Deep slate report 20260924

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_20260924.json	301958
forensic deep audit	loaded	probability, provenance, starter and decision contracts	/app/output/reports/PREDICTION_SYSTEM_DEEP_AUDIT_20260924.json	196012
continuous learning	loaded	pipeline-stage quality, calibration, probability lineage, score distribution, and official starter comparisons	/app/output/learning/continuous_learning_daily_20260924.json	2023270
detailed report	loaded	full game and simulation rendering	/app/output/reports/detailed_20260924.html	285081
ultra detailed player audit	loaded	player-level prediction trace	/app/output/reports/20260924_ultra_detailed_player_audit.html	25971500
structural audit	loaded	all invariant findings	/app/output/reports/audit_report_20260924.html	1530211
integrity revision audit	missing	post-Luna remediation proof for immutable revision artifacts	/app/output/reports/audit_report_20260924_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 (12)

Severity	Domain	Finding and evidence	Affected	Action & guardrail
HIGH	pipeline_stages	An intermediate stage materially degraded winner direction historical_ml → variance_applied changed winner accuracy by -0.2500 (-3 hits)	CHW@KC, MIA@CHC, NYM@TEX	Do thisList the exact games whose winner flipped at this stage, and keep both the before-stage and after-stage picks in the trace. Don't concludeEven if a later stage recovers the pick, this unstable step is still a risk; do not treat it as harmless.
HIGH	calibration	Published confidence buckets were poorly aligned on this slate weighted absolute calibration error 0.1701 across 12 games	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@OAK, LAA@SEA, MIA@CHC, MIL@PHI, NYM@TEX, SD@LAD, STL@PIT, TB@NYY	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 = 0.0040/0.0349/0.0031	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@OAK, LAA@SEA, MIA@CHC, MIL@PHI, NYM@TEX, SD@LAD, STL@PIT, TB@NYY	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 12/12 games structurally audited	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@OAK, LAA@SEA, MIA@CHC, MIL@PHI, NYM@TEX, SD@LAD, STL@PIT, TB@NYY	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 5 material movement(s), all explained by named score-coherence and/or recency transformations	ARI@COL, CHW@KC, CIN@ATL, HOU@OAK, LAA@SEA	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	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@OAK, LAA@SEA, MIA@CHC, MIL@PHI, NYM@TEX, SD@LAD, STL@PIT, TB@NYY	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	3/11	no	param
pipeline_winner_instability	1	6/11	yes	code/diagnostic
simulation_inert_on_slate	11	11/11	yes	code/diagnostic
single_slate_calibration_instability	1	8/11	yes	param
weak_run_model_signal	11	11/11	yes	code/diagnostic

Metric	Direction	Series (old → new)
winner_accuracy	improving	[0.7, 0.4, 0.6667, 0.4444, 0.6667, 0.7333, 0.6, 0.3333, 0.4, 0.5, 0.6667]
weighted_calibration_error	improving	[0.2812, 0.2795, 0.1661, 0.1016, 0.1247, 0.3458, 0.3184, 0.2028, 0.1881, 0.1108, 0.1701]
margin_bias	improving	[-0.4, -1.667, -1.133, -2.111, -2.0, -0.667, -1.533, -2.667, -1.867, -1.188, -1.083]
team_run_error_on_losses	improving	[6.333, 5.556, 6.4, 3.6, 5.2, 6.5, 6.5, 3.0, 6.222, 5.75, 3.5]

Stage	Winner hits	Accuracy	Team-run MAE	Total-run MAE	Margin MAE
historical_ml	9/12	75%	2.57	4.25	2.213
variance_applied	6/12	50%	2.5	4.167	2.167
enhanced_prediction	8/12	66.7%	2.167	3.833	2
embedded_savant_run_environment	8/12	66.7%	2.208	3.75	2.083
fundamental_edge_target_review	8/12	66.7%	2.208	3.75	2.083
capped_target	8/12	66.7%	2.208	3.75	2.083
simulation_result	8/12	66.7%	2.208	3.75	2.083
published_final	8/12	66.7%	2.208	3.75	2.083

Transition	Δ accuracy	Δ hits	Δ team-run MAE	Δ margin MAE	Games whose winner/tie changed
historical_ml → variance_applied	-0.25	-3	-0.0701	-0.0464	CHW@KC: CHW → TIE; MIA@CHC: CHC → TIE; NYM@TEX: TEX → TIE
variance_applied → enhanced_prediction	0.1667	2	-0.3333	-0.1667	CHW@KC: TIE → CHW; MIA@CHC: TIE → CHC; NYM@TEX: TIE → NYM
enhanced_prediction → embedded_savant_run_environment	0	0	0.0416	0.0833
embedded_savant_run_environment → fundamental_edge_target_review	0	0	0	0
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
ARI@COL	hit	0.716	0.661	0.661	0.6175	0.56	score coherence followed by matchup-recency guardrail
CHW@KC	hit	0.5873	0.62	0.62	0.62	0.5446	score coherence followed by matchup-recency guardrail
CIN@ATL	miss	0.6478	0.5973	0.5973	0.5857	0.56	score coherence followed by matchup-recency guardrail
CLE@BOS	hit	0.5	0.5073	0.5073	0.5073	0.5062	published probability matches the calibrated/governed value
HOU@OAK	hit	0.6661	0.6344	0.6344	0.6344	0.56	score coherence followed by matchup-recency guardrail
LAA@SEA	miss	0.6618	0.6366	0.6366	0.6053	0.56	score coherence followed by matchup-recency guardrail
MIA@CHC	hit	0.5425	0.5263	0.5263	0.5263	0.519	score coherence followed by matchup-recency guardrail
MIL@PHI	hit	0.5639	0.5543	0.5543	0.5543	0.5392	score coherence followed by matchup-recency guardrail
NYM@TEX	miss	0.5029	0.5246	0.5246	0.5246	0.5209	published probability matches the calibrated/governed value
SD@LAD	miss	0.5283	0.5186	0.5186	0.5186	0.5098	score coherence followed by matchup-recency guardrail
STL@PIT	hit	0.5644	0.5381	0.5381	0.5381	0.5275	score coherence followed by matchup-recency guardrail
TB@NYY	hit	0.5616	0.5366	0.5366	0.5366	0.5311	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
24	4	11	1	8	1.68	-0.67

Game	Trained p(home)	Blended p(home)	Blend favoured	Published p(home)	Published side	Artifact says winner changed	Outcome
CLE@BOS	0.522	0.507	home	0.494	away	yes	hit

Slate	Games	Winner acc	Mean starter ERA error	Mean lineup runs vs proj	Sides broke on pitching	Sides broke on offense
20260923	16	0.500	-0.49	-0.91	4	11
20260922	15	0.400	-0.32	-0.97	3	11
20260921	3	0.333	-1.08	-0.50
20260920	15	0.600	-0.89	-1.00	2	10
20260919	15	0.733	1.36	-0.10	8	7
20260918	15	0.667	3.95	0.30	6	5
20260917	9	0.444	2.16	-0.22	2	3
20260916	15	0.667	2.09	-0.23	4	5
20260915	15	0.400	-1.58	-0.70	1	12
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

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

ARI@COLHIT

Predicted: ARI 6-5 (conf -0.12, moderate)Actual: ARI 12-8Brier 0.194total bias -9

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 9; total bias -9
pregame home-minus-away evidence: season -1.157, recent -0.666, pitcher -0.153, composite -1.320
confidence pipeline reached 0.661 before later publication/recency controls; published home probability was 0.440
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 9 (11 vs 20)
ARI offense under-projected by 6
COL offense under-projected by 3
Tanner Gordon (COL) allowed 7 more ER than projected (10 vs 3)

Team	Starter	Projected	Actual	ER err
ARI	Eduardo Rodriguez	3 ER / 17 outs	2 ER / 18 outs	-1
COL	Tanner Gordon	3 ER / 13 outs	10 ER / 10 outs	7

Decision internals: run factors home×0.928 / away×1.081; 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 CF): 4-for-5, 1 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 2.03]
Geraldo Perdomo (ARI SS): 3-for-5, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.56]
Pavin Smith (ARI 1B): 2-for-3, 0 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.11]
Kyle Karros (COL 3B): 2-for-5, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.07]
Nolan Arenado (ARI 3B): 3-for-5, 1 HR, 3 RBI (projected 1.0 H, 1.0 HR) [run impact 0.94]

Starter miss: Tanner Gordon (COL) posted 27.00 ERA vs projected 6.23 (10 ER in 10 outs); he is materially worse pitching home (9.29 ERA there vs 3.00 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

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

No settled pitching line for this club. The forecast named Eduardo Rodriguez (ARI) at a projected 4.76 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 13 hits and 2 HR, against a projected 4 runs / 9 hits (1 runs vs projection).

Who did the damage to him:

Cole Carrigg: 4-for-5, 1 HR, 3 RBI [+2.03 runs vs projection] season xwOBA 0.31, average - a good day, not a model error
Kyle Karros: 2-for-5, 1 HR, 1 RBI [+1.07 runs vs projection] season xwOBA 0.337, above average - real projection error
TJ Rumfield: 2-for-4, 0 HR, 1 RBI [+0.78 runs vs projection] season xwOBA 0.333, above average - real projection error

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

HOME pitching

Tanner Gordon (COL) threw 10 outs, 10 ER = 27 ERA against a projected 6.23 ERA (20.77 ERA error)

Bullpen behind him: 2 ER over 14 outs.

The AWAY lineup he faced produced 11 runs on 14 hits and 2 HR, against a projected 4 runs / 9 hits (7 runs vs projection).

Who did the damage to him:

Geraldo Perdomo: 3-for-5, 0 HR, 1 RBI [+1.56 runs vs projection] season xwOBA 0.326, above average - real projection error
Pavin Smith: 2-for-3, 0 HR, 2 RBI [+1.11 runs vs projection] season xwOBA 0.279, weak - a good day, not a model error
Nolan Arenado: 3-for-5, 1 HR, 3 RBI [+0.94 runs vs projection] season xwOBA 0.296, below average - a good day, not a model error

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

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)
COL	Tanner Gordon (Starter)	10 / 13	27 / 6.23 (+20.77)	10 / 3	12 / 7	2 / 1	4 / 1	1.03	1 / 1	3.77	venue home: 9.29 vs 3 ERA (worse here) · opp run factor 1.08 (average) · 23 prior appearances this season
COL	Jimmy Herget (Middle Reliever (High Leverage))	8 / 4	6.75 / 0 (+6.75)	2 / 0	4 / 0	1 / 0	1 / 0	0.59	4 / 0	1.01
COL	Jordan Romano (Closer)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.54	1 / 3	1.17
COL	Nick Frasso (Reliever)	3 / 1	0 / 81 (-81.00)	0 / 3	1 / 2	0 / 0	0 / 1	n/a	0 / 1	n/a

