simulation box score: JJ Bleday: HR current=0 box=1, RBI current=0 box=4, R current=0 box=1, BB current=1 box=0; Carlos Jorge: AB current=4 box=3, 2B current=1 box=0, RBI current=1 box=0, R current=0 box=1; Héctor Rodríguez: H current=1 box=0, SO current=1 box=2; Sal Stewart: H current=1 box=2; Juan Brito: AB current=3 box=4, 2B current=0 box=1, R current=1 box=0; Elly De La Cruz: HR current=1 box=0, RBI current=1 box=0, BB current=1 box=0, SO current=2 box=1; Will Banfield: SO current=1 box=0; Matt McLain: AB current=3 box=4, SO current=1 box=2	CIN
info	CURRENT_FINALIZED_MATCH_BOX	LAD model-prediction batting totals differ from final simulation box score (AB: current=33, sim=39; H: current=8, sim=12; 2B: current=1, sim=3; HR: current=1, sim=2; RBI: current=4, sim=9; R: current=5, sim=9; BB: current=4, sim=2; SO: current=7, sim=9).	LAD
info	CURRENT_PLAYER_MATCH_BOX	LAD current finalized hitter lines are model predictions and differ from the simulation box score: Josue De Paula: AB current=3 box=5, RBI current=0 box=3, BB current=1 box=0; Teoscar Hernández: H current=1 box=3, 2B current=0 box=1, RBI current=1 box=2, BB current=0 box=1; Kyle Tucker: AB current=4 box=5, 2B current=0 box=1, BB current=1 box=0, SO current=1 box=2; Tommy Edman: R current=0 box=1; Freddie Freeman: AB current=4 box=5, H current=1 box=2, R current=0 box=1, BB current=1 box=0; Mookie Betts: H current=1 box=2, HR current=0 box=2, RBI current=1 box=2, HBP current=0 box=1; Max Muncy: HR current=1 box=0; Hunter Feduccia: AB current=3 box=4, R current=0 box=1, SO current=1 box=3	LAD
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

MIA@ARIMISS

Predicted: ARI 4-6 (conf +0.09, lean)Actual: MIA 4-2Brier 0.299total bias +4

Root cause: Small edge on the losing side

Evidence-backed diagnosis

published system edge 0.09 (lean); this is not the half-scale distance from 50%
team-run absolute error 4; total bias +4
pregame home-minus-away evidence: season +0.172, recent +0.229, pitcher +0.105, composite +0.743
confidence pipeline reached 0.555 before later publication/recency controls; published home probability was 0.547
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 4 (10 vs 6)
ARI offense over-projected by 4

Team	Starter	Projected	Actual	ER err
MIA	Janson Junk	3 ER / 14 outs	1 ER / 9 outs	-2
ARI	Michael Soroka	1 ER / 15 outs	1 ER / 15 outs	0

Decision internals: run factors home×1.039 / away×0.934; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Gabriel Moreno (ARI C): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.75]
Ketel Marte (ARI 2B): 2-for-6, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]
Heriberto Hernández (MIA LF): 1-for-5, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Nolan Arenado (ARI 3B): 1-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Jakob Marsee (MIA CF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Pitcher versus batter: what actually broke this game

AWAY pitching

Janson Junk (MIA) threw 9 outs, 1 ER = 3 ERA against a projected 5.79 ERA (-2.79 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The HOME lineup he faced produced 1 runs on 8 hits and 1 HR, against a projected 4 runs / 8 hits (-3 runs vs projection).

Who did the damage to him:

Gabriel Moreno: 2-for-4, 1 HR, 1 RBI [+1.75 runs vs projection] season xwOBA 0.359, elite - real projection error
Ketel Marte: 2-for-6, 0 HR, 1 RBI [+1.09 runs vs projection] season xwOBA 0.345, elite - real projection error

Verdict: offense: the lineup scored well below its projection

HOME pitching

No settled pitching line for this club. The forecast named Michael Soroka (ARI) at a projected 1.8 ERA, but the per-player feed carries no pitching row for him in this game, so the actual-versus-projected comparison cannot be made on this side. That is a source-coverage gap in the settled box, not a forecasting failure.

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 3 runs on 4 hits and 1 HR, against a projected 5 runs / 8 hits (-2 runs vs projection).

Who did the damage to him:

Griffin Conine: 1-for-2, 1 HR, 1 RBI [+0.6 runs vs projection] season xwOBA 0.364, elite - real projection error
Deyvison De Los Santos: 1-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.2, weak - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIA	Janson Junk (Starter)	9 / 14	3 / 5.79	1 / 3	5 / 5	1 / 1	1 / 2	1.24	4 / 8	2.55	venue away: 4.78 vs 4.46 ERA (ok) · opp run factor 1.04 (average) · 23 prior appearances this season
MIA	Calvin Faucher (Closer)	3 / 5	0 / 16.2 (-16.20)	0 / 3	1 / 2	0 / 0	1 / 2	0.51	1 / 0	0.97
MIA	Tyler Zuber (Setup Man)	3 / 5	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.45	1 / 0	1.13

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Gabriel Moreno (C)	ARI	4 / 4	2 / 1	0 / 1	0 / 0	1 / 0	1 / 1	1 / 1	2 / 0	1 / 0	1.75	0.359 elite	projection too low
Ketel Marte (2B)	ARI	6 / 4	2 / 1	2 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 0	0 / 0	1.09	0.345 elite	projection too low
Heriberto Hernández (LF)	MIA	5 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-0.93	0.351 elite	bad day
Nolan Arenado (3B)	ARI	3 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	0 / 1	-0.93	0.291 below average	projection too high
Jakob Marsee (CF)	MIA	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	-0.8	0.309 average	no signal
Kyle Stowers (DH)	MIA	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 1	2 / 1	-0.78	0.326 above average	bad day
Griffin Conine (RF)	MIA	2 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 1	0 / 1	0 / 1	0.6	0.364 elite	projection too low
Agustín Ramírez (C)	MIA	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.282 weak	—
Deyvison De Los Santos (1B)	MIA	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.47	0.2 weak	—
Graham Pauley (3B)	MIA	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.247 weak	—
Lars Nootbaar (LF)	ARI	5 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	4 / 1	-0.47	0.351 elite	—
Pavin Smith (1B)	ARI	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.253 weak	—
Javier Sanoja (2B)	MIA	5 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 0	-0.31	0.295 below average	—
Corbin Carroll (RF)	ARI	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	-0.31	0.343 elite	—
Otto Lopez (SS)	MIA	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	2 / 0	1 / 0	0.19	0.327 above average	—
Geraldo Perdomo (SS)	ARI	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 0	0	0.326 above average	—
Jordan Lawlar (CF)	ARI	5 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	4 / 1	0	0.332 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.172	positive favors home; negative favors away
Recent form	0.229	positive favors home; negative favors away
Starting-pitcher matchup	0.105	positive favors home; negative favors away
Composite	0.743	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.66-5.25	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.547	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	4-6	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-6	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-6	n/a	Capped at max 12 realistic runs
simulation_result	4-6	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.66-5.25	4-5	-0.589	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 matchup_recency_guardrail)
embedded_savant_run_environment	4-5	4-6	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-6	4-6	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-6	4-6	0	Capped at max 12 realistic runs
simulation_result	4-6	4-6	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	Ketel Marte	batter / 2B	worse	predicted_AB -2.96
ARI	Geraldo Perdomo	batter / SS	better	predicted_RBI 2.7
ARI	Lars Nootbaar	batter / LF	worse	predicted_AB 1.805
ARI	Gabriel Moreno	batter / C	worse	predicted_AB -1.73
MIA	Agustín Ramírez	batter / C	better	predicted_RBI 1.644
MIA	Heriberto Hernández	batter / LF	worse	predicted_AB -1.55
MIA	Deyvison De Los Santos	batter / 1B	deboosted	predicted_AB 1.5
MIA	Otto Lopez	batter / SS	worse	predicted_H -1.46