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 (CF)	COL	5 / 4	4 / 1	0 / 1	0 / 0	1 / 0	3 / 1	1 / 1	0 / 0	0 / 1	2.03	0.31 average	no signal
Geraldo Perdomo (SS)	ARI	5 / 4	3 / 1	0 / 0	1 / 0	0 / 0	1 / 0	3 / 0	1 / 1	1 / 0	1.56	0.326 above average	projection too low
Pavin Smith (1B)	ARI	3 / 3	2 / 1	1 / 0	0 / 0	0 / 0	2 / 0	2 / 0	1 / 0	0 / 1	1.11	0.279 weak	good day
Kyle Karros (3B)	COL	5 / 4	2 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 1	1 / 1	1.07	0.337 above average	projection too low
Nolan Arenado (3B)	ARI	5 / 4	3 / 1	0 / 0	0 / 0	1 / 1	3 / 1	1 / 1	0 / 0	1 / 1	0.94	0.296 below average	good day
Hunter Goodman (DH)	COL	5 / 4	1 / 1	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	0 / 0	0 / 1	-0.93	0.32 average	no signal
TJ Rumfield (1B)	COL	4 / 4	2 / 1	1 / 0	0 / 0	0 / 0	1 / 1	2 / 0	1 / 1	0 / 0	0.78	0.333 above average	projection too low
Ketel Marte (2B)	ARI	2 / 4	1 / 1	1 / 1	0 / 0	0 / 0	1 / 1	2 / 1	2 / 0	0 / 0	0.66	0.342 elite	projection too low
Jake McCarthy (LF)	COL	3 / 4	1 / 1	0 / 0	0 / 1	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.62	0.325 above average	bad day
Lars Nootbaar (LF)	ARI	6 / 4	3 / 1	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	0 / 1	0 / 1	0.61	0.338 above average	projection too low
Tim Tawa (CF)	ARI	5 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.47	0.312 average	—
Brett Sullivan (C)	COL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	0 / 0	-0.47	0.284 below average	—
Ezequiel Tovar (SS)	COL	5 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	-0.47	0.28 weak	—
Corbin Carroll (RF)	ARI	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 1	0.33	0.345 elite	—
Gabriel Moreno (DH)	ARI	6 / 4	1 / 2	0 / 1	0 / 0	1 / 0	3 / 1	1 / 1	0 / 1	1 / 0	-0.18	0.362 elite	—
Connor Norby (2B)	COL	3 / 4	2 / 1	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0.14	0.304 average	—
Jordan Beck (RF)	COL	5 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	4 / 1	0	0.248 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-1.157	positive favors home; negative favors away
Recent form	-0.666	positive favors home; negative favors away
Starting-pitcher matchup	-0.153	positive favors home; negative favors away
Composite	-1.32	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.64-3.67	n/a	Base ML historical model prediction
variance_applied	6-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-5	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	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	5.64-3.67	6-4	0.03	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-4	6-5	-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	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
COL	Nick Frasso	pitcher / Reliever	worse	predicted_ER 3
COL	Connor Norby	batter / 2B	better	predicted_H 2.76
COL	TJ Rumfield	batter / 1B	boosted	predicted_SO 2.43
ARI	Pavin Smith	batter / 1B	worse	predicted_AB 2.229
COL	Kyle Karros	batter / 3B	worse	predicted_H 2.14
ARI	Lars Nootbaar	batter / LF	worse	predicted_AB 1.771
ARI	Tim Tawa	batter / CF	better	predicted_RBI 1.549
COL	Jake McCarthy	batter / LF	worse	predicted_H -1.42

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	COL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	COL
info	CURRENT_FINALIZED_MATCH_BOX	ARI model pitcher projections differ from final simulation pitching box score (H: current=13, sim=12).	ARI
info	CURRENT_FINALIZED_MATCH_BOX	ARI model-prediction batting totals differ from final simulation box score (AB: current=33, sim=37; H: current=10, sim=11; 2B: current=2, sim=1; 3B: current=0, sim=2; RBI: current=5, sim=6; R: current=4, sim=6; SO: current=6, sim=8; SF: current=0, sim=1).	ARI
info	CURRENT_PLAYER_MATCH_BOX	ARI current finalized hitter lines are model predictions and differ from the simulation box score: Lars Nootbaar: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, R current=0 box=1; Gabriel Moreno: H current=2 box=1, 2B current=1 box=0, RBI current=1 box=0, SO current=0 box=1; Pavin Smith: AB current=3 box=5, SO current=1 box=2; Ketel Marte: AB current=4 box=5, H current=1 box=2, 2B current=1 box=0, 3B current=0 box=1; Nolan Arenado: HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Tim Tawa: HR current=0 box=1, RBI current=1 box=2, R current=0 box=1, BB current=0 box=1; Geraldo Perdomo: AB current=4 box=3, RBI current=0 box=1, R current=0 box=1, SF current=0 box=1; Corbin Carroll: H current=1 box=2, 3B current=0 box=1, RBI current=0 box=2, SO current=1 box=0	ARI
info	CURRENT_FINALIZED_MATCH_BOX	COL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=38; H: current=9, sim=12; 2B: current=2, sim=1; 3B: current=1, sim=0; HR: current=1, sim=0; R: current=4, sim=5; BB: current=3, sim=1; SO: current=7, sim=11).	COL
info	CURRENT_PLAYER_MATCH_BOX	COL current finalized hitter lines are model predictions and differ from the simulation box score: Hunter Goodman: H current=1 box=0, HR current=1 box=0, SO current=1 box=2, SF current=0 box=1; Cole Carrigg: 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=2; TJ Rumfield: RBI current=1 box=0, R current=0 box=1, SO current=0 box=3; Jake McCarthy: AB current=4 box=5, H current=1 box=0, 3B current=1 box=0, RBI current=1 box=0; Jordan Beck: AB current=3 box=4; Kyle Karros: H current=1 box=3, RBI current=0 box=1, R current=0 box=1, BB current=1 box=0; Connor Norby: AB current=4 box=5, H current=1 box=3, RBI current=0 box=2, R current=0 box=1; Brett Sullivan: AB current=3 box=4, H current=1 box=2, R current=0 box=1, SO current=0 box=1	COL
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

Team	Starter	Projected	Actual	ER err
CHW	David Sandlin	3 ER / 15 outs	1 ER / 18 outs	-2
KC	Randy Dobnak	2 ER / 17 outs	4 ER / 4 outs	2

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
KC	Randy Dobnak (Starter)	4 / 17	27 / 3.18 (+23.82)	4 / 2	6 / 2	0 / 2	1 / 5	1.33	1 / 6	3.37	venue home: 3.86 vs 2.73 ERA (worse here) · opp run factor 1.01 (average) · 14 prior appearances this season
KC	Alex Lange (Middle Reliever (High Leverage))	3 / 3	18 / 9 (+9.00)	2 / 1	2 / 2	1 / 0	2 / 0	0.55	1 / 1	0.87

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Carter Jensen (C)	KC	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	2 / 1	-1.73	0.31 average	no signal
Salvador Perez (DH)	KC	3 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	1 / 0	1 / 1	1.26	0.311 average	no signal
Vinnie Pasquantino (1B)	KC	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	-0.8	0.335 above average	bad day
Chase Meidroth (2B)	CHW	3 / 3	2 / 1	1 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0.78	0.299 below average	good day
Michael Massey (2B)	KC	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 0	-0.78	0.306 average	no signal
John Rave (RF)	KC	4 / 3	2 / 1	1 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0.78	0.311 average	no signal
Miguel Vargas (3B)	CHW	5 / 4	2 / 1	0 / 0	0 / 0	0 / 0	2 / 1	2 / 1	1 / 1	3 / 1	0.47	0.38 elite	—
Kyle Isbel (CF)	KC	3 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.47	0.253 weak	—
Isaac Collins (LF)	KC	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	-0.33	0.294 below average	—
Tristan Peters (CF)	CHW	5 / 3	1 / 1	0 / 1	0 / 0	0 / 0	2 / 0	0 / 0	0 / 0	1 / 0	-0.31	0.315 average	—
Bobby Witt Jr. (SS)	KC	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0.31	0.378 elite	—
Colson Montgomery (SS)	CHW	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	0	0.299 below average	—
Nick Loftin (3B)	KC	4 / 4	1 / 1	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.294 below average	—

Signal	Home-away edge	Meaning
Season quality	-0.923	positive favors home; negative favors away
Recent form	-0.738	positive favors home; negative favors away
Starting-pitcher matchup	-0.046	positive favors home; negative favors away
Composite	-1.153	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	5.02-4.5	n/a	Base ML historical model prediction
variance_applied	5-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	7-5	0.545	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, starter_quality_favours_the_other_side, matchup_recency_guardrail)
embedded_savant_run_environment	7-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	7-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	7-5	n/a	Capped at max 12 realistic runs
simulation_result	7-5	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	5.02-4.5	5-5	-0.516	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-5	7-5	2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, starter_quality_favours_the_other_side, matchup_recency_guardrail)
embedded_savant_run_environment	7-5	7-5	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	7-5	7-5	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	7-5	7-5	-0	Capped at max 12 realistic runs
simulation_result	7-5	7-5	-0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
KC	Isaac Collins	batter / LF	worse	predicted_SO 3.173
CHW	Braden Montgomery	batter / RF	better	predicted_RBI 2.568
KC	Carter Jensen	batter / C	worse	predicted_RBI 2.3
KC	Kyle Isbel	batter / CF	worse	predicted_SO 2.217
CHW	David Sandlin	pitcher / Starter	boosted	predicted_H 2
CHW	Sean Newcomb	pitcher / Setup Man	deboosted	predicted_H -2
CHW	Tristan Peters	batter / CF	boosted	predicted_AB 1.9
CHW	Chase Meidroth	batter / 2B	better	predicted_H 1.87

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	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	CHW model-prediction batting totals differ from final simulation box score (AB: current=31, sim=37; H: current=8, sim=10; 2B: current=2, sim=3; HR: current=1, sim=2; RBI: current=5, sim=7; R: current=5, sim=7; BB: current=4, sim=7; SO: current=7, sim=9).	CHW
info	CURRENT_PLAYER_MATCH_BOX	CHW current finalized hitter lines are model predictions and differ from the simulation box score: Sam Antonacci: RBI current=0 box=1, BB current=0 box=1, SO current=0 box=2; Munetaka Murakami: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Kyle Teel: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Tristan Peters: AB current=3 box=5, H current=1 box=2, R current=0 box=1, SO current=0 box=1; Andrew Benintendi: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, R current=0 box=1; Braden Montgomery: AB current=3 box=4, 2B current=0 box=1, RBI current=0 box=3, BB current=1 box=0; Miguel Vargas: AB current=4 box=3, H current=1 box=2, HR current=0 box=1, BB current=1 box=2; Colson Montgomery: H current=0 box=1, RBI current=1 box=0, R current=0 box=1, BB current=0 box=2	CHW
info	CURRENT_FINALIZED_MATCH_BOX	KC model pitcher projections differ from final simulation pitching box score (ER: current=6, sim=7).	KC
info	CURRENT_FINALIZED_MATCH_BOX	KC 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; HR: current=1, sim=0; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=4; SO: current=7, sim=14).	KC
info	CURRENT_PLAYER_MATCH_BOX	KC current finalized hitter lines are model predictions and differ from the simulation box score: Carter Jensen: AB current=4 box=5, HR current=1 box=0, RBI current=1 box=3, BB current=1 box=0; Salvador Perez: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Vinnie Pasquantino: RBI current=1 box=0, BB current=1 box=0, SO current=0 box=2; Bobby Witt Jr.: AB current=4 box=3, H current=1 box=2, BB current=0 box=1; Michael Massey: AB current=4 box=3, 2B current=1 box=0, R current=0 box=1, BB current=0 box=1; Nick Loftin: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=0 box=1; John Rave: H current=1 box=0, R current=1 box=0, SO current=1 box=0, HBP current=0 box=1; Kyle Isbel: AB current=3 box=4, H current=1 box=0, SO current=1 box=3	KC
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
CIN	Brady Singer	4 ER / 12 outs	5 ER / 9 outs	1
ATL	Tyler Mahle	4 ER / 18 outs	5 ER / 15 outs	1

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
CIN	Brady Singer (Starter)	9 / 12	15 / 9 (+6.00)	5 / 4	7 / 7	1 / 1	2 / 1	3.06	1 / 7	4.27	venue away: 6.37 vs 5.08 ERA (worse here) · opp run factor 1.08 (strong) · 30 prior appearances this season
CIN	Brock Burke (Middle Reliever (High Leverage))	3 / 5	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.52	0 / 1	1.04
CIN	Emilio Pagán (Closer)	3 / 4	0 / 13.5 (-13.50)	0 / 2	1 / 2	0 / 1	0 / 0	0.28	1 / 0	1.28
ATL	Tyler Mahle (Starter)	15 / 18	9 / 6 (+3.00)	5 / 4	6 / 6	2 / 0	2 / 4	1.04	4 / 2	5.17	venue home: 2.29 vs 5.28 ERA (ok) · opp run factor 1.05 (weak) · 26 prior appearances this season

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Matt Olson (1B)	ATL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	0 / 1	-1.73	0.354 elite	bad day
Héctor Rodríguez (RF)	CIN	4 / 4	2 / 1	1 / 0	0 / 0	1 / 0	2 / 0	2 / 0	0 / 0	1 / 1	1.71	0.245 weak	good day
Austin Riley (3B)	ATL	2 / 3	1 / 1	0 / 0	0 / 0	1 / 0	2 / 0	1 / 0	1 / 0	0 / 1	1.26	0.299 below average	good day
Michael Harris II (CF)	ATL	5 / 4	3 / 1	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	0 / 0	0 / 1	0.94	0.34 elite	projection too low
Carlos Jorge (CF)	CIN	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	1 / 1	0.93	0.238 weak	good day
JJ Bleday (LF)	CIN	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	-0.8	0.339 elite	bad day
Tyler Stephenson (C)	CIN	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 1	1 / 1	-0.78	0.326 above average	bad day
Mauricio Dubón (SS)	ATL	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	0 / 0	0 / 0	0.47	0.303 average	—
Drake Baldwin (DH)	ATL	5 / 4	2 / 1	1 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	0.45	0.366 elite	—
Sal Stewart (1B)	CIN	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 0	0 / 1	0.33	0.346 elite	—
Elly De La Cruz (SS)	CIN	4 / 4	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	2 / 1	-0.33	0.375 elite	—
Mike Yastrzemski (LF)	ATL	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	1 / 1	0.33	0.31 average	—
Sean Murphy (C)	ATL	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	0 / 1	0.19	0.252 weak	—
Ronald Acuña Jr. (RF)	ATL	5 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	0.14	0.368 elite	—
Eugenio Suárez (DH)	CIN	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	2 / 1	1 / 0	0 / 0	1 / 1	0	0.299 below average	—
Ke'Bryan Hayes (3B)	CIN	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.29 below average	—
Matt McLain (2B)	CIN	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0	0.296 below average	—
Ozzie Albies (2B)	ATL	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	3 / 0	0	0.278 weak	—

Signal	Home-away edge	Meaning
Season quality	1.445	positive favors home; negative favors away
Recent form	0.59	positive favors home; negative favors away
Starting-pitcher matchup	0.032	positive favors home; negative favors away
Composite	1.373	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	3.76-5.68	n/a	Base ML historical model prediction
variance_applied	4-6	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-6	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	5-7	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-7	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-7	n/a	Capped at max 12 realistic runs
simulation_result	5-7	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	3.76-5.68	4-6	0.079	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-6	5-6	-1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	5-6	5-7	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-7	5-7	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-7	5-7	0	Capped at max 12 realistic runs
simulation_result	5-7	5-7	0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
ATL	Mike Yastrzemski	batter / LF	worse	predicted_RBI 1.78
CIN	Tyler Stephenson	batter / C	worse	predicted_BB 1.624
CIN	Carlos Jorge	batter / CF	worse	predicted_AB 1.5
CIN	Eugenio Suárez	batter / DH	better	predicted_RBI 1.45
ATL	Michael Harris II	batter / CF	worse	predicted_AB -1.43
CIN	Elly De La Cruz	batter / SS	better	predicted_RBI 1.39
ATL	Matt Olson	batter / 1B	better	predicted_RBI 1.39
ATL	Ozzie Albies	batter / 2B	worse	predicted_H 1.16

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	CIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CIN
info	CURRENT_FINALIZED_MATCH_BOX	ATL model pitcher projections differ from final simulation pitching box score (BB: current=4, sim=5).	ATL
info	CURRENT_FINALIZED_MATCH_BOX	ATL model-prediction batting totals differ from final simulation box score (3B: current=0, sim=1; HR: current=1, sim=2; RBI: current=4, sim=7; R: current=5, sim=7; BB: current=3, sim=1; SO: current=7, sim=10).	ATL
info	CURRENT_PLAYER_MATCH_BOX	ATL current finalized hitter lines are model predictions and differ from the simulation box score: Drake Baldwin: 2B current=0 box=1, BB current=1 box=0, SO current=1 box=2; Mauricio Dubón: 3B current=0 box=1, RBI current=0 box=1, R current=0 box=1; Matt Olson: RBI current=1 box=2, BB current=1 box=0; Ozzie Albies: H current=1 box=2, RBI current=0 box=1; Ronald Acuña Jr.: AB current=4 box=3, RBI current=1 box=0; Michael Harris II: 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=2; Mike Yastrzemski: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Austin Riley: H current=1 box=0, SO current=1 box=2	ATL
info	CURRENT_FINALIZED_MATCH_BOX	CIN model pitcher projections differ from final simulation pitching box score (ER: current=6, sim=7).	CIN
info	CURRENT_FINALIZED_MATCH_BOX	CIN model-prediction batting totals differ from final simulation box score (AB: current=33, sim=34; H: current=8, sim=7; 2B: current=1, sim=4; HR: current=1, sim=0; RBI: current=4, sim=5; R: current=3, sim=5; BB: current=3, sim=5; SO: current=9, sim=4).	CIN
info	CURRENT_PLAYER_MATCH_BOX	CIN current finalized hitter lines are model predictions and differ from the simulation box score: Carlos Jorge: 2B current=0 box=1, R current=0 box=1, HBP current=0 box=1; Tyler Stephenson: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Sal Stewart: 2B current=0 box=1, RBI current=1 box=0, BB current=0 box=1, SO current=1 box=0; Héctor Rodríguez: RBI current=0 box=1, R current=0 box=1; Elly De La Cruz: AB current=4 box=5, 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=2; JJ Bleday: AB current=4 box=3, H current=1 box=0, R current=1 box=0, SO current=1 box=0; Eugenio Suárez: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, RBI current=1 box=2; Ke'Bryan Hayes: AB current=3 box=4, H current=0 box=1, SO current=1 box=0	CIN
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
CLE	Daniel Espino	2 ER / 9 outs	0 ER / 3 outs	-2
BOS	Ranger Suarez	1 ER / 17 outs	1 ER / 15 outs	0

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
CLE	Daniel Espino (Starter)	3 / 9	0 / 6 (-6.00)	0 / 2	0 / 6	0 / 1	0 / 1	4.59	2 / 6	4.06	venue away: 0 vs 27 ERA (ok) · opp run factor 0.98 (average) · 12 prior appearances this season
CLE	Cade Smith (Closer)	3 / 6	0 / 0	0 / 0	0 / 0	0 / 0	0 / 2	0.36	3 / 1	1.66
CLE	Hunter Gaddis (Setup Man)	3 / 6	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0.25	1 / 1	0.86
CLE	Erik Sabrowski (Middle Reliever (High Leverage))	3 / 6	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0.42	2 / 0	0.91
BOS	Ranger Suarez (Starter)	15 / 17	1.8 / 1.59	1 / 1	3 / 2	1 / 0	1 / 3	1.94	5 / 5	4.31	venue home: 3.81 vs 2.85 ERA (worse here) · opp run factor 0.97 (average) · 27 prior appearances this season
BOS	Erik Miller (Setup Man)	3 / 3	0 / 18 (-18.00)	0 / 2	0 / 2	0 / 1	0 / 1	0.78	1 / 1	1.25

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Chase DeLauter (RF)	CLE	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	3 / 1	-1.11	0.346 elite	bad day
Adley Rutschman (C)	BOS	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	3 / 1	-1.11	0.328 above average	bad day
Wilyer Abreu (RF)	BOS	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-1.11	0.35 elite	bad day
Mickey Gasper (DH)	BOS	2 / 3	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	1 / 0	1 / 1	-1.07	0.36 elite	bad day
Steven Kwan (CF)	CLE	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	-0.47	0.327 above average	—
Jo Adell (DH)	CLE	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.324 above average	—
David Fry (1B)	CLE	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.47	0.264 weak	—
Travis Bazzana (2B)	CLE	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.302 below average	—
Patrick Bailey (C)	CLE	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.304 average	—
Brayan Rocchio (SS)	CLE	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	-0.47	0.295 below average	—
Trevor Story (SS)	BOS	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.47	0.25 weak	—
Nick Sogard (1B)	BOS	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	-0.33	0.31 average	—
José Ramírez (3B)	CLE	4 / 4	2 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.17	0.327 above average	—
Caleb Durbin (3B)	BOS	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	-0.14	0.295 below average	—
Ceddanne Rafaela (CF)	BOS	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	-0.14	0.284 below average	—
Angel Martínez (LF)	CLE	4 / 4	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 0	0 / 1	0	0.306 average	—
Isiah Kiner-Falefa (2B)	BOS	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.296 below average	—
Jarren Duran (LF)	BOS	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0	0.301 below average	—

Signal	Home-away edge	Meaning
Season quality	0.171	positive favors home; negative favors away
Recent form	-0.41	positive favors home; negative favors away
Starting-pitcher matchup	0.014	positive favors home; negative favors away
Composite	-0.337	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.56-4.21	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-3	0.506	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	4-2	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-2	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-2	n/a	Capped at max 12 realistic runs
simulation_result	4-2	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	4.56-4.21	5-4	0.651	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	4-3	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	4-3	4-2	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-2	4-2	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-2	4-2	-0	Capped at max 12 realistic runs
simulation_result	4-2	4-2	-0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
BOS	Mickey Gasper	batter / DH	worse	predicted_2B 1.857
CLE	David Fry	batter / 1B	worse	predicted_BB 1.676
CLE	Daniel Espino	pitcher / Starter	worse	predicted_IP -1.667
CLE	José Ramírez	batter / 3B	worse	predicted_AB -1.66
CLE	Chase DeLauter	batter / RF	worse	predicted_RBI 1.45
BOS	Trevor Story	batter / SS	worse	predicted_AB -1.35
CLE	Erik Sabrowski	pitcher / Middle Reliever (High Leverage)	better	predicted_IP 1.333
BOS	Nick Sogard	batter / 1B	worse	predicted_AB 1.317