Structural audit findings

Severity	Rule	Finding	Subject
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	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_FINALIZED_MATCH_BOX	ARI model-prediction batting totals differ from final simulation box score (AB: current=33, sim=31; H: current=8, sim=7; 2B: current=2, sim=1; 3B: current=0, sim=1; RBI: current=4, sim=6; R: current=4, sim=6; BB: current=3, sim=5; SO: current=5, sim=8).	ARI
info	CURRENT_PLAYER_MATCH_BOX	ARI current finalized hitter lines are model predictions and differ from the simulation box score: Lars Nootbaar: AB current=4 box=5, H current=1 box=0, BB current=1 box=0; Corbin Carroll: 2B current=1 box=0, RBI current=1 box=2, BB current=1 box=0, SO current=1 box=2; Geraldo Perdomo: H current=1 box=2, 2B current=0 box=1, 3B current=0 box=1, RBI current=0 box=3; Jesús Sánchez: R current=0 box=1, BB current=0 box=1, SO current=0 box=1; Gabriel Moreno: AB current=4 box=3, H current=1 box=2, 2B current=1 box=0, RBI current=1 box=0; Ketel Marte: AB current=4 box=2, HR current=0 box=1, BB current=0 box=2, SO current=0 box=1; Pavin Smith: AB current=3 box=2, R current=0 box=1, BB current=0 box=1, SO current=1 box=0; Nolan Arenado: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0	ARI
info	CURRENT_FINALIZED_MATCH_BOX	MIA model-prediction batting totals differ from final simulation box score (AB: current=33, sim=32; H: current=8, sim=5; 2B: current=2, sim=0; 3B: current=0, sim=2; HR: current=1, sim=0; R: current=5, sim=4; BB: current=3, sim=4; SO: current=7, sim=6).	MIA
info	CURRENT_PLAYER_MATCH_BOX	MIA current finalized hitter lines are model predictions and differ from the simulation box score: Kyle Stowers: 2B current=1 box=0, 3B current=0 box=1, R current=0 box=1, BB current=1 box=0; Javier Sanoja: H current=1 box=2, 2B current=1 box=0, 3B current=0 box=1, RBI current=1 box=0; Griffin Conine: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Agustín Ramírez: AB current=3 box=4, RBI current=0 box=2, R current=0 box=1, SO current=1 box=0; Heriberto Hernández: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Deyvison De Los Santos: AB current=3 box=4, H current=0 box=1, SO current=1 box=0; Jakob Marsee: AB current=4 box=3, H current=1 box=0, RBI current=0 box=1; Otto Lopez: H current=1 box=0, R current=1 box=0, SO current=0 box=2	MIA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

MIL@PITHIT

Predicted: MIL 5-2 (conf -0.12, moderate)Actual: MIL 5-1Brier 0.194total bias +1

Root cause: Strong correct call

Evidence-backed diagnosis

published system edge 0.12 (moderate); this is not the half-scale distance from 50%
team-run absolute error 1; total bias +1
pregame home-minus-away evidence: season -0.970, recent -0.414, pitcher -0.026, composite -1.235
confidence pipeline reached 0.642 before later publication/recency controls; published home probability was 0.440
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

No large single-factor errors.

Team	Starter	Projected	Actual	ER err
MIL	Jacob Misiorowski	2 ER / 18 outs	0 ER / 15 outs	-2
PIT	Lake Bachar	2 ER / 15 outs	1 ER / 6 outs	-1

Decision internals: run factors home×0.978 / away×1.005; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Rafael Flores Jr. (PIT C): 4-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 1.41]
Jackson Chourio (MIL LF): 0-for-5, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Joey Ortiz (MIL SS): 3-for-4, 0 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.25]
Bryan Reynolds (PIT RF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Oneil Cruz (PIT CF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Starter miss: Jacob Misiorowski (MIL) posted 0.00 ERA vs projected 3.00 (0 ER in 15 outs); he is materially worse pitching away (2.39 ERA there vs 1.61 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

Jacob Misiorowski (MIL) threw 15 outs, 0 ER = 0 ERA against a projected 3 ERA (-3 ERA error)

Bullpen behind him: 0 ER over 9 outs.

The HOME lineup he faced produced 1 runs on 9 hits and 1 HR, against a projected 4 runs / 7 hits (-3 runs vs projection).

Who did the damage to him:

Rafael Flores Jr.: 4-for-4, 0 HR, 0 RBI [+1.41 runs vs projection] season xwOBA 0.335, above average - real projection error
Nick Gonzales: 2-for-3, 0 HR, 0 RBI [+0.8 runs vs projection] season xwOBA 0.326, above average - real projection error
Ryan O'Hearn: 1-for-3, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.315, average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Lake Bachar (PIT) threw 6 outs, 1 ER = 4.5 ERA against a projected 3.6 ERA (0.9 ERA error)

Bullpen behind him: 4 ER over 3 outs.

The AWAY lineup he faced produced 5 runs on 8 hits and 0 HR, against a projected 6 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

Joey Ortiz: 3-for-4, 0 HR, 2 RBI [+1.25 runs vs projection] season xwOBA 0.3, below average - a good day, not a model error
David Hamilton: 1-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.274, weak - a good day, not a model error
William Contreras: 1-for-5, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.336, above average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIL	Jacob Misiorowski (Starter)	15 / 18	0 / 3 (-3.00)	0 / 2	5 / 5	0 / 0	1 / 4	2.14	7 / 7	7.4	venue away: 2.39 vs 1.61 ERA (worse here) · opp run factor 0.98 (average) · 27 prior appearances this season
MIL	Aaron Ashby (Middle Reliever (High Leverage))	6 / 3	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0.55	2 / 0	1.3
MIL	Trevor Megill (Closer)	3 / 3	0 / 0	0 / 0	2 / 0	0 / 0	0 / 0	0.29	0 / 1	1.23
PIT	Lake Bachar (Starter)	6 / 15	4.5 / 3.6	1 / 2	3 / 4	0 / 0	2 / 4	2.45	1 / 4	4.75	venue home: 0.93 vs 2.53 ERA (ok) · opp run factor 1 (strong) · 48 prior appearances this season
PIT	Gregory Soto (Middle Reliever (High Leverage))	1 / 3	108 / 0 (+108.00)	4 / 0	2 / 1	0 / 0	2 / 0	0.44	1 / 1	1.19
PIT	Yohan Ramírez (Middle Reliever (High Leverage))	2 / 3	0 / 18 (-18.00)	0 / 2	0 / 2	0 / 1	0 / 0	0.5	1 / 1	1.12

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Rafael Flores Jr. (C)	PIT	4 / 4	4 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	1.41	0.335 above average	projection too low
Jackson Chourio (LF)	MIL	5 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-1.4	0.342 elite	bad day
Joey Ortiz (SS)	MIL	4 / 3	3 / 1	1 / 0	0 / 0	0 / 0	2 / 0	0 / 0	0 / 0	0 / 1	1.25	0.3 below average	good day
Bryan Reynolds (RF)	PIT	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	-1.11	0.359 elite	bad day
Oneil Cruz (CF)	PIT	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	4 / 2	-0.8	0.358 elite	bad day
Nick Gonzales (3B)	PIT	3 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0.8	0.326 above average	projection too low
Christian Yelich (DH)	MIL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	-0.47	0.312 average	—
David Hamilton (3B)	MIL	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0.47	0.274 weak	—
Jake Bauers (1B)	MIL	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	2 / 1	-0.33	0.38 elite	—
Ryan O'Hearn (DH)	PIT	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	0.33	0.315 average	—
Spencer Horwitz (1B)	PIT	2 / 4	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	-0.33	0.311 average	—
William Contreras (C)	MIL	5 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	0 / 1	0.31	0.336 above average	—
Luis Lara (RF)	MIL	3 / 4	1 / 1	1 / 0	0 / 0	0 / 0	2 / 0	1 / 0	0 / 0	0 / 1	0.31	0.301 below average	—
Konnor Griffin (SS)	PIT	5 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.31	0.303 below average	—
Brice Turang (2B)	MIL	1 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	4 / 1	1 / 1	0.21	0.345 elite	—
Garrett Mitchell (CF)	MIL	3 / 4	1 / 1	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	0	0.356 elite	—
Jake Mangum (LF)	PIT	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.287 below average	—
Brandon Lowe (2B)	PIT	5 / 4	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 0	2 / 2	0	0.321 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.97	positive favors home; negative favors away
Recent form	-0.414	positive favors home; negative favors away
Starting-pitcher matchup	-0.026	positive favors home; negative favors away
Composite	-1.235	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.92-3.48	n/a	Base ML historical model prediction
variance_applied	5-3	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	5-2	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-2	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-2	n/a	Capped at max 12 realistic runs
simulation_result	5-2	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.92-3.48	5-3	0.553	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	5-3	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	5-3	5-2	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-2	5-2	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-2	5-2	-0	Capped at max 12 realistic runs
simulation_result	5-2	5-2	-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
MIL	Jacob Misiorowski	pitcher / Starter	better	predicted_SO 4
MIL	Aaron Ashby	pitcher / Middle Reliever (High Leverage)	worse	predicted_SO -2
MIL	Abner Uribe	pitcher / Setup Man	worse	predicted_SO -2
MIL	Luis Lara	batter / RF	better	predicted_2B 1.87
MIL	Garrett Mitchell	batter / CF	worse	predicted_AB 1.714
MIL	William Contreras	batter / C	worse	predicted_RBI -1.7
PIT	Spencer Horwitz	batter / 1B	better	predicted_RBI 1.59
MIL	Jake Bauers	batter / 1B	worse	predicted_AB -1.49