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	BOS pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	BOS
info	CURRENT_PITCHER_DUPLICATION	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	BOS model-prediction batting totals differ from final simulation box score (AB: current=32, sim=34; H: current=8, sim=7; 3B: current=0, sim=1; RBI: current=4, sim=2; R: current=5, sim=2).	BOS
info	CURRENT_PLAYER_MATCH_BOX	BOS current finalized hitter lines are model predictions and differ from the simulation box score: Nick Sogard: AB current=4 box=5, H current=1 box=0, BB current=1 box=0; Mickey Gasper: AB current=3 box=2, H current=1 box=2, 2B current=0 box=2, HR current=1 box=0; Adley Rutschman: 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Wilyer Abreu: 2B current=1 box=0, HR current=0 box=1, BB current=1 box=0, SO current=1 box=2; Caleb Durbin: AB current=3 box=4, H current=1 box=0, R current=1 box=0, SO current=0 box=1; Isiah Kiner-Falefa: AB current=3 box=4, H current=0 box=1; Ceddanne Rafaela: R current=0 box=1, SO current=1 box=2; Jarren Duran: AB current=3 box=4, 3B current=0 box=1, R current=1 box=0, SO current=1 box=0	BOS
info	CURRENT_FINALIZED_MATCH_BOX	CLE model-prediction batting totals differ from final simulation box score (H: current=8, sim=5; 2B: current=2, sim=1; HR: current=1, sim=2; BB: current=3, sim=4).	CLE
info	CURRENT_PLAYER_MATCH_BOX	CLE current finalized hitter lines are model predictions and differ from the simulation box score: Steven Kwan: AB current=4 box=3, H current=1 box=0, R current=1 box=0; Jo Adell: H current=1 box=0, RBI current=1 box=0, SO current=1 box=2; David Fry: AB current=3 box=2, BB current=0 box=2; José Ramírez: AB current=4 box=3, 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Angel Martínez: 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Chase DeLauter: 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=2, BB current=1 box=0; Travis Bazzana: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=1, R current=0 box=1; Patrick Bailey: AB current=3 box=4, H current=1 box=0	CLE
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
HOU	Peter Lambert	1 ER / 17 outs	4 ER / 8 outs	3
OAK	Mason Barnett	3 ER / 11 outs	4 ER / 18 outs	1

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
HOU	Peter Lambert (Starter)	8 / 17	13.5 / 1.59 (+11.91)	4 / 1	6 / 4	2 / 0	2 / 1	1.78	6 / 9	5.2	venue away: 3.17 vs 3.99 ERA (ok) · opp run factor 1.05 (weak) · 27 prior appearances this season
HOU	Bryan Abreu (Middle Reliever (High Leverage))	4 / 3	0 / 9 (-9.00)	0 / 1	1 / 2	0 / 0	1 / 1	0.34	0 / 2	0.86
HOU	Bryan King (Reliever)	4 / 0	0 / n/a	0 / 1	1 / 1	0 / 0	0 / 0	n/a	3 / 1	n/a
OAK	Mason Barnett (Starter)	18 / 11	6 / 7.36	4 / 3	4 / 6	2 / 1	2 / 1	3.2	8 / 4	3.9	venue home: 8 vs 12.19 ERA (ok) · opp run factor 1 (average) · 19 prior appearances this season
OAK	Scott Blewett (Middle Reliever (High Leverage))	3 / 6	9 / 9	1 / 2	3 / 3	1 / 0	0 / 0	0.41	0 / 3	1.16
OAK	Elvis Alvarado (Closer)	3 / 4	9 / 6.75	1 / 1	2 / 2	1 / 0	0 / 0	0.43	0 / 1	1.04

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Lucas Spence (CF)	HOU	4 / 3	3 / 1	0 / 0	0 / 0	3 / 0	5 / 0	3 / 0	0 / 1	0 / 1	3.4	0.231 weak	good day
Zack Gelof (3B)	OAK	3 / 4	2 / 1	0 / 0	0 / 0	2 / 0	2 / 1	2 / 1	1 / 0	0 / 1	2.66	0.322 above average	projection too low
Christian Walker (1B)	HOU	4 / 4	2 / 1	0 / 0	0 / 0	1 / 0	1 / 1	2 / 0	0 / 0	1 / 1	1.4	0.298 below average	good day
Jeff McNeil (1B)	OAK	4 / 4	2 / 1	1 / 1	0 / 0	1 / 0	3 / 1	1 / 0	0 / 0	1 / 0	1.4	0.3 below average	good day
Jose Altuve (2B)	HOU	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	-1.11	0.272 weak	projection too high
Yordan Alvarez (LF)	HOU	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	1 / 1	-0.93	0.444 elite	bad day
Isaac Paredes (3B)	HOU	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 1	1 / 1	0 / 0	0.93	0.323 above average	projection too low
Jeremy Peña (SS)	HOU	5 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-0.93	0.326 above average	bad day
Shea Langeliers (DH)	OAK	5 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 0	1 / 1	-0.93	0.358 elite	bad day
Henry Bolte (CF)	OAK	5 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	4 / 1	-0.8	0.318 average	no signal
Yainer Diaz (DH)	HOU	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.78	0.312 average	no signal
Jonah Heim (C)	OAK	3 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 1	0.64	0.283 weak	good day
Lawrence Butler (RF)	OAK	5 / 4	3 / 1	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	0.61	0.328 above average	projection too low
Carlos Cortes (LF)	OAK	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	-0.47	0.305 average	—
Donovan Walton (2B)	OAK	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	1 / 1	-0.33	0.326 above average	—
Joshua Kuroda-Grauer (SS)	OAK	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	2 / 1	-0.14	0.32 average	—
Christian Vázquez (C)	HOU	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0	0.233 weak	—
Cam Smith (RF)	HOU	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	2 / 1	0	0.318 average	—

Signal	Home-away edge	Meaning
Season quality	-0.848	positive favors home; negative favors away
Recent form	-0.162	positive favors home; negative favors away
Starting-pitcher matchup	0.05	positive favors home; negative favors away
Composite	-0.758	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	5.72-3.8	n/a	Base ML historical model prediction
variance_applied	6-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	7-5	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	7-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	7-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	7-5	n/a	Capped at max 12 realistic runs
simulation_result	7-5	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	5.72-3.8	6-4	0.088	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-4	7-5	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	7-5	7-5	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	7-5	7-5	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	7-5	7-5	-0	Capped at max 12 realistic runs
simulation_result	7-5	7-5	-0	Final inning-by-inning simulation result

Compared against packet 20260923. 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.

The adjustment the system made in response

Luna deep post-mortem (full)

1. Evidence receipt and gaps

Sources received

All seven manifest entries were checked against the supplied evidence:

Source	Status	Purpose
`prediction review` — `/app/output/reports/prediction_review_20260924.json`	Loaded, 301,958 bytes	Official outcomes and scoring errors
`forensic deep audit` — `/app/output/reports/PREDICTION_SYSTEM_DEEP_AUDIT_20260924.json`	Loaded, 196,012 bytes	Probability, provenance, starter, and decision contracts
`continuous learning` — `/app/output/learning/continuous_learning_daily_20260924.json`	Loaded, 2,023,270 bytes	Stage quality, calibration, lineage, score distribution, starter comparisons
`detailed report` — `/app/output/reports/detailed_20260924.html`	Loaded, 285,081 bytes	Full game and simulation rendering
`ultra detailed player audit` — `/app/output/reports/20260924_ultra_detailed_player_audit.html`	Loaded, 25,971,500 bytes	Player-level prediction trace
`structural audit` — `/app/output/reports/audit_report_20260924.html`	Loaded, 1,530,211 bytes	Invariant findings
`integrity revision audit` — `/app/output/reports/audit_report_20260924_integrity_revision.html`	Missing	Post-Luna remediation proof for immutable revision artifacts

The missing integrity-revision audit is a provenance gap, but it did not prevent the supplied per-game structural checks from running. It does mean immutable revision/remediation claims cannot be independently confirmed from this receipt.

Games reviewed

All 12 games in the slate were reviewed, and the coverage matrix reports complete game-level coverage:

Expected games: 12
Canonical artifacts loaded: 12
Continuous-learning games: 12
Forensic deep-audit games: 12
Starter comparisons: 12
Structural audits: 12
Games structurally audited: 12

The 12 games were:

ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@OAK, LAA@SEA, MIA@CHC, MIL@PHI, NYM@TEX, SD@LAD, STL@PIT, and TB@NYY.

Internal gaps and contradictions

Missing integrity revision artifact. The manifest explicitly marks the integrity revision audit as missing.
Pipeline trace contradiction for later winner direction. The stored transitions say that after enhanced_prediction, no later stage changed direction, but at least CIN@ATL is shown with an enhanced-stage score favoring CIN while the published prediction favors ATL. The probability trace also reports a published ATL probability of 0.560. This requires reconciliation in the raw trace; the supplied evidence does not fully establish whether the later winner changed, whether score and probability paths diverged, or whether the transition summary omitted a change.
Stage summary versus per-game result ambiguity. historical_ml has 9/12 winner hits, variance_applied 6/12, and final stages 8/12. The aggregate transition record correctly identifies a three-game regression at variance application, but it does not list all winner states for every later stage.
Calibration labels are internally consistent with the stated definition. The definition is published confidence minus observed hit rate. Thus 0.5248 − 0.75 = −0.2252 is underconfident, while 0.56 − 0.50 = +0.06 is overconfident.
No market cross-check exists. No source-labelled moneyline was available for any game, so market agreement/disagreement cannot be used diagnostically.
Starter identity coverage was complete, but not every per-player feed line was complete. For example, projected starter lines were missing for Eduardo Rodriguez and David Sandlin in the settled per-player feed, despite game-level starter comparisons being present.

---

2. Executive diagnosis

Overall result

Winner accuracy: 8/12, 66.67%
Misses: 4/12
Mean Brier score: approximately 0.240
Published weighted absolute calibration error: 0.1701
Historical validation winner accuracy: 0.592357
Walk-forward accuracy: 0.57551
Walk-forward Brier: 0.241034
Holdout run-model \(R²\): away 0.034855, home 0.003987, total 0.003064
Final-stage team-run MAE: 2.2083
Final-stage total-run MAE: 3.75
Final-stage margin MAE: 2.0833
Final-stage winner accuracy: 8/12

The slate beat the stated walk-forward baseline and was directionally successful, but the score forecasts were weak, the probability buckets were poorly aligned, and the system had a material intermediate winner-direction regression.

Separate diagnoses

Winner accuracy

The final published pick was correct in 8 of 12 games. This is useful directional performance, but it does not imply that the numerical prediction was good.

The four losses were:

CIN@ATL
LAA@SEA
NYM@TEX
SD@LAD

Two losses were classified as wrong-side conviction calls, one as a run-environment miss, and one as toss-up variance:

Wrong conviction side: 2
Run environment overestimated: 1
Genuine toss-up variance: 1

Calibration quality

Calibration was the most important winner-probability weakness on this slate.