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	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_FINALIZED_MATCH_BOX	MIL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; 2B: current=2, sim=3; HR: current=1, sim=2; R: current=6, sim=5; SO: current=9, sim=7; HBP: 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: William Contreras: AB current=4 box=5, 2B current=0 box=1, RBI current=1 box=0, R current=1 box=0; Garrett Mitchell: AB current=4 box=5, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Brice Turang: 2B current=1 box=0, HR current=0 box=1; Luis Lara: H current=1 box=2, 2B current=0 box=2, RBI current=0 box=2, R current=0 box=1; Jackson Chourio: AB current=4 box=5; Jake Bauers: AB current=3 box=2, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Christian Yelich: AB current=4 box=3; David Hamilton: BB current=0 box=1	MIL
info	CURRENT_FINALIZED_MATCH_BOX	PIT model-prediction batting totals differ from final simulation box score (AB: current=35, sim=32; H: current=7, sim=5; 2B: current=2, sim=1; HR: current=1, sim=0; RBI: current=5, sim=2; R: current=4, sim=2; BB: current=3, sim=4; SO: current=11, sim=8).	PIT
info	CURRENT_PLAYER_MATCH_BOX	PIT current finalized hitter lines are model predictions and differ from the simulation box score: Bryan Reynolds: H current=1 box=0, 2B current=1 box=0, R current=1 box=0, BB current=1 box=0; Konnor Griffin: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Ryan O'Hearn: H current=1 box=0, RBI current=1 box=0; Oneil Cruz: AB current=4 box=3, RBI current=1 box=0, SO current=2 box=1; Jake Mangum: H current=0 box=1, BB current=0 box=1; Brandon Lowe: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Rafael Flores Jr.: AB current=4 box=3, RBI current=1 box=0, BB current=0 box=1, SO current=1 box=0; Spencer Horwitz: H current=0 box=1, 2B current=0 box=1, RBI current=0 box=2, R current=0 box=1	PIT
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

NYY@MINHIT

Predicted: NYY 6-3 (conf -0.14, moderate)Actual: NYY 8-1Brier 0.185total bias +0

Root cause: Solid correct call

Evidence-backed diagnosis

published system edge 0.14 (moderate); this is not the half-scale distance from 50%
team-run absolute error 4; total bias +0
pregame home-minus-away evidence: season -1.318, recent -1.012, pitcher -0.046, composite -1.619
confidence pipeline reached 0.653 before later publication/recency controls; published home probability was 0.430
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

No large single-factor errors.

Team	Starter	Projected	Actual	ER err
NYY	Max Fried	1 ER / 18 outs	1 ER / 21 outs	0
MIN	Bailey Ober	4 ER / 15 outs	3 ER / 16 outs	-1

Decision internals: run factors home×0.992 / away×1.038; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Cody Bellinger (NYY LF): 3-for-4, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.56]
Ben Rice (NYY DH): 1-for-6, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.26]
Heliot Ramos (NYY RF): 2-for-4, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.11]
Anthony Volpe (NYY 2B): 2-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 1.11]
Josh Bell (MIN DH): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]

Pitcher versus batter: what actually broke this game

AWAY pitching

Max Fried (NYY) threw 21 outs, 1 ER = 1.29 ERA against a projected 1.5 ERA (-0.21 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 1 runs on 7 hits and 0 HR, against a projected 4 runs / 7 hits (-3 runs vs projection).

Who did the damage to him:

Emmanuel Rodriguez: 2-for-3, 0 HR, 1 RBI [+0.78 runs vs projection] season xwOBA 0.501, elite - real projection error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Bailey Ober (MIN) threw 16 outs, 3 ER = 5.06 ERA against a projected 7.2 ERA (-2.14 ERA error)

Bullpen behind him: 1 ER over 6 outs.

The AWAY lineup he faced produced 8 runs on 14 hits and 2 HR, against a projected 4 runs / 8 hits (4 runs vs projection).

Who did the damage to him:

Cody Bellinger: 3-for-4, 1 HR, 2 RBI [+1.56 runs vs projection] season xwOBA 0.338, above average - real projection error
Heliot Ramos: 2-for-4, 0 HR, 1 RBI [+1.11 runs vs projection] season xwOBA 0.327, above average - real projection error
Anthony Volpe: 2-for-4, 0 HR, 0 RBI [+1.11 runs vs projection] season xwOBA 0.29, below average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
NYY	Max Fried (Starter)	21 / 18	1.29 / 1.5	1 / 1	6 / 1	0 / 0	0 / 3	1.59	6 / 6	5.91	venue away: 2.47 vs 3.66 ERA (ok) · opp run factor 0.99 (average) · 18 prior appearances this season
MIN	Bailey Ober (Starter)	16 / 15	5.06 / 7.2	3 / 4	9 / 5	1 / 1	2 / 2	1.8	3 / 6	3.74	venue home: 3.16 vs 5.84 ERA (ok) · opp run factor 1.04 (strong) · 22 prior appearances this season
MIN	Andrew Morris (Middle Reliever (High Leverage))	3 / 3	9 / 18 (-9.00)	1 / 2	1 / 3	0 / 0	1 / 0	0.37	0 / 0	0.87
MIN	Yoendrys Gómez (Closer)	3 / 3	0 / 0	0 / 0	1 / 1	0 / 0	0 / 1	0.67	2 / 1	0.88

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Cody Bellinger (LF)	NYY	4 / 4	3 / 1	0 / 1	0 / 0	1 / 0	2 / 1	1 / 1	1 / 1	1 / 0	1.56	0.338 above average	projection too low
Ben Rice (DH)	NYY	6 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	0 / 1	1 / 1	-1.26	0.377 elite	bad day
Heliot Ramos (RF)	NYY	4 / 3	2 / 1	1 / 0	0 / 0	0 / 0	1 / 0	2 / 0	1 / 0	1 / 1	1.11	0.327 above average	projection too low
Anthony Volpe (2B)	NYY	4 / 3	2 / 1	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	1.11	0.29 below average	good day
Josh Bell (DH)	MIN	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	-1.11	0.335 above average	bad day
George Lombard Jr. (SS)	NYY	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 1	0.8	0.267 weak	good day
Emmanuel Rodriguez (CF)	MIN	3 / 4	2 / 1	1 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	0 / 1	0.78	0.501 elite	projection too low
Spencer Jones (CF)	NYY	5 / 4	1 / 1	0 / 0	0 / 0	1 / 0	3 / 1	1 / 0	0 / 1	0 / 1	0.6	0.333 above average	projection too low
Ali Sánchez (C)	NYY	5 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.47	0.25 weak	—
Kaelen Culpepper (SS)	MIN	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	-0.47	0.304 average	—
Brooks Lee (3B)	MIN	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.47	0.275 weak	—
Austin Martin (LF)	MIN	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	-0.33	0.311 average	—
Luis García Jr. (1B)	NYY	5 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	1 / 0	-0.31	0.352 elite	—
Kody Clemens (2B)	MIN	4 / 4	1 / 1	0 / 0	1 / 0	0 / 1	0 / 1	0 / 1	0 / 0	0 / 1	-0.31	0.337 above average	—
Luke Keaschall (RF)	MIN	4 / 4	3 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0.3	0.316 average	—
José Caballero (3B)	NYY	5 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	1 / 1	0	0.281 weak	—
Royce Lewis (1B)	MIN	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0	0.312 average	—
Victor Caratini (C)	MIN	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.317 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-1.318	positive favors home; negative favors away
Recent form	-1.012	positive favors home; negative favors away
Starting-pitcher matchup	-0.046	positive favors home; negative favors away
Composite	-1.619	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.96-3.02	n/a	Base ML historical model prediction
variance_applied	5-3	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	0.57	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
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	4.96-3.02	5-3	0.062	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	5-3	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
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
NYY	José Caballero	batter / 3B	better	predicted_RBI 2.581
MIN	Kody Clemens	batter / 2B	worse	predicted_SO 2.083
NYY	Heliot Ramos	batter / RF	better	predicted_2B 1.838
MIN	Emmanuel Rodriguez	batter / CF	worse	predicted_AB -1.62
NYY	Ali Sánchez	batter / C	worse	predicted_AB 1.5
NYY	Spencer Jones	batter / CF	worse	predicted_AB 1.43
MIN	Luke Keaschall	batter / RF	worse	predicted_SO 1.414
NYY	Cody Bellinger	batter / LF	better	predicted_SO 1.367