0.50–0.55 bucket: average confidence 0.5248, observed hit rate 0.75, eight games; the model was underconfident in this bucket.
0.55–0.60 bucket: average confidence 0.56, observed hit rate 0.50, four games; the model was overconfident in this bucket.
Weighted absolute calibration error: 0.1701 across 12 settled games.
High-conviction miss rate: 0.5, with 2 misses among 4 high-conviction games.
The high-raw-probability cluster had 2 hits and 1 miss; the current slate did not meet the persistent high-raw-cluster streak condition.

The correct response is not to treat all probabilities as unusable. The evidence supports bounded confidence-cap and variance shadow experiments, not a wholesale calibration rewrite based on one slate.

Score accuracy

Score accuracy was materially weaker than winner accuracy.

The model predicted every margin at two runs or less:

Predicted margins ≥ 3: 0/12
Actual margins ≥ 4: 3/12
Actual margins: four one-run games, five two-run games, two four-run games, and one eight-run game.
Signed margin errors included underestimates of 3 runs in ARI@COL, 6 in CHW@KC, and 3 in MIL@PHI.

The final-stage run model has near-zero holdout explanatory power. Exact scores should therefore remain low-confidence outputs. The model correctly identified several winners while badly missing the score magnitude.

Starter accuracy

Starter forecasts were mixed:

Several losses were caused by outlier starter performance: Tanner Gordon, Peter Lambert, Kumar Rocker, and Paul Skenes’ unexpectedly strong outing.
ARI@COL and CHW@KC had very large starter errors, but the winner was still correct.
CIN@ATL had both starters only one earned run from projection, so the loss was not primarily a starter-model failure.
LAA@SEA had both starters close enough to projection that the loss was better classified as variance or wrong-side conviction than starter error.
NYM@TEX was affected by Kumar Rocker allowing 1 ER versus a projected 4 ER, but the stored primary diagnosis correctly emphasizes the lower-than-projected run environment.
SD@LAD had Nick Pivetta outperforming projection and the game remained a toss-up.

The cross-slate starter ERA finding remains nonpersistent, so a strong starter-weight change is not justified solely by this slate.

Artifact integrity

Artifact integrity was strong for the game artifacts:

All 12 structural audits passed.
Structural audit summary: critical 0, error 0, warning 0, info 91.
Each per-game audit had zero critical/error/warning findings.
Forecast-file integrity was reported clean for every game.

However, the missing integrity-revision audit remains a receipt gap. It is a presentation/provenance gap, not evidence that the picks themselves were corrupted.

Causal issue register reconciliation

`pipeline_winner_instability` — High

Evidence: historical_ml → variance_applied reduced winner accuracy by 0.2500, or 3 hits.
Affected games: CHW@KC, MIA@CHC, NYM@TEX.
Action: list exact pre- and post-stage winner states and retain both in the trace.
Guardrail: later recovery does not make the unstable step harmless.

This was a real model/process risk. The variance stage changed three directions to ties, and the subsequent enhanced stage recovered two correctly and moved one incorrectly. The intermediate regression must remain visible even where the final result recovered.

`single_slate_calibration_instability` — High

Evidence: weighted absolute calibration error 0.1701.
Affected games: all 12.
Action: monitor the same buckets over the rolling window before changing calibration weights.
Guardrail: a 12-game slate cannot establish a stable calibration curve.

This is a real current-slate warning, but not sufficient by itself for a permanent calibration rewrite.

`weak_run_model_signal` — High

Evidence: home/away/total holdout \(R²\) 0.0040/0.0349/0.0031.
Affected games: all 12.
Action: improve features/target and run an isolated training backtest.
Guardrail: small positive \(R²\) does not make exact scores reliable.

This is both a current-slate and persistent cross-slate model risk.

`complete_slate_audit_coverage` — Info

Evidence: 12/12 games structurally audited.
Guardrail: coverage does not cancel findings inside audited games.
Action: retain the per-game audit receipt.

Coverage is complete; the slate is not therefore “clean” in predictive terms.

`probability_publication_delta_explained` — Info

Evidence: five material movements were attributed to score-coherence and/or recency transformations; no unexplained material delta was reported.
Guardrail: a published probability above calibrated probability is not a governor bypass when a named transformation explains it.
Action: continue exposing every probability layer.

This supports artifact lineage integrity for the five material movements.

`simulation_inert_on_slate` — Info on this slate, escalated persistent failure across slates

Evidence this slate: zero flip candidates and zero published flips.
Guardrail: zero winner flips is not automatically a defect.
Action: measure simulation contribution over rolling slates.

Cross-slate evidence now elevates this from a single-slate observation to an established code/diagnostic issue.

---

3. Every game

The published probability below is the probability of the published side. A home-minus-away edge of 0.12 means a 56% home / 44% away split when the home side was favored, or 44% home / 56% away when the away side was favored. It is moderate, not near certainty and not a near coin flip.

ARI@COL — correct winner, wrong score

Predicted: ARI 6, COL 5; ARI winner
Actual: ARI 12, COL 8; ARI winner
Published pick probability: ARI 0.56
Confidence edge: −0.12 home-minus-away, moderate
Bucket: moderate
Brier: 0.1936
Team-run absolute error: 9
Total-run bias: −9; total was projected 11 versus actual 20
Predicted margin: ARI by 1; actual margin: ARI by 4

Edges:

Season home-minus-away: −1.1569
Recent: −0.666
Pitcher: −0.1533
Composite: −1.3201

Starters:

Eduardo Rodriguez: projected 3 ER/17 outs, actual 2 ER/18 outs; identity matched. The detailed player feed lacked a settled projected line for him in one matchup section.
Tanner Gordon: projected 3 ER/13 outs, actual 10 ER/10 outs; ERA 27.00 versus projected 6.23, ER error +7.

Controls and score path:

Governor reached 0.661 before score-coherence and recency controls; published home probability was 0.440, therefore ARI 0.560.
Savant adjustment applied; large run-environment swing flag.
Score path: historical model approximately ARI 4 / COL 5, variance stage 5/5, later score target 6/5.
Volatility action: supported blowout-margin expansion.

Player adjustments:

Cole Carrigg: +2.03 run impact versus projection.
Geraldo Perdomo: +1.56.
Pavin Smith: +1.11.
Kyle Karros: +1.07.
Nolan Arenado: +0.94.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 8 info.
Primary cause: none assigned because the winner was correct.
Secondary cause: none.
Forensic diagnosis: right winner, wrong score. Tanner Gordon’s severe home/venue underperformance and under-projected offense drove the score error. The evidence supports a shadow test of context.venue_adjustment_weight, not a claim that this one start proves the adjustment is generally valid.

CHW@KC — correct winner, wrong score

Predicted: CHW 7, KC 5; CHW winner
Actual: CHW 9, KC 1; CHW winner
Published pick probability: CHW 0.54462
Confidence edge: −0.0892, lean
Bucket: lean
Brier: 0.207371
Team-run absolute error: 6
Total-run bias: +2
Predicted margin: CHW by 2; actual margin: CHW by 8

Edges:

Season: −0.9229
Recent: −0.738
Pitcher: −0.0461
Composite: −1.1535

Starters:

David Sandlin: projected 3 ER/15 outs, actual 1 ER/18 outs; detailed per-player feed lacked a settled projected line.
Randy Dobnak: projected 2 ER/17 outs, actual 4 ER/4 outs; ERA 27.00 versus 3.18, ER error +2, outs error −13.

Controls and path:

Experience control applied and shrank confidence toward 0.5; winner unchanged.
Published probability followed the governed value after score/recency controls.
Score path increased toward a CHW advantage and supported blowout-margin expansion.

Player adjustments:

Carter Jensen: −1.73.
Salvador Perez: +1.26.
Vinnie Pasquantino: −0.80.
Chase Meidroth: +0.78.
Michael Massey: −0.78.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 8 info.
Primary/secondary causes: none assigned because the winner was correct.
Classification: right winner, wrong score. KC’s starter and lineup both underperformed the projected scoring environment.

This game is also one of the three documented winner-direction regressions at historical_ml → variance_applied: CHW became a tie before recovering to CHW at enhanced prediction.

CIN@ATL — wrong winner and wrong score

Predicted: CIN 5, ATL 7; ATL winner
Actual: CIN 7, ATL 6; CIN winner
Published pick probability: ATL 0.560
Confidence edge: +0.12, moderate
Bucket: moderate
Brier: 0.3136
Team-run absolute error: 3
Total-run bias: −1
Predicted margin: ATL by 2; actual margin: CIN by 1

Edges:

Season: +1.445
Recent: +0.590
Pitcher: +0.032
Composite: +1.3726

Starters:

Brady Singer: projected 4 ER/12 outs, actual 5 ER/9 outs; ERA error +6.
Tyler Mahle: projected 4 ER/18 outs, actual 5 ER/15 outs; ERA error +3.
Both starters were only one earned run off in absolute terms, so the starter projection was broadly accurate relative to this miss.

Controls and path:

Confidence reached 0.5973 before controls; published ATL probability was 0.560.
Savant adjustment applied; no volatility adjustment.
Historical score path initially favored ATL, variance moved to a tie, enhanced prediction moved toward CIN, and later stored published score was again CIN 5 / ATL 7. This creates the noted transition-trace contradiction.
Recent form favored ATL: ATL 6–4 with +6 run differential versus CIN 4–6 and −14.

Player adjustments:

Mike Yastrzemski: largest adverse change, +1.78 projected RBI delta.
Tyler Stephenson: +1.624 projected BB delta.
Carlos Jorge: +1.50 projected AB delta.
Eugenio Suárez: +1.45 favorable RBI delta.
Michael Harris II: −1.43 AB delta.
Elly De La Cruz and Matt Olson each had +1.39 RBI-related changes.
Material player-adjustment count: 23.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 9 info.
Primary: wrong_conviction_side.
Secondary: none.
The miss was not primarily caused by a gross starter projection failure. It was a moderate-confidence commitment to ATL that lost a close game.

CLE@BOS — correct winner, wrong score

Predicted: CLE 4, BOS 2; CLE winner
Actual: CLE 1, BOS 0; CLE winner
Published pick probability: CLE 0.50619
Confidence edge: −0.0124, toss-up
Bucket: toss_up
Brier: 0.243848
Team-run absolute error: 5
Total-run bias: +5
Predicted margin: CLE by 2; actual margin: CLE by 1

Edges:

Season: +0.1714
Recent: −0.410
Pitcher: +0.014
Composite: −0.3374

Starters:

Daniel Espino: projected 2 ER/9 outs, actual 0 ER/3 outs; ERA error −6. The audit calls this an outlier with no credible explanatory signal.
Ranger Suarez: projected 1 ER/17 outs, actual 1 ER/15 outs; essentially accurate.

Controls and path:

Governed probability 0.5073; published away probability 0.50619.
Score volatility: no change.
Both lineups scored below projection.
The toss-up went the model’s way.

Player adjustments:

Chase DeLauter, Adley Rutschman, Wilyer Abreu, and Mickey Gasper all had negative run impacts from underproduction.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 7 info.
Primary/secondary causes: none.
Classification: correct winner, wrong score; mostly low-run variance and an unexpectedly short, scoreless Espino appearance.