Structural audit findings

Severity	Rule	Finding	Subject
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_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	MIN model-prediction batting totals differ from final simulation box score (AB: current=34, sim=30; H: current=7, sim=3; 2B: current=2, sim=1; HR: current=1, sim=0; RBI: current=4, sim=3; R: current=4, sim=3; SO: current=7, sim=11; HBP: current=0, sim=1).	MIN
info	CURRENT_PLAYER_MATCH_BOX	MIN current finalized hitter lines are model predictions and differ from the simulation box score: Luke Keaschall: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, R current=1 box=0; Kaelen Culpepper: AB current=4 box=3, H current=1 box=0, R current=1 box=0, HBP current=0 box=1; Royce Lewis: RBI current=0 box=1, R current=0 box=1; Josh Bell: RBI current=1 box=2, BB current=1 box=0, SO current=1 box=0; Kody Clemens: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Austin Martin: BB current=1 box=0; Brooks Lee: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Victor Caratini: AB current=3 box=4, H current=0 box=1, R current=0 box=1, SO current=1 box=0	MIN
info	CURRENT_FINALIZED_MATCH_BOX	NYY model-prediction batting totals differ from final simulation box score (AB: current=31, sim=36; H: current=8, sim=9; 2B: current=2, sim=3; RBI: current=4, sim=6; R: current=4, sim=6; BB: current=3, sim=4; SO: current=7, sim=8; 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: Heliot Ramos: AB current=3 box=4, H current=1 box=2, 2B current=0 box=2, RBI current=0 box=1; Luis García Jr.: AB current=4 box=5, 2B current=1 box=0, SO current=0 box=1; Spencer Jones: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Cody Bellinger: SO current=0 box=2; Anthony Volpe: R current=0 box=1, BB current=0 box=1; José Caballero: HR current=0 box=1, RBI current=0 box=3, BB current=0 box=1; Ben Rice: AB current=4 box=5, H current=1 box=2, HR current=1 box=0, RBI current=1 box=0; George Lombard Jr.: H current=1 box=0, BB current=0 box=1	NYY
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

OAK@TBHIT

Predicted: TB 4-8 (conf +0.33, strong)Actual: TB 1-2Brier 0.111total bias +9

Root cause: Right winner, wrong score

Evidence-backed diagnosis

published system edge 0.33 (strong); this is not the half-scale distance from 50%
team-run absolute error 9; total bias +9
pregame home-minus-away evidence: season +1.499, recent +0.710, pitcher +0.129, composite +2.005
confidence pipeline reached 0.667 before later publication/recency controls; published home probability was 0.667
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 9 (12 vs 3)
OAK offense over-projected by 3
TB offense over-projected by 6
Jack Perkins (OAK) allowed 4 fewer ER than projected (0 vs 4)
Griffin Jax (TB) allowed 4 fewer ER than projected (0 vs 4)

Team	Starter	Projected	Actual	ER err
OAK	Jack Perkins	4 ER / 12 outs	0 ER / 20 outs	-4
TB	Griffin Jax	4 ER / 15 outs	0 ER / 15 outs	-4

Decision internals: run factors home×1.062 / away×0.934; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Chandler Simpson (TB LF): 3-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 1.25]
Donovan Walton (OAK 2B): 0-for-2, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Shea Langeliers (OAK DH): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Lawrence Butler (OAK RF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Richie Palacios (TB 2B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Starter miss: Jack Perkins (OAK) posted 0.00 ERA vs projected 9.00 (0 ER in 20 outs); he is materially worse pitching away (8.10 ERA there vs 7.42 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

Jack Perkins (OAK) threw 20 outs, 0 ER = 0 ERA against a projected 9 ERA (-9 ERA error)

Bullpen behind him: 0 ER over 3 outs.

The HOME lineup he faced produced 2 runs on 6 hits and 1 HR, against a projected 4 runs / 8 hits (-2 runs vs projection).

Who did the damage to him:

Chandler Simpson: 3-for-3, 0 HR, 0 RBI [+1.25 runs vs projection] season xwOBA 0.264, weak - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

HOME pitching

Griffin Jax (TB) threw 15 outs, 0 ER = 0 ERA against a projected 7.2 ERA (-7.2 ERA error)

Bullpen behind him: 0 ER over 10 outs.

The AWAY lineup he faced produced 1 runs on 4 hits and 0 HR, against a projected 4 runs / 8 hits (-3 runs vs projection).

Who did the damage to him:

Alika Williams: 1-for-3, 0 HR, 0 RBI [+0.78 runs vs projection] season xwOBA 0.243, weak - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
OAK	Jack Perkins (Starter)	20 / 12	0 / 9 (-9.00)	0 / 4	4 / 5	0 / 2	1 / 1	2.37	8 / 2	4.23	venue away: 8.1 vs 7.42 ERA (worse here) · opp run factor 1.06 (average) · 34 prior appearances this season
OAK	Hogan Harris (Reliever)	3 / 3	0 / 18 (-18.00)	0 / 2	1 / 3	0 / 1	0 / 0	n/a	1 / 0	n/a
TB	Griffin Jax (Starter)	15 / 15	0 / 7.2 (-7.20)	0 / 4	2 / 5	0 / 1	0 / 2	1.3	7 / 4	4.01	venue home: 3.99 vs 3.14 ERA (worse here) · opp run factor 0.93 (weak) · 31 prior appearances this season
TB	Bryan Baker (Closer)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.33	0 / 2	0.77
TB	Kevin Kelly (Setup Man)	3 / 3	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0.13	3 / 1	0.62
TB	Cam Booser (Middle Reliever (High Leverage))	4 / 2	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.14	1 / 1	0.84

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Chandler Simpson (LF)	TB	3 / 4	3 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	1.25	0.264 weak	good day
Donovan Walton (2B)	OAK	2 / 3	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	-1.11	0.336 above average	bad day
Shea Langeliers (DH)	OAK	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-0.93	0.364 elite	bad day
Lawrence Butler (RF)	OAK	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	2 / 1	-0.8	0.325 above average	bad day
Richie Palacios (2B)	TB	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	-0.8	0.309 average	no signal
Alika Williams (SS)	OAK	3 / 3	1 / 0	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.78	0.243 weak	good day
Yandy Díaz (DH)	TB	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.78	0.35 elite	bad day
Carlos Cortes (LF)	OAK	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.305 average	—
Jonah Heim (C)	OAK	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.29 below average	—
Zack Gelof (3B)	OAK	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.47	0.313 average	—
Cedric Mullins (CF)	TB	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.266 weak	—
Henry Bolte (CF)	OAK	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	3 / 1	-0.33	0.317 average	—
Junior Caminero (3B)	TB	4 / 4	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	2 / 1	-0.33	0.362 elite	—
Jonathan Aranda (1B)	TB	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.33	0.348 elite	—
Jeff McNeil (1B)	OAK	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 0	-0.31	0.305 average	—
Liam Hicks (C)	TB	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 0	-0.14	0.317 average	—
Jorge Mateo (SS)	TB	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.29 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	1.499	positive favors home; negative favors away
Recent form	0.71	positive favors home; negative favors away
Starting-pitcher matchup	0.129	positive favors home; negative favors away
Composite	2.005	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.9-5.74	n/a	Base ML historical model prediction
variance_applied	4-6	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-8	0.667	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	4-8	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-8	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-8	n/a	Capped at max 12 realistic runs
simulation_result	4-8	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.9-5.74	4-6	0.159	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-6	4-8	2	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	4-8	4-8	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-8	4-8	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-8	4-8	0	Capped at max 12 realistic runs
simulation_result	4-8	4-8	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
OAK	Jack Perkins	pitcher / Starter	worse	predicted_IP -4
OAK	Hogan Harris	pitcher / Reliever	worse	predicted_H 3
OAK	Seth Johnson	pitcher / Reliever	better	predicted_H 2
OAK	Donovan Walton	batter / 2B	better	predicted_RBI 1.53
OAK	Carlos Cortes	batter / LF	worse	predicted_SO 1.477
OAK	Zack Gelof	batter / 3B	worse	predicted_RBI 1.41
TB	Liam Hicks	batter / C	better	predicted_RBI 1.35
TB	Jorge Mateo	batter / SS	worse	predicted_AB 1.333

Structural audit findings

Severity	Rule	Finding	Subject
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_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	OAK model pitcher projections differ from final simulation pitching box score (ER: current=7, sim=8).	OAK
info	CURRENT_FINALIZED_MATCH_BOX	OAK model-prediction batting totals differ from final simulation box score (AB: current=32, sim=33; H: current=8, sim=6; BB: current=3, sim=2; SO: current=8, sim=9).	OAK
info	CURRENT_PLAYER_MATCH_BOX	OAK current finalized hitter lines are model predictions and differ from the simulation box score: Henry Bolte: H current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=1 box=2; Jeff McNeil: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, SO current=0 box=1; Shea Langeliers: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Carlos Cortes: AB current=3 box=4, 2B current=0 box=1, SO current=1 box=2; Jonah Heim: AB current=3 box=4, H current=1 box=0; Alika Williams: AB current=3 box=2, H current=0 box=1, R current=0 box=1, BB current=0 box=1; Donovan Walton: AB current=3 box=4, 2B current=1 box=0, HR current=0 box=1, RBI current=0 box=2; Zack Gelof: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=2, SO current=1 box=0	OAK
info	CURRENT_FINALIZED_MATCH_BOX	TB model-prediction batting totals differ from final simulation box score (AB: current=32, sim=35; H: current=9, sim=11; 2B: current=1, sim=2; HR: current=1, sim=4; RBI: current=4, sim=8; R: current=5, sim=8; BB: current=3, sim=1; SO: current=7, sim=3).	TB
info	CURRENT_PLAYER_MATCH_BOX	TB current finalized hitter lines are model predictions and differ from the simulation box score: Junior Caminero: BB current=1 box=0; Yandy Díaz: H current=1 box=2, HR current=0 box=1, RBI current=1 box=2, SO current=1 box=0; Liam Hicks: AB current=3 box=4, H current=1 box=2, HR current=0 box=1, RBI current=1 box=2; Jonathan Aranda: BB current=1 box=0, SO current=1 box=0; Victor Mesa Jr.: AB current=4 box=3, H current=1 box=2, HR current=0 box=1, RBI current=0 box=2; Chandler Simpson: H current=1 box=2; Jorge Mateo: AB current=3 box=4, 2B current=0 box=1, R current=0 box=1, SO current=1 box=2; Richie Palacios: AB current=3 box=4, H current=1 box=0, R current=0 box=1, BB current=1 box=0	TB
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

PHI@WASMISS

Predicted: PHI 6-5 (conf -0.07, lean)Actual: WAS 3-6Brier 0.284total bias +2

Root cause: Starting-pitcher projection was wrong

Evidence-backed diagnosis

published system edge 0.07 (lean); this is not the half-scale distance from 50%
team-run absolute error 4; total bias +2
pregame home-minus-away evidence: season -0.317, recent -0.518, pitcher +0.098, composite -0.268
confidence pipeline reached 0.568 before later publication/recency controls; published home probability was 0.467
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

PHI offense over-projected by 3
Cristopher Sánchez (PHI) allowed 4 more ER than projected (5 vs 1)

Team	Starter	Projected	Actual	ER err
PHI	Cristopher Sánchez	1 ER / 19 outs	5 ER / 19 outs	4
WAS	Jackson Kent	3 ER / 13 outs	3 ER / 15 outs	0

Decision internals: run factors home×1.115 / away×1.016; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Luis Arraez (PHI 2B): 2-for-3, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.42]
Bryce Harper (PHI RF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Bryson Stott (PHI 3B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Dylan Crews (WAS RF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]
Trea Turner (PHI SS): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.64]

Starter miss: Cristopher Sánchez (PHI) posted 7.11 ERA vs projected 1.42 (5 ER in 19 outs); he is materially worse pitching away (3.68 ERA there vs 2.21 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

Cristopher Sánchez (PHI) threw 19 outs, 5 ER = 7.11 ERA against a projected 1.42 ERA (5.69 ERA error)

Bullpen behind him: 1 ER over 2 outs.

The HOME lineup he faced produced 5 runs on 8 hits and 1 HR, against a projected 5 runs / 8 hits (0 runs vs projection).

Who did the damage to him:

Brady House: 1-for-3, 0 HR, 0 RBI [+0.64 runs vs projection] season xwOBA 0.338, above average - real projection error
Andrew Pinckney: 1-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.326, above average - a good day, not a model error
Andrés Chaparro: 2-for-3, 1 HR, 1 RBI [+0.45 runs vs projection] season xwOBA 0.337, above average - a good day, not a model error

Verdict: both: the starter was hit hard AND the lineup underproduced

HOME pitching

Jackson Kent (WAS) threw 15 outs, 3 ER = 5.4 ERA against a projected 6.23 ERA (-0.83 ERA error) only 6 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 0 ER over 4 outs.

The AWAY lineup he faced produced 3 runs on 5 hits and 1 HR, against a projected 5 runs / 9 hits (-2 runs vs projection).

Who did the damage to him:

Luis Arraez: 2-for-3, 0 HR, 1 RBI [+1.42 runs vs projection] season xwOBA 0.303, below average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
PHI	Cristopher Sánchez (Starter)	19 / 19	7.11 / 1.42 (+5.69)	5 / 1	8 / 3	1 / 0	2 / 2	1.4	5 / 6	5.95	venue away: 3.68 vs 2.21 ERA (worse here) · opp run factor 1.11 (strong) · 30 prior appearances this season
PHI	Jonathan Bowlan (Setup Man)	2 / 3	13.5 / 0 (+13.50)	1 / 0	1 / 0	1 / 0	0 / 0	0.19	2 / 1	1.2
WAS	Jackson Kent (Starter)	15 / 13	5.4 / 6.23	3 / 3	4 / 5	1 / 0	3 / 3	3.3	4 / 4	3.62	venue home: 3.27 vs 8.82 ERA (ok) · opp run factor 1.02 (average) · 6 prior appearances this season
WAS	Brad Lord (Setup Man)	1 / 4	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.78	0 / 1	1.66
WAS	Erik Tolman (Closer)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.51	0 / 0	1.1

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Luis Arraez (2B)	PHI	3 / 4	2 / 1	2 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0 / 0	1.42	0.303 below average	good day
Bryce Harper (RF)	PHI	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.8	0.389 elite	bad day
Bryson Stott (3B)	PHI	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.8	0.327 above average	bad day
Dylan Crews (RF)	WAS	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.78	0.312 average	no signal
Trea Turner (SS)	PHI	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	-0.64	0.291 below average	projection too high
Brady House (3B)	WAS	3 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	1 / 1	0.64	0.338 above average	projection too low
Bryan De La Cruz (LF)	PHI	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.275 weak	—
J.T. Realmuto (C)	PHI	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	-0.47	0.323 above average	—
Derek Hill (CF)	PHI	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.292 below average	—
Jacob Young (CF)	WAS	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	-0.47	0.314 average	—
Andrew Pinckney (LF)	WAS	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	2 / 1	0.47	0.326 above average	—
Andrés Chaparro (DH)	WAS	3 / 4	2 / 1	1 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	0 / 1	0.45	0.337 above average	—
Harry Ford (C)	WAS	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	1 / 1	-0.33	0.289 below average	—
Yohandy Morales (1B)	WAS	3 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 1	1 / 1	0 / 0	1 / 1	0.31	0.332 above average	—
Alec Bohm (1B)	PHI	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	1 / 1	0 / 0	1 / 0	0 / 0	0.02	0.324 above average	—
Kyle Schwarber (DH)	PHI	3 / 4	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	2 / 1	1 / 1	0 / 1	0	0.365 elite	—
Jorbit Vivas (2B)	WAS	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	2 / 1	0	0.295 below average	—
Nasim Nuñez (SS)	WAS	1 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0	0.268 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.317	positive favors home; negative favors away
Recent form	-0.518	positive favors home; negative favors away
Starting-pitcher matchup	0.098	positive favors home; negative favors away
Composite	-0.268	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.9-3.5	n/a	Base ML historical model prediction
variance_applied	6-3	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-5	0.533	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	6-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-5	n/a	Capped at max 12 realistic runs
simulation_result	6-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	5.9-3.5	6-3	0.591	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-3	6-5	-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	6-5	6-5	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-5	6-5	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-5	6-5	-0	Capped at max 12 realistic runs
simulation_result	6-5	6-5	-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
WAS	Yohandy Morales	batter / 1B	worse	predicted_SO 3.111
WAS	Nasim Nuñez	batter / SS	better	predicted_RBI 2.722
PHI	Bryan De La Cruz	batter / LF	boosted	predicted_RBI 2.516
PHI	Bryce Harper	batter / RF	worse	predicted_BB 2.311
WAS	Jackson Kent	pitcher / Starter	worse	predicted_H 2
WAS	Luis Perales	pitcher / Middle Reliever (High Leverage)	better	predicted_H -2
PHI	Kyle Schwarber	batter / DH	better	predicted_2B 1.88
PHI	Trea Turner	batter / SS	better	predicted_H 1.77

Structural audit findings

Severity	Rule	Finding	Subject
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_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	PHI model-prediction batting totals differ from final simulation box score (AB: current=32, sim=35; H: current=9, sim=8; HR: current=1, sim=2; RBI: current=5, sim=6; R: current=5, sim=6; BB: current=4, sim=5; SO: current=7, sim=8).	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=2, H current=1 box=0, RBI current=1 box=0, BB current=1 box=3; Alec Bohm: 2B current=1 box=0, R current=0 box=1, SO current=0 box=2; Bryan De La Cruz: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=3, R current=0 box=1; Bryson Stott: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Luis Arraez: RBI current=0 box=1, R current=0 box=1, BB current=0 box=1; Trea Turner: AB current=4 box=5, H current=1 box=3, 2B current=1 box=0, HR current=0 box=1; Kyle Schwarber: H current=1 box=2, 2B current=0 box=2, HR current=1 box=0, RBI current=1 box=0; J.T. Realmuto: AB current=3 box=4, H current=1 box=0, R current=1 box=0	PHI
info	CURRENT_FINALIZED_MATCH_BOX	WAS model-prediction batting totals differ from final simulation box score (AB: current=31, sim=34; H: current=8, sim=7; 2B: current=1, sim=2; RBI: current=4, sim=5; SO: current=8, sim=9).	WAS
info	CURRENT_PLAYER_MATCH_BOX	WAS current finalized hitter lines are model predictions and differ from the simulation box score: Andrés Chaparro: AB current=4 box=3, HR current=1 box=0, SO current=1 box=2; Yohandy Morales: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=4; Jorbit Vivas: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, R current=0 box=1; Dylan Crews: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Brady House: H current=1 box=0, RBI current=1 box=0; Nasim Nuñez: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=3, R current=0 box=1; Jacob Young: AB current=3 box=4, H current=1 box=0; Harry Ford: RBI current=0 box=1	WAS
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