HOU@OAK — correct winner and close score

Predicted: HOU 7, OAK 5; HOU winner
Actual: HOU 7, OAK 5; HOU winner
Published pick probability: HOU 0.56
Confidence edge: −0.12, moderate
Bucket: moderate
Brier: 0.1936
Team-run absolute error: 0
Total-run bias: 0
Predicted and actual margin: HOU by 2

Edges:

Season: −0.8484
Recent: −0.162
Pitcher: +0.0496
Composite: −0.7580

Starters:

Peter Lambert: projected 1 ER/17 outs, actual 4 ER/8 outs; ERA error +11.91.
Mason Barnett: projected 3 ER/11 outs, actual 4 ER/18 outs; ER error +1, outs +7.

Controls and path:

Governed probability 0.6344 before publication; score coherence and recency produced published HOU probability 0.56.
Blowout-margin expansion was supported.
The score remained exactly correct despite very poor Lambert starter accuracy because HOU’s offense compensated.

Player adjustments:

Lucas Spence: +3.40.
Zack Gelof: +2.66 and identified as a learnable strong-hitter under-rating.
Christian Walker: +1.40.
Jeff McNeil: +1.40.
Jose Altuve: −1.11.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 8 info.
Primary/secondary causes: none.
Strong correct call, but score correctness should not conceal a large pitcher error.

LAA@SEA — wrong winner and wrong score

Predicted: LAA 4, SEA 5; SEA winner
Actual: LAA 6, SEA 4; LAA winner
Published pick probability: SEA 0.560
Confidence edge: +0.12, moderate
Bucket: moderate
Brier: 0.3136
Team-run absolute error: 3
Total-run bias: −1
Predicted margin: SEA by 1; actual margin: LAA by 2

Edges:

Season: +0.4043
Recent: +0.614
Pitcher: +0.1394
Composite: +0.8625

Starters:

Grayson Rodriguez: projected 5 ER/15 outs, actual 4 ER/15 outs.
Bryan Woo: projected 3 ER/17 outs, actual 4 ER/16 outs.
Both were within one earned run; starter error was not the main cause.

Controls and path:

Governed probability 0.6366; published SEA probability 0.56 after score-coherence and recency controls.
Savant adjustment and supported blowout-margin expansion favored SEA.
Model had SEA recent run differential +13 versus LAA −18, and recent scoring 5.4 versus 3.6.

Player adjustments:

Josh Naylor: +2.57 projected RBI change.
Lazaro Montes: +1.70 RBI change.
Randy Arozarena: −1.70 AB change.
Zach Neto: +1.64 deboost-related change.
Julio Rodríguez: actual run impact +1.40 and classified as a learnable under-rating.
Material change count: 22.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 7 info.
Primary: wrong_conviction_side.
Secondary: none.
The evidence does not support blaming the starters; this was a moderate-confidence wrong-side commitment in a close run environment.

MIA@CHC — correct winner, wrong score

Predicted: MIA 4, CHC 5; CHC winner
Actual: MIA 1, CHC 2; CHC winner
Published pick probability: CHC 0.519035
Confidence edge: +0.0381, toss-up
Bucket: toss_up
Brier: 0.231327
Team-run absolute error: 6
Total-run bias: +6
Predicted margin: CHC by 1; actual margin: CHC by 1

Edges:

Season: +0.6258
Recent: −0.098
Pitcher: +0.0807
Composite: +0.6772

Starters:

Tyler Phillips: projected 1 ER/15 outs, actual 0 ER/12 outs.
Matthew Boyd: projected 3 ER/17 outs, actual 1 ER/21 outs.
Both outperformed their ER projections, causing the low total.

Controls and path:

Governed probability 0.5263; published CHC probability 0.5190.
Starter experience control was not applied.
Variance action: no change.

Player adjustments:

Heriberto Hernández: −1.73.
Seiya Suzuki: +1.09 and identified as a learnable strong-hitter under-rating.
Pete Crow-Armstrong: −0.93.
Michael Busch: −0.80.
Both lineups scored substantially below projection.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 8 info.
Primary/secondary causes: none.
Correct winner, wrong score. This is the second documented variance-stage winner regression: historical CHC moved to a tie at variance and recovered to CHC at enhanced prediction.

MIL@PHI — correct winner, wrong score

Predicted: MIL 5, PHI 4; MIL winner
Actual: MIL 5, PHI 1; MIL winner
Published pick probability: MIL 0.53921
Confidence edge: −0.0784, lean
Bucket: lean
Brier: 0.212327
Team-run absolute error: 3
Total-run bias: +3
Predicted margin: MIL by 1; actual margin: MIL by 4

Edges:

Season: −0.955
Recent: −0.530
Pitcher: +0.0056
Composite: −0.8254

Starters:

Shane Drohan: projected 1 ER/14 outs, actual 1 ER/8 outs.
Andrew Painter: projected 5 ER/15 outs, actual 1 ER/18 outs; ER error −4.
Painter’s strong outing materially reduced PHI’s scoring.

Controls and path:

Governed probability 0.5543; published MIL probability 0.53921.
No volatility change.
Score environment was too high, especially for PHI.

Player adjustments:

Sal Frelick: +0.97.
Cooper Pratt: +0.80.
Trea Turner: −0.78.
Brice Turang: −0.64.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 7 info.
Primary/secondary causes: none.
Correct winner, wrong margin, mainly from PHI starter and offense underproduction.

NYM@TEX — wrong winner and wrong score

Predicted: NYM 5, TEX 4; NYM winner
Actual: NYM 1, TEX 3; TEX winner
Published pick probability: NYM 0.520878
Confidence edge: −0.0418, lean
Bucket: lean
Brier: 0.271314
Team-run absolute error: 5
Total-run bias: +5
Predicted margin: NYM by 1; actual margin: TEX by 2

Edges:

Season: +0.110
Recent: −0.018
Pitcher: −0.0998
Composite: −0.3436

Starters:

Zac Thornton: projected 2 ER/17 outs, actual 2 ER/15 outs; close.
Kumar Rocker: projected 4 ER/14 outs, actual 1 ER/15 outs; ERA error −5.91.

Controls and path:

Governed probability 0.5246; published NYM probability 0.520878.
No recency guardrail was applied.
Score path moved from approximately 4/4 to 5/4 for NYM.
The model had just seen NYM win the prior two head-to-head games, but TEX won this game.
Score-volatility action: no change.

Player adjustments:

Francisco Lindor: +2.53 RBI-related change.
Josh Jung: +1.76 hit change and actual +2.20 run impact; identified as a learnable under-rating.
Justin Foscue: −2.50 AB change.
Juan Soto: −1.305 deboosted BB change and actual negative run impact.
Material change count: 24.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 7 info.
Primary: run_environment_over.
Secondary: none.
The winner miss began with a near-even directional call and was worsened by a much lower run environment than projected. Rocker’s unexpectedly strong performance contributed, but the stored diagnosis properly identifies the broad run-environment error.

This is the third documented variance-stage instability game: historical evidence favored TEX, variance moved to a tie, and enhanced prediction moved to NYM—the wrong direction relative to the actual result.

SD@LAD — wrong winner and wrong score

Predicted: SD 4, LAD 5; LAD winner
Actual: SD 4, LAD 2; SD winner
Published pick probability: LAD 0.509796
Confidence edge: +0.0196, toss-up
Bucket: toss_up
Brier: 0.259892
Team-run absolute error: 3
Total-run bias: +3
Predicted margin: LAD by 1; actual margin: SD by 2

Edges:

Season: +0.6803
Recent: −0.002
Pitcher: −0.0449
Composite: +0.4549

Starters:

Nick Pivetta: projected 2 ER/15 outs, actual 0 ER/14 outs.
Tyler Glasnow: projected 2 ER/18 outs, actual 3 ER/19 outs.
Pivetta’s strong start and the low-scoring game were not supported by a credible predictive signal in the audit.

Controls and path:

Starter-experience control shrank probability from 0.518595 to 0.513559; winner unchanged.
Published LAD probability was 0.509796.
Score path remained approximately SD 4 / LAD 5.
Low-score probability was 0.6925, but volatility action was still no change.

Player adjustments:

Mookie Betts: +1.58 and classified as a learnable strong-hitter under-rating.
Dustin Harris: +2.16 hit-related change.
Hunter Feduccia: +2.28.
Manny Machado: −1.73.
Material change count: 24.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 8 info.
Primary: toss_up_variance.
Secondary: none.
This was a genuine low-confidence flip, not a strong model conviction failure.

STL@PIT — correct winner, wrong score

Predicted: STL 4, PIT 5; PIT winner
Actual: STL 1, PIT 2; PIT winner
Published pick probability: PIT 0.52755
Confidence edge: +0.0551, lean
Bucket: lean
Brier: 0.223209
Team-run absolute error: 6
Total-run bias: +6
Predicted margin: PIT by 1; actual margin: PIT by 1

Edges:

Season: +0.3532
Recent: +0.462
Pitcher: −0.0209
Composite: +0.4438

Starters:

Kyle Leahy: projected 2 ER/15 outs, actual 1 ER/6 outs.
Paul Skenes: projected 4 ER/17 outs, actual 1 ER/19 outs; ER error −3.

Controls and path:

Governed PIT probability 0.5381; published 0.52755.
No volatility change.
Both lineups scored three runs below projection.

Player adjustments:

Jordan Walker: −1.07.
Brandon Lowe: −0.93.
Spencer Horwitz: −0.80.
Leonardo Bernal: −0.78.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 7 info.
Primary/secondary causes: none.
Correct winner, wrong score. Skenes’ unexpectedly strong performance and broad offensive underproduction caused the six-run team-error total.

TB@NYY — correct winner, wrong score

Predicted: TB 2, NYY 4; NYY winner
Actual: TB 4, NYY 6; NYY winner
Published pick probability: NYY 0.531136
Confidence edge: +0.0623, lean
Bucket: lean
Brier: 0.219833
Team-run absolute error: 4
Total-run bias: −4
Predicted margin: NYY by 2; actual margin: NYY by 2

Edges:

Season: +0.185
Recent: −0.042
Pitcher: +0.0045
Composite: +0.4612

Starters:

Ian Seymour: projected 0 ER/15 outs, actual 0 ER/15 outs.
Cam Schlittler: projected 1 ER/19 outs, actual 1 ER/9 outs.
ER projections were exact; outs were wrong for Schlittler.

Controls and path:

Governed probability 0.5366; published NYY probability 0.531136.
No volatility change.
The total was four runs higher than projected, with both offenses contributing.

Player adjustments:

Austin Wells: +1.73.
George Lombard Jr.: +1.71.
Taylor Walls: +1.11.
Ben Rice: −1.40.
Cody Bellinger: −1.11.

Audit and causes:

Structural audit: PASS; 0 critical/error/warning, 7 info.
Primary/secondary causes: none.
Correct winner, wrong score. The winner call was solid; the run total was too low.

---

4. Miss reconstruction

Earliest wrong-direction stages for every loss

CIN@ATL

Historical model score: approximately CIN 3.76, ATL 5.68, initially favoring ATL—the wrong result.
Variance stage: moved to a tie, an intermediate winner-direction regression.
Enhanced prediction: moved toward CIN, which was directionally correct.
Published score/probability: stored final prediction again favors ATL, creating a trace inconsistency.
Diagnosis: bad initial weighting toward ATL, followed by documented pipeline instability; starter projections were not the principal issue.