SD@COLMISS

Predicted: SD 9-5 (conf -0.16, moderate)Actual: COL 3-9Brier 0.336total bias +2

Root cause: Starting-pitcher projection was wrong

Evidence-backed diagnosis

published system edge 0.16 (moderate); this is not the half-scale distance from 50%
team-run absolute error 10; total bias +2
pregame home-minus-away evidence: season -1.479, recent -1.546, pitcher +0.005, composite -1.909
confidence pipeline reached 0.698 before later publication/recency controls; published home probability was 0.420
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

SD offense over-projected by 6
COL offense under-projected by 4
Kyle Freeland (COL) allowed 4 fewer ER than projected (3 vs 7)
committed to SD at 0.16 confidence and it lost

Team	Starter	Projected	Actual	ER err
SD	Walker Buehler	3 ER / 16 outs	2 ER / 14 outs	-1
COL	Kyle Freeland	7 ER / 14 outs	3 ER / 12 outs	-4

Decision internals: run factors home×1.027 / away×1.022; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Cole Carrigg (COL SS): 4-for-5, 2 HR, 5 RBI (projected 1.0 H, 0.0 HR) [run impact 2.94]
Troy Johnston (COL RF): 3-for-3, 0 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.56]
Austin Hays (SD DH): 2-for-3, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]
Luis Campusano (SD C): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.07]

Starter miss: Kyle Freeland (COL) posted 6.75 ERA vs projected 13.50 (3 ER in 12 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

What to do: Variance — no model change is warranted for this start.

Pitcher versus batter: what actually broke this game

AWAY pitching

Walker Buehler (SD) threw 14 outs, 2 ER = 3.86 ERA against a projected 5.06 ERA (-1.2 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 8 runs on 16 hits and 2 HR, against a projected 5 runs / 9 hits (3 runs vs projection).

Who did the damage to him:

Cole Carrigg: 4-for-5, 2 HR, 5 RBI [+2.94 runs vs projection] season xwOBA 0.309, average - a good day, not a model error
Troy Johnston: 3-for-3, 0 HR, 2 RBI [+1.56 runs vs projection] season xwOBA 0.323, above average - real projection error

Verdict: offense: the lineup scored well below its projection

HOME pitching

Kyle Freeland (COL) threw 12 outs, 3 ER = 6.75 ERA against a projected 13.5 ERA (-6.75 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 3 runs on 10 hits and 3 HR, against a projected 5 runs / 10 hits (-2 runs vs projection).

Who did the damage to him:

Austin Hays: 2-for-3, 1 HR, 1 RBI [+1.09 runs vs projection] season xwOBA 0.289, below average - a good day, not a model error
Luis Campusano: 2-for-4, 1 HR, 1 RBI [+1.07 runs vs projection] season xwOBA 0.332, above average - real projection error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
SD	Walker Buehler (Starter)	14 / 16	3.86 / 5.06	2 / 3	7 / 3	1 / 0	1 / 2	1.42	1 / 3	4.12	venue away: 5.23 vs 4.11 ERA (worse here) · opp run factor 1.03 (weak) · 29 prior appearances this season
COL	Kyle Freeland (Starter)	12 / 14	6.75 / 13.5 (-6.75)	3 / 7	8 / 11	3 / 1	0 / 2	1.42	1 / 5	3.72	venue home: 5.75 vs 6.96 ERA (ok) · opp run factor 1.02 (average) · 23 prior appearances this season

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Cole Carrigg (SS)	COL	5 / 4	4 / 1	1 / 1	0 / 0	2 / 0	5 / 1	2 / 1	0 / 1	1 / 1	2.94	0.309 average	no signal
Troy Johnston (RF)	COL	3 / 3	3 / 1	2 / 0	0 / 0	0 / 0	2 / 0	1 / 0	0 / 0	0 / 1	1.56	0.323 above average	projection too low
Austin Hays (DH)	SD	3 / 4	2 / 1	0 / 1	0 / 0	1 / 0	1 / 1	1 / 0	0 / 0	0 / 1	1.09	0.289 below average	good day
Luis Campusano (C)	SD	4 / 3	2 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 1	0 / 1	1.07	0.332 above average	projection too low
Ty France (1B)	SD	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 0	-0.47	0.336 above average	—
Samad Taylor (LF)	SD	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	-0.47	0.303 below average	—
Jackson Merrill (CF)	SD	4 / 4	2 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	0 / 1	-0.46	0.347 elite	—
Xander Bogaerts (SS)	SD	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 0	-0.33	0.319 average	—
Mickey Moniak (DH)	COL	4 / 4	1 / 1	0 / 0	1 / 0	0 / 1	0 / 1	1 / 1	0 / 0	1 / 1	-0.31	0.278 weak	—
Jake McCarthy (CF)	COL	5 / 4	2 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	0 / 1	0.16	0.324 above average	—
Fernando Tatis Jr. (RF)	SD	5 / 4	1 / 2	0 / 1	0 / 0	1 / 0	1 / 1	1 / 1	0 / 0	1 / 1	0.15	0.377 elite	—
TJ Rumfield (1B)	COL	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0.14	0.336 above average	—
Manny Machado (3B)	SD	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	0	0.335 above average	—
Jake Cronenworth (2B)	SD	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.268 weak	—
Adael Amador (2B)	COL	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	0 / 1	0	0.285 below average	—
Kyle Karros (3B)	COL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	0 / 1	0	0.337 above average	—
Zac Veen (LF)	COL	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0	0.293 below average	—
Brett Sullivan (C)	COL	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	1 / 0	0 / 0	0 / 0	0	0.282 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-1.479	positive favors home; negative favors away
Recent form	-1.546	positive favors home; negative favors away
Starting-pitcher matchup	0.005	positive favors home; negative favors away
Composite	-1.909	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.32-4.53	n/a	Base ML historical model prediction
variance_applied	5-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	8-5	0.58	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	9-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	9-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	9-5	n/a	Capped at max 12 realistic runs
simulation_result	9-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	5.32-4.53	5-5	-0.792	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-5	8-5	3	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	8-5	9-5	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	9-5	9-5	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	9-5	9-5	-0	Capped at max 12 realistic runs
simulation_result	9-5	9-5	-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
SD	Luis Campusano	batter / C	better	predicted_RBI 2.6
SD	Austin Hays	batter / DH	boosted	predicted_H 2.08
SD	Fernando Tatis Jr.	batter / RF	better	predicted_RBI 1.97
COL	Troy Johnston	batter / RF	worse	predicted_2B 1.75
COL	Zac Veen	batter / LF	better	predicted_BB 1.688
SD	Jackson Merrill	batter / CF	worse	predicted_AB -1.55
COL	Kyle Freeland	pitcher / Starter	worse	predicted_IP -1.467
SD	Jake Cronenworth	batter / 2B	better	predicted_AB 1.45

Structural audit findings

Severity	Rule	Finding	Subject
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_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_FINALIZED_MATCH_BOX	COL model pitcher projections differ from final simulation pitching box score (ER: current=8, sim=9).	COL
info	CURRENT_FINALIZED_MATCH_BOX	COL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=36; 2B: current=2, sim=5; HR: current=1, sim=0; BB: current=3, sim=4; SO: current=7, sim=6).	COL
info	CURRENT_PLAYER_MATCH_BOX	COL current finalized hitter lines are model predictions and differ from the simulation box score: Jake McCarthy: AB current=4 box=5, H current=1 box=2, 2B current=1 box=0, SO current=1 box=0; Cole Carrigg: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Mickey Moniak: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Adael Amador: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; TJ Rumfield: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, RBI current=0 box=1; Kyle Karros: 2B current=0 box=1, RBI current=0 box=1, BB current=1 box=0, SO current=1 box=2; Troy Johnston: H current=1 box=2, 2B current=0 box=2, BB current=0 box=1, SO current=1 box=0; Zac Veen: AB current=3 box=2, 2B current=0 box=1, RBI current=0 box=1, R current=0 box=1	COL
info	CURRENT_FINALIZED_MATCH_BOX	SD model-prediction batting totals differ from final simulation box score (AB: current=33, sim=41; H: current=10, sim=14; 2B: current=2, sim=6; RBI: current=5, sim=9; R: current=5, sim=9; BB: current=3, sim=5).	SD
info	CURRENT_PLAYER_MATCH_BOX	SD current finalized hitter lines are model predictions and differ from the simulation box score: Austin Hays: AB current=4 box=5, H current=1 box=3, R current=0 box=1; Ty France: H current=1 box=2, 2B current=0 box=1, R current=0 box=1, BB current=0 box=1; Manny Machado: RBI current=1 box=0, SO current=1 box=0; Fernando Tatis Jr.: AB current=4 box=6, H current=2 box=1, RBI current=1 box=3; Luis Campusano: AB current=3 box=5, H current=1 box=2, 2B current=0 box=1, HR current=0 box=1; Xander Bogaerts: SO current=0 box=1; Jackson Merrill: HR current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Samad Taylor: AB current=3 box=4, 2B current=0 box=1, RBI current=0 box=1, BB current=0 box=1	SD
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