LAA@SEA

Historical and early score path favored SEA.
Recent, season, pitcher, and composite edges all favored SEA.
The eventual actual result was LAA.
No clearly identifiable later stage moved the model toward LAA; the earliest wrong direction was the historical/form-weighted winner assessment.
Diagnosis: bad weighting or irreducible variance; not a starter miss.

NYM@TEX

Historical model was effectively TEX by a small amount: approximately NYM 4.32, TEX 4.35.
Variance stage moved to a tie.
Enhanced prediction moved to NYM, which was the wrong direction.
Later stages retained NYM.
Diagnosis: winner-direction regression during enhanced prediction after a near-even historical assessment; low run environment and Rocker’s unexpectedly strong start amplified the error.

SD@LAD

Historical and enhanced stages favored LAD by a small margin.
Actual SD won.
Published probability was only 0.5098 for LAD after experience shrinkage.
No stage shows a material, high-confidence directional failure.
Diagnosis: toss-up variance, with low scoring and unexpectedly strong Pivetta performance.

Cause classification

Loss	Bad input	Bad weighting	Calibration	Starter error	Offense/bullpen	Run environment	Irreducible variance
CIN@ATL	No clear defect	Yes: ATL season/recent conviction; pipeline trace instability	Moderate-confidence overcommitment	Limited; both starters close in ER	Some player deviations	Minor; total close	Some
LAA@SEA	No structural defect	Yes: form/quality favored SEA	Moderate-confidence overcommitment	No material starter miss	LAA offense exceeded projection	Minor	Material
NYM@TEX	No structural defect	Yes: enhanced stage moved from tie toward NYM	Lean call, not extreme	Rocker materially outperformed	NYM offense underproduced	Major; actual total 4 vs projected 9	Some
SD@LAD	No structural defect	No strong evidence of bad weighting	Correctly treated as toss-up	Pivetta outperformed	Both lineups underproduced	Moderate low-run environment	Major

The ARI@COL and CHW@KC stage regressions are not losses, but they demonstrate why final winner accuracy alone is insufficient: variance application lost three winner hits overall before enhanced prediction recovered two.

A 0.12 edge is moderate. It should not be described as a near coin flip, but it also does not represent a highly reliable prediction. On this slate both 0.12-confidence calls in CIN@ATL and LAA@SEA lost.

Pipeline-stage quality across all stages

Stage	Winner accuracy	Winner hits	Team-run MAE	Total-run MAE	Margin MAE
historical_ml	75.0%	9	2.5701	4.2504	2.2131
variance_applied	50.0%	6	2.5000	4.1667	2.1667
enhanced_prediction	66.67%	8	2.1667	3.8333	2.0000
embedded_savant_run_environment	66.67%	8	2.2083	3.7500	2.0833
fundamental_edge_target_review	66.67%	8	2.2083	3.7500	2.0833
capped_target	66.67%	8	2.2083	3.7500	2.0833
simulation_result	66.67%	8	2.2083	3.7500	2.0833
published_final	66.67%	8	2.2083	3.7500	2.0833

Transitions:

historical_ml → variance_applied: winner accuracy −25 percentage points, 3 fewer hits; CHW, MIA, and NYM moved to ties.
variance_applied → enhanced_prediction: winner accuracy +16.67 points, 2 recovered hits; CHW and MIA recovered, NYM moved to NYM.
enhanced_prediction → embedded_savant_run_environment: no winner changes; small run-metric deterioration.
All later transitions: no winner changes and no run-metric changes.

A later recovery does not erase the variance-stage regression. The system must preserve both before- and after-stage winner states.

---

5. Cross-game counterfactuals

Lever tension: widen run distributions

Lever: score.variance_multiplier / score.variance_floor

Helps:
CIN@ATL: lost a 0.12-confidence pick; wider tails could reduce overconfidence.
LAA@SEA: same rationale.
Hurts: none explicitly identified in the tension register.
Net evidence: 2 helped, 0 hurt on this slate.

This supports a bounded shadow experiment, not an unrestricted increase. The lever primarily changes spread and tail behavior; it should not be used as a substitute for fixing the near-zero run-model signal.

Lever tension: pitcher form weights and prior-game weighting

Lever: player.pitcher.v4 form weights / player.pitcher_prior_games

Helps:
ARI@COL: Gordon ER error 7.
HOU@OAK: Lambert ER error 3.
MIL@PHI: Painter ER error 4.
NYM@TEX: Rocker ER error 3.
STL@PIT: Skenes ER error 3.
Hurts:
CIN@ATL: both starter projections were reasonably accurate.
LAA@SEA: both starters were close.
TB@NYY: ER projections were exact.
Net evidence: 5 helped, 3 hurt.

The net slate count favors some investigation, but cross-slate starter_era_error is explicitly nonpersistent: only 5 of 12 prior slates had the same sign, with z = 1.04. Therefore, do not make a direct permanent weight increase yet. Use a shadow test only if the system can isolate credible venue/form evidence.

Lever tension: recent and season run-mean weighting

Lever: recent.* / season.* run-mean weighting

Helps:
ARI@COL
CHW@KC
CLE@BOS
MIA@CHC
NYM@TEX
STL@PIT
TB@NYY
Hurts:
HOU@OAK, where team-run error was already zero.
Net evidence: 7 helped, 1 hurt.

This is directionally favorable on the slate, but it must be reconciled with the persistent cross-slate finding that projected lineup runs are too high. The safer response is to lower the projected run environment or widen tails in a bounded experiment, not simply increase recent/season weight without examining feature bias.

Tempting one-game fixes that must not be made

Do not reverse the SEA pick solely because LAA won; the starter lines were close and the game was a moderate-confidence call.
Do not permanently increase pitcher weights because of Tanner Gordon, Randy Dobnak, Peter Lambert, or Paul Skenes; these include outlier starts and mixed directional effects.
Do not tune directly to individual player deviations. Of 77 material player deviations, only 16, or 20.8%, were supported by the hitter’s own season contact quality.
Do not treat CLE@BOS or SD@LAD as evidence that all low-confidence picks should be reversed; both were genuine toss-up outcomes.
Do not use the exact score of HOU@OAK to conclude the run model is healthy; it had exact team runs despite a 3-run starter error for Lambert.
Do not make a permanent calibration change from the 12-game bucket results alone.

---

6. Data and contract defects

Audit findings by severity

Across all supplied structural audits:

Critical: 0
Error: 0
Warning: 0
Info: 91

Every individual game reported structural audit PASS with zero critical, error, and warning findings.

Critical and error defects

There are no critical or error findings in the game artifacts.

The missing integrity_revision_audit is a manifest-level missing source, not a reported critical finding inside the game audits. It remains a provenance deficiency.

Warnings

There are no warning findings.

Informational findings

The principal informational findings are:

Eight to nine informational notes per game regarding forecast-file integrity and related diagnostics.
Five material calibrated-to-published probability movements with named transformations.
Zero simulation winner flips on this slate.
Complete 12/12 structural coverage.
Starter experience and venue-split diagnostics where applicable.
Large Savant run-environment swings on several games.

Prediction-model defects versus presentation/artifact bugs

Prediction-model/process defects

Near-zero run-model holdout signal:
Total \(R²\): 0.003064
Home \(R²\): 0.003987
Away \(R²\): 0.034855
Variance stage winner-direction regression: 9 historical hits to 6 variance-stage hits.
Persistent simulation inertness across 11 consecutive slates.
Persistent broad run-model weakness across 11 consecutive slates.
Current-slate calibration error of 0.1701.
Persistent negative lineup-run delta across 9 of 12 slates, mean −0.328, z = −2.25.

Presentation/provenance defects

Missing integrity revision audit.
Incomplete per-player starter-line fields for Eduardo Rodriguez and David Sandlin.
Ambiguous later-stage winner transition reporting, particularly for CIN@ATL.
Probability and score paths are not always sufficiently explicit to reconcile final side selection from every intermediate state.

These presentation/trace defects matter because they limit forensic reproducibility, even though they do not prove that the published forecast artifact was corrupted.

Coverage reconciliation

The slate is structurally covered:

12 expected and 12 audited.
12 starter comparisons.
12 continuous-learning records.
12 forensic records.

“Complete coverage” means every game was audited; it does not mean every prediction was good. The coverage finding is therefore valid but cannot override the high-severity model findings.

Probability-lineage audit

The supplied lineage includes:

Raw model pick probability.
Calibrated pick probability and empirical sample size.
Governor input and output.
Score-coherence probability.
Recency guardrail.
Published probability.

Material examples:

ARI@COL: raw 0.716, calibrated 0.661, score coherence 0.617535, published pick probability 0.56; named score-coherence and matchup-recency transformation.
CHW@KC: calibrated 0.620, score coherence 0.619987, published 0.54462; named transformation.
CIN@ATL: calibrated 0.5973, score coherence 0.585689, published 0.56.
HOU@OAK: calibrated 0.6344, score coherence 0.634358, published 0.56.
LAA@SEA: calibrated 0.6366, score coherence 0.605337, published 0.56.
CLE@BOS: published probability matches the calibrated/governed value.
NYM@TEX: no recency guardrail; published probability matches calibrated/governed value.
SD@LAD: experience shrinkage reduced the probability before publication.

The evidence records named transformations for all five material movements and marks unexplained_material_delta: false. No probability-governor bypass is established.

There is, however, a separate probability-reversion event:

CLE@BOS: blend favored home and changed the winner flag, with blended home probability 0.506608, trained home probability 0.521703, and published home probability 0.49381.
The published side was away and the outcome was an away hit.
One blend-stage overwrite occurred across 12 artifacts.

This was not a final prediction failure, but it warrants continued trace visibility.

---

7. Cross-slate persistence

The evidence includes 10 prior slates and an 11-slate trend window from 2026-09-14 through 2026-09-24.

Finding streaks

Finding	Consecutive-slate count	Present now?	Interpretation
`high_raw_probability_cluster`	0	No	Appeared in 3 slates within the window, but not consecutively present now
`pipeline_winner_instability`	1	Yes	Current-slate issue, also incidence in 6/11 slates; not yet escalated as a persistent streak
`simulation_inert_on_slate`	11	Yes	Established persistent failure
`single_slate_calibration_instability`	1	Yes	Current-slate issue; incidence in 8/11, but not an 11-slate consecutive streak
`weak_run_model_signal`	11	Yes	Established persistent failure

Escalated findings

Simulation inertness

Streak: 11 consecutive slates
Target lever: simulation.allow_score
Parameter-actionable: No
Metric harmed: team_run_mae, total_run_mae
Direction: confirm the simulation is running and not silently no-oping.

The self-assessment gives a concrete suspected root cause: the simulation distribution is built from the same draws used to find a target-realizing play-by-play path; the loop can early-exit after roughly two selected draws. That makes the derived win rate non-probabilistic and explains repeated zero flip candidates.

Weak run-model signal

Streak: 11 consecutive slates
Target levers registered: training.xgb.n_estimators, training.xgb.max_depth, training.xgb.learning_rate
Parameter-actionable: explicitly No for the current diagnosis
Metrics harmed: team_run_mae, total_run_mae
Direction: investigate features and target through an isolated training backtest.