SEA@LAAMISS

Predicted: SEA 5-4 (conf -0.06, lean)Actual: LAA 1-2Brier 0.281total bias +6

Root cause: Starting-pitcher projection was wrong

Evidence-backed diagnosis

published system edge 0.06 (lean); this is not the half-scale distance from 50%
team-run absolute error 6; total bias +6
pregame home-minus-away evidence: season -0.177, recent -0.166, pitcher -0.098, composite -0.534
confidence pipeline reached 0.569 before later publication/recency controls; published home probability was 0.470
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 6 (9 vs 3)
SEA offense over-projected by 4
Ryan Johnson (LAA) allowed 4 fewer ER than projected (1 vs 5)

Team	Starter	Projected	Actual	ER err
SEA	Logan Gilbert	1 ER / 17 outs	2 ER / 18 outs	1
LAA	Ryan Johnson	5 ER / 16 outs	1 ER / 18 outs	-4

Decision internals: run factors home×0.999 / away×1.097; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Josh Lowe (LAA RF): 2-for-3, 1 HR, 1 RBI (projected 0.0 H, 0.0 HR) [run impact 1.87]
Zach Neto (LAA SS): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
J.P. Crawford (SEA SS): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Mike Trout (LAA CF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]
Dominic Canzone (SEA DH): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.64]

Starter miss: Ryan Johnson (LAA) posted 1.50 ERA vs projected 8.44 (1 ER in 18 outs); he faced a strong offense (opponent run factor 1.10, bucket weak).

What to do: Consider raising context.opponent_strength_weight; validate across the rolling window before promotion.

Pitcher versus batter: what actually broke this game

AWAY pitching

Logan Gilbert (SEA) threw 18 outs, 2 ER = 3 ERA against a projected 1.59 ERA (1.41 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 2 runs on 7 hits and 1 HR, against a projected 3 runs / 7 hits (-1 runs vs projection).

Who did the damage to him:

Josh Lowe: 2-for-3, 1 HR, 1 RBI [+1.87 runs vs projection] season xwOBA 0.286, below average - a good day, not a model error
Moisés Ballesteros: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.293, below average - a good day, not a model error
Vaughn Grissom: 1-for-3, 0 HR, 1 RBI [+0.31 runs vs projection] season xwOBA 0.308, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

HOME pitching

Ryan Johnson (LAA) threw 18 outs, 1 ER = 1.5 ERA against a projected 8.44 ERA (-6.94 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The AWAY lineup he faced produced 1 runs on 7 hits and 0 HR, against a projected 5 runs / 7 hits (-4 runs vs projection).

Who did the damage to him:

Josh Naylor: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.315, average - a good day, not a model error
Julio Rodríguez: 1-for-4, 0 HR, 1 RBI [+0.31 runs vs projection] season xwOBA 0.336, above average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
SEA	Logan Gilbert (Starter)	18 / 17	3 / 1.59	2 / 1	6 / 5	1 / 0	2 / 2	2.05	7 / 8	7.07	venue away: 2.7 vs 4.41 ERA (ok) · opp run factor 1 (weak) · 29 prior appearances this season
LAA	Ryan Johnson (Starter)	18 / 16	1.5 / 8.44 (-6.94)	1 / 5	6 / 7	0 / 2	0 / 2	2.15	7 / 6	4.16	venue home: 3.18 vs 5.56 ERA (ok) · opp run factor 1.1 (weak) · 19 prior appearances this season
LAA	Ryan Watson (Setup Man)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.2	0 / 1	0.99
LAA	Tayler Saucedo (Middle Reliever (High Leverage))	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.15	1 / 1	0.78

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Josh Lowe (RF)	LAA	3 / 3	2 / 0	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	1 / 1	1.87	0.286 below average	good day
Zach Neto (SS)	LAA	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	1 / 2	-1.4	0.31 average	no signal
J.P. Crawford (SS)	SEA	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	-0.8	0.322 above average	bad day
Mike Trout (CF)	LAA	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	-0.78	0.378 elite	bad day
Dominic Canzone (DH)	SEA	4 / 4	1 / 1	0 / 0	1 / 0	0 / 1	0 / 1	1 / 1	0 / 1	0 / 1	-0.64	0.351 elite	bad day
Randy Arozarena (LF)	SEA	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.64	0.347 elite	bad day
Josh Naylor (1B)	SEA	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0.47	0.315 average	—
Moisés Ballesteros (DH)	LAA	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	0.47	0.293 below average	—
Denzer Guzman (3B)	LAA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	3 / 1	-0.47	0.267 weak	—
Christian Moore (2B)	LAA	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 2	-0.47	0.236 weak	—
Cal Raleigh (C)	SEA	4 / 4	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.33	0.314 average	—
Julio Rodríguez (CF)	SEA	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	1 / 1	0.31	0.336 above average	—
Vaughn Grissom (1B)	LAA	3 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	0 / 1	0.31	0.308 average	—
Wade Meckler (LF)	LAA	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	1 / 1	0.14	0.316 average	—
Brock Rodden (3B)	SEA	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.227 weak	—
Cole Young (2B)	SEA	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0	0.315 average	—
Lazaro Montes (RF)	SEA	2 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	0	0.365 elite	—
Tyler Heineman (C)	LAA	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.241 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.177	positive favors home; negative favors away
Recent form	-0.166	positive favors home; negative favors away
Starting-pitcher matchup	-0.098	positive favors home; negative favors away
Composite	-0.534	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.92-2.85	n/a	Base ML historical model prediction
variance_applied	5-3	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	4.92-2.85	5-3	-0.067	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	5-4	-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	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
LAA	Zach Neto	batter / SS	better	predicted_RBI 3.3
LAA	Josh Lowe	batter / RF	worse	predicted_SO 2.062
SEA	Brock Rodden	batter / 3B	better	predicted_RBI 1.8
SEA	Cal Raleigh	batter / C	worse	predicted_AB -1.58
SEA	Julio Rodríguez	batter / CF	worse	predicted_RBI 1.42
LAA	Tyler Heineman	batter / C	worse	predicted_AB 1.311
LAA	Christian Moore	batter / 2B	worse	predicted_SO -1.259
SEA	Dominic Canzone	batter / DH	worse	predicted_H -1.24

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	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_FINALIZED_MATCH_BOX	LAA 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; R: current=3, sim=4; BB: current=3, sim=2; SO: current=11, sim=10).	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=3, 2B current=0 box=1, RBI current=1 box=0, R current=0 box=1; Wade Meckler: AB current=4 box=3, H current=1 box=0, R current=1 box=0; Moisés Ballesteros: H current=1 box=0, RBI current=1 box=0; Josh Lowe: AB current=3 box=4, SO current=1 box=3; Denzer Guzman: R current=0 box=1; Mike Trout: H current=1 box=2, 2B current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Zach Neto: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, RBI current=1 box=4; Christian Moore: AB current=3 box=4, SO current=2 box=0	LAA
info	CURRENT_FINALIZED_MATCH_BOX	SEA model-prediction batting totals differ from final simulation box score (2B: current=1, sim=2; HR: current=1, sim=2; RBI: current=4, sim=5; BB: current=4, sim=2).	SEA
info	CURRENT_PLAYER_MATCH_BOX	SEA current finalized hitter lines are model predictions and differ from the simulation box score: Cal Raleigh: AB current=4 box=3, RBI current=1 box=0, R current=1 box=0, SO current=1 box=2; Dominic Canzone: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Randy Arozarena: 2B current=1 box=0, HR current=0 box=1, BB current=1 box=0, SO current=1 box=0; J.P. Crawford: R current=0 box=1, BB current=1 box=0; Julio Rodríguez: HR current=0 box=1, RBI current=0 box=2, SO current=1 box=2; Brock Rodden: AB current=3 box=4, H current=0 box=1, RBI current=0 box=2, R current=0 box=1; Josh Naylor: H current=1 box=2, 2B current=0 box=1, R current=0 box=1; Cole Young: 2B current=0 box=1, R current=1 box=0, SO current=1 box=0	SEA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

SF@STLMISS

Predicted: STL 4-5 (conf +0.11, moderate)Actual: SF 10-3Brier 0.306total bias -4

Root cause: Backed the wrong side with real conviction

Evidence-backed diagnosis

published system edge 0.11 (moderate); this is not the half-scale distance from 50%
team-run absolute error 8; total bias -4
pregame home-minus-away evidence: season +0.469, recent +0.041, pitcher +0.057, composite +0.574
confidence pipeline reached 0.574 before later publication/recency controls; published home probability was 0.553
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs under-projected by 4 (9 vs 13)
SF offense under-projected by 6
committed to STL at 0.11 confidence and it lost

Team	Starter	Projected	Actual	ER err
SF	Blade Tidwell	4 ER / 15 outs	2 ER / 20 outs	-2
STL	Andre Pallante	4 ER / 17 outs	6 ER / 15 outs	2

Decision internals: run factors home×0.985 / away×0.928; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Turner Hill (SF LF): 2-for-4, 1 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Jordan Walker (STL RF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Leonardo Bernal (STL C): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Christian Koss (SF SS): 3-for-5, 0 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 0.94]
Iván Herrera (STL DH): 0-for-2, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Starter miss: Blade Tidwell (SF) posted 2.70 ERA vs projected 7.20 (2 ER in 20 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

What to do: Variance — no model change is warranted for this start.

Pitcher versus batter: what actually broke this game

AWAY pitching

Blade Tidwell (SF) threw 20 outs, 2 ER = 2.7 ERA against a projected 7.2 ERA (-4.5 ERA error)

Bullpen behind him: 0 ER over 3 outs.

The HOME lineup he faced produced 3 runs on 4 hits and 1 HR, against a projected 4 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

José Fermín: 1-for-2, 0 HR, 0 RBI [+0.78 runs vs projection] season xwOBA 0.289, below average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Andre Pallante (STL) threw 15 outs, 6 ER = 10.8 ERA against a projected 6.35 ERA (4.45 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 9 runs on 11 hits and 2 HR, against a projected 4 runs / 8 hits (5 runs vs projection).

Who did the damage to him:

Turner Hill: 2-for-4, 1 HR, 3 RBI [+1.4 runs vs projection] season xwOBA 0.318, average - a good day, not a model error
Christian Koss: 3-for-5, 0 HR, 2 RBI [+0.94 runs vs projection] season xwOBA 0.251, weak - a good day, not a model error
Bryce Eldridge: 2-for-4, 1 HR, 3 RBI [+0.78 runs vs projection] season xwOBA 0.354, elite - real projection error

Verdict: pitching: the starter gave up runs at a far worse rate than projected

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
SF	Blade Tidwell (Starter)	20 / 15	2.7 / 7.2 (-4.50)	2 / 4	3 / 1	1 / 0	1 / 5	2.74	4 / 5	4.07	venue away: 2.3 vs 8.05 ERA (ok) · opp run factor 0.98 (average) · 15 prior appearances this season
SF	Dylan Smith (Closer)	3 / 4	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0.49	0 / 1	1.14
STL	Andre Pallante (Starter)	15 / 17	10.8 / 6.35 (+4.45)	6 / 4	8 / 5	1 / 2	2 / 2	1.92	2 / 5	3.24	venue home: 4.02 vs 2.75 ERA (worse here) · opp run factor 0.93 (weak) · 26 prior appearances this season

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Turner Hill (LF)	SF	4 / 3	2 / 1	0 / 0	0 / 0	1 / 0	3 / 1	1 / 0	1 / 1	0 / 1	1.4	0.318 average	no signal
Jordan Walker (RF)	STL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-1.4	0.343 elite	bad day
Leonardo Bernal (C)	STL	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	-1.11	0.405 elite	bad day
Christian Koss (SS)	SF	5 / 3	3 / 1	0 / 0	0 / 0	0 / 0	2 / 0	0 / 1	0 / 0	0 / 1	0.94	0.251 weak	good day
Iván Herrera (DH)	STL	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	1 / 1	-0.8	0.332 above average	bad day
Bryce Eldridge (1B)	SF	4 / 4	2 / 1	1 / 0	0 / 0	1 / 1	3 / 1	2 / 1	1 / 1	1 / 1	0.78	0.354 elite	projection too low
Drew Gilbert (CF)	SF	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	1 / 0	-0.78	0.329 above average	bad day
José Fermín (3B)	STL	2 / 3	1 / 0	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.78	0.289 below average	good day
Osleivis Basabe (3B)	SF	3 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0 / 1	0.64	0.276 weak	good day
Grant McCray (RF)	SF	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 0	0 / 1	0.47	0.308 average	—
Drew Cavanaugh (C)	SF	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	3 / 1	-0.47	0.317 average	—
Alec Burleson (1B)	STL	3 / 4	1 / 1	0 / 1	0 / 0	1 / 0	1 / 1	1 / 1	0 / 1	2 / 0	0.29	0.376 elite	—
Bryan Torres (LF)	STL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	-0.14	0.294 below average	—
Nathan Church (CF)	STL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	-0.14	0.289 below average	—
Jonah Cox (2B)	SF	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	0 / 0	2 / 1	0	0.261 weak	—
Jung Hoo Lee (DH)	SF	5 / 4	1 / 1	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0	0.317 average	—
Thomas Saggese (2B)	STL	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.315 average	—
Masyn Winn (SS)	STL	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0	0.299 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.469	positive favors home; negative favors away
Recent form	0.041	positive favors home; negative favors away
Starting-pitcher matchup	0.057	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	3.95-4.78	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.553	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	3.95-4.78	4-5	0.176	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
STL	Iván Herrera	batter / DH	worse	predicted_AB -2.38
STL	Leonardo Bernal	batter / C	better	predicted_AB -2.19
STL	Masyn Winn	batter / SS	worse	predicted_AB -1.615
SF	Bryce Eldridge	batter / 1B	better	predicted_RBI 1.481
STL	Nathan Church	batter / CF	worse	predicted_SO 1.373
SF	Turner Hill	batter / LF	worse	predicted_SO 1.292
SF	Osleivis Basabe	batter / 3B	worse	predicted_SO 1.286
SF	Grant McCray	batter / RF	better	predicted_AB 1.28

Structural audit findings

Severity	Rule	Finding	Subject
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_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	SF model-prediction batting totals differ from final simulation box score (AB: current=32, sim=33; H: current=8, sim=6; 2B: current=2, sim=0; HR: current=1, sim=2; SO: current=7, sim=9).	SF
info	CURRENT_PLAYER_MATCH_BOX	SF current finalized hitter lines are model predictions and differ from the simulation box score: Bryce Eldridge: RBI current=1 box=2, BB current=1 box=0, SO current=1 box=2; Jonah Cox: H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Jung Hoo Lee: H current=1 box=2, 2B current=1 box=0, R current=0 box=1; Osleivis Basabe: H current=1 box=0, SO current=1 box=2; Turner Hill: H current=1 box=0, RBI current=1 box=0, SO current=1 box=2; Grant McCray: AB current=3 box=4, H current=0 box=1, HR current=0 box=1, RBI current=0 box=1; Drew Cavanaugh: H current=1 box=0, SO current=1 box=0; Christian Koss: AB current=3 box=4, R current=1 box=0	SF
info	CURRENT_FINALIZED_MATCH_BOX	STL model-prediction batting totals differ from final simulation box score (AB: current=31, sim=27; H: current=8, sim=3; 2B: current=2, sim=1; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=5; SO: current=6, sim=8; HBP: current=0, sim=2).	STL
info	CURRENT_PLAYER_MATCH_BOX	STL current finalized hitter lines are model predictions and differ from the simulation box score: Bryan Torres: H current=1 box=0, BB current=0 box=1, SO current=1 box=0; Iván Herrera: AB current=4 box=2, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Thomas Saggese: H current=1 box=0, RBI current=0 box=1, R current=0 box=1, BB current=0 box=1; Leonardo Bernal: AB current=4 box=3, 2B current=1 box=0, HR current=0 box=1, SO current=1 box=0; Jordan Walker: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; José Fermín: H current=0 box=1, 2B current=0 box=1, SO current=0 box=1; Alec Burleson: 2B current=1 box=0, BB current=1 box=0, SO current=0 box=2; Nathan Church: H current=1 box=0, RBI current=0 box=1, R current=0 box=1, SO current=0 box=2	STL
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.

Day over day: did it improve or worsen

Compared against packet 20260914. Last recorded change: v4.30.0 (2026-09-08) Hitter shrinkage is derived from the between-hitter spread rather than fitted to slate error, so an established extreme profile is no longer flattened toward the league.

Metric	Prior	Current	Δ	Result
brier_score	0.2411	0.2426	+0.001416	regressed
accuracy	0.6333	0.6	-0.0333	regressed
team_run_mae	2.5083	2.5542	+0.0459	regressed
total_run_mae	3.8167	3.7917	-0.025	improved
calibration_error	0.1244	0.084	-0.0404	improved
starter_er_mae	1.7778	1.7739	-0.0039	improved
starter_outs_mae	3.2644	3.2874	+0.023	regressed

Deep slate report — 20260915

Process, end to end

Slate summary

What drove the losses

Source coverage receipt

High-level causal issue register

Cross-slate trends (persistent-pattern memory)

Escalated persistent patterns

Finding streaks across the window

Metric direction across the window

Pipeline stage regression

Probability lineage and calibration

Score distribution and margin compression

Per-game evidence coverage

Every game, in depth (15)

ATL@CHCMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

BAL@NYMMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

BOS@TEXMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

CHW@CLEMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

DET@TORHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

KC@HOUHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

LAD@CINHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

MIA@ARIMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

MIL@PITHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

NYY@MINHIT

Player-level breakdown (projected vs actual)