The recommended action is a code/data change, not nightly hyperparameter tuning. Candidate features named in the evidence are:

Source-labelled no-vig market probability when available
Park/weather run environment
Bullpen rest/fatigue
Platoon-adjusted lineup quality

Metric trends

Every supplied metric trend is labeled improving, but the levels remain weak:

Margin bias: improving; mean −1.4833
Mean team-run error on losses: improving; mean 5.3237
Weighted calibration error: improving; mean 0.2081
Winner accuracy: improving; mean 0.5556

“Improving” does not mean solved. In particular, the current weighted calibration error is still 0.1701 and run-model \(R²\) remains near zero.

Top-confidence-pick streak

Current top-confidence loss streak: 0
The 11-slate sequence ends with hits on 2026-09-23 and 2026-09-24.
The sequence still contains misses on 2026-09-14, 2026-09-19, and 2026-09-21–22, so the absence of a current streak is not evidence that top-confidence reliability is strong.

Cross-slate structural blocks

Margin distribution

Current slate:

Actual margins ≥ 4: 3
Predicted margins ≤ 2: 12/12
Actual margins: [1,1,1,1,2,2,2,2,2,4,4,8]
The model systematically compressed tails, especially in ARI@COL, CHW@KC, and MIL@PHI.

Pipeline transitions

The recurring concern is not simply final accuracy; it is intermediate winner-direction regression. Current slate evidence directly confirms the variance-stage regression.

Calibration monitoring

Current bucket behavior is mixed:

Underconfidence at 0.50–0.55.
Overconfidence at 0.55–0.60.
Overall weighted error remains high.

High raw cluster

Three current-slate games in the high raw cluster.
Hits: 2
Misses: 1
Current finding streak: 0, so it is not an escalated persistent failure.

Simulation control

12 games observed.
Zero simulation flip candidates.
Zero published flips.
Current status: diagnostic only without independent support.
Cross-slate streak: 11, so this is an established code/diagnostic failure rather than single-slate noise.

---

8. Adjustments for today

Recommendations are ranked by persistence and scope.

1. Fix and independently validate simulation — highest evidence strength

Registered target: simulation.allow_score
Direction: code/diagnostic action; do not force a numeric parameter move
Bounded change: separate score-target realization from winner-probability estimation. Run an independent, unsteered batch of approximately 200–500 draws, or a closed-form Skellam/Poisson-difference calculation, without early exit. Register batch size.
Expected metric: more informative simulation contribution; improve team_run_mae, total_run_mae, and especially winner-direction diagnostics.
Helps: all games where simulation was inert, particularly close games such as CIN@ATL, NYM@TEX, and SD@LAD.
Hurts/regression risk: additional compute and possible score-distribution noise; no lever-tension evidence shows a specific game harmed.
Rollback monitor: retain the current no-op path as a control and compare flip rate, winner accuracy, Brier, and run MAE over a shadow slate window.

This is the safest persistent-pattern action because the evidence identifies an architectural cause and explicitly says not to force a parameter move.

2. Isolated run-model feature/target backtest — persistent model weakness

Registered targets: training.xgb.n_estimators, training.xgb.max_depth, training.xgb.learning_rate
Direction: no direct nightly parameter move; code/data investigation first.
Bounded change: add the named feature-contract candidates and retrain under a new model schema version; compare same-fingerprint holdout \(R²\) and MAE before any promotion.
Expected metrics: increase away/home/total run \(R²\), reduce team_run_mae and total_run_mae.
Helps: all games affected by compressed margins and poor score forecasts, especially ARI@COL, CHW@KC, MIL@PHI, NYM@TEX, STL@PIT, and TB@NYY.
Hurts/regression risk: market or weather features may overfit or become unavailable; require source-labelled availability and rollback if holdout performance does not improve.

This is a code/data action because the evidence explicitly states that no bounded nightly knob can repair near-zero signal.

3. Bounded shadow increase in run variance — strongest single-slate actionable driver

Registered parameter: score.variance_multiplier or, if the implementation uses a floor as the spread lever, score.variance_floor
Direction: increase spread modestly, with the floor interpreted according to the system’s implementation.
Suggested magnitude: one small shadow step, for example +0.05 to score.variance_multiplier, rather than a large change.
Expected metric: reduce high-conviction miss rate and weighted calibration error; improve margin-tail coverage. Track Brier, calibration error, margin MAE, team-run MAE, and total-run MAE.
Helps: CIN@ATL and LAA@SEA, the two 0.12-confidence losses explicitly identified by the lever tension; also helps the broader compressed-margin pattern.
Hurts: no explicit current-slate hurt in the lever register, but HOU@OAK had exact team runs and could be disturbed; all correctly scored close games are potential regression cases.
Gate: shadow only; rollback if Brier, total-run MAE, or winner accuracy deteriorates over the registered evaluation window.

This is not a claim that wider variance would have changed the actual winners. It is a bounded calibration/spread experiment motivated by the 3/12 large-margin outcomes and 0/12 predicted margins above two runs.

4. Hold pitcher-weight changes

Registered lever: player.pitcher.v4 form weights / player.pitcher_prior_games
Direction: abstain for now
Reason: the current slate gives a 5-help versus 3-hurt tension, but cross-slate starter ERA bias is nonpersistent.
Diagnostic action: evaluate credible venue splits and prior-game sample sizes in a shadow report.
Expected metric if later justified: starter_er_mae, starter_outs_mae, and indirectly total_run_mae.
Risk: disturbing accurate starter projections in CIN@ATL, LAA@SEA, and TB@NYY.

5. Do not change confidence calibration weights permanently

Registered levers: confidence.empirical_max_weight, confidence.base_cap, confidence.moderate_cap
Direction: no permanent move from this slate.
Reason: current calibration error is high, but the 12-game bucket sample has conflicting underconfidence and overconfidence behavior, and the causal register explicitly requires a rolling-window watch.
Shadow option: if the system’s existing auto-action mechanism permits it, test a small cap reduction paired with the variance experiment; evaluate Brier and calibration error separately rather than using winner accuracy alone.

---

9. Bottom line

Known

The slate produced 8 correct winners out of 12.
Winner accuracy exceeded the stated walk-forward baseline, but that does not make the prediction system well calibrated or numerically accurate.
Exact and team-run scores were materially weaker than winners; predicted margins were too compressed.
The run model has a persistent near-zero holdout signal across 11 slates.
The variance stage materially degraded winner direction from 9/12 to 6/12 before later recovery.
Published probability movements have recorded transformations; no material probability bypass is established.
All 12 games received structural QA with zero critical, error, or warning findings.
Two persistent escalated failures are established: simulation inertness and weak run-model signal.
The current slate’s largest practical model risk is overconfident spread/run behavior, especially on moderate-confidence calls.

Uncertain

Whether changing pitcher venue/form weighting would improve future performance; cross-slate evidence is not persistent enough.
Whether the current 0.1701 calibration error represents a stable calibration curve problem or slate noise.
How much a real independent simulation distribution would change winner probabilities, because the current simulation has not supplied an independent signal.
The exact later-stage winner path for CIN@ATL, because the stored transition summary and per-game published result do not fully reconcile.
Whether the missing integrity-revision audit contains any additional provenance information.

Single safest next action

Run the bounded simulation code/diagnostic repair in shadow: separate independent winner-probability draws from target-realization draws, remove the early-exit dependence, register the batch size, and compare its winner/Brier/run-metric contribution against the current control before changing model parameters.

Team	Player	Role	Net effect	Largest change
HOU	Christian Walker	batter / 1B	worse	predicted_SO 3.04
OAK	Jeff McNeil	batter / 1B	better	predicted_RBI 2.594
HOU	Lucas Spence	batter / CF	better	predicted_H 2.538
OAK	Jonah Heim	batter / C	worse	predicted_SO 2.324
OAK	Carlos Cortes	batter / LF	worse	predicted_AB 2.154
OAK	Shea Langeliers	batter / DH	boosted	predicted_SO 1.98
OAK	Joshua Kuroda-Grauer	batter / SS	boosted	predicted_H 1.7
HOU	Yainer Diaz	batter / DH	better	predicted_RBI 1.523

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	OAK pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	OAK
info	CURRENT_FINALIZED_MATCH_BOX	HOU model-prediction batting totals differ from final simulation box score (AB: current=33, sim=41; H: current=9, sim=11; 2B: current=2, sim=4; HR: current=2, sim=1; RBI: current=5, sim=7; R: current=4, sim=7; BB: current=4, sim=1; SO: current=7, sim=13).	HOU
info	CURRENT_PLAYER_MATCH_BOX	HOU current finalized hitter lines are model predictions and differ from the simulation box score: Yordan Alvarez: 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=2, SO current=1 box=2; Yainer Diaz: AB current=4 box=5, H current=1 box=2, HR current=0 box=1, RBI current=1 box=2; Isaac Paredes: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, BB current=1 box=0; Christian Walker: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, SO current=1 box=4; Lucas Spence: AB current=3 box=4, H current=1 box=3, 2B current=0 box=1, RBI current=0 box=1; Jose Altuve: AB current=4 box=5, H current=1 box=2, 2B current=1 box=0, RBI current=0 box=1; Jeremy Peña: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Christian Vázquez: AB current=3 box=4, R current=0 box=1, SO current=0 box=1	HOU
info	CURRENT_FINALIZED_MATCH_BOX	OAK model pitcher projections differ from final simulation pitching box score (ER: current=6, sim=7).	OAK
info	CURRENT_FINALIZED_MATCH_BOX	OAK model-prediction batting totals differ from final simulation box score (AB: current=32, sim=42; H: current=9, sim=12; 2B: current=2, sim=3; R: current=4, sim=5; BB: current=3, sim=2; SO: current=7, sim=15; HBP: current=0, sim=1).	OAK
info	CURRENT_PLAYER_MATCH_BOX	OAK current finalized hitter lines are model predictions and differ from the simulation box score: Shea Langeliers: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, HR current=1 box=0; Henry Bolte: AB current=4 box=5, 2B current=0 box=1, R current=1 box=0, BB current=1 box=0; Jeff McNeil: H current=1 box=2, 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=3; Carlos Cortes: AB current=3 box=5, SO current=0 box=2; Joshua Kuroda-Grauer: AB current=3 box=5, H current=1 box=2, 2B current=0 box=1, RBI current=0 box=1; Lawrence Butler: H current=1 box=2, 2B current=1 box=0, RBI current=1 box=0; Donovan Walton: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Zack Gelof: AB current=4 box=5, H current=1 box=2, RBI current=1 box=0	OAK
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
LAA	Grayson Rodriguez	5 ER / 15 outs	4 ER / 15 outs	-1
SEA	Bryan Woo	3 ER / 17 outs	4 ER / 16 outs	1

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
LAA	Grayson Rodriguez (Starter)	15 / 15	7.2 / 9	4 / 5	5 / 5	1 / 2	5 / 3	1.75	2 / 5	5.15	venue away: 4.54 vs 7.1 ERA (ok) · opp run factor 1.1 (weak) · 19 prior appearances this season
LAA	Ryan Watson (Setup Man)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.25	2 / 0	1.14
SEA	Bryan Woo (Starter)	16 / 17	6.75 / 4.76	4 / 3	7 / 3	0 / 0	3 / 1	1.36	8 / 7	5.88	venue home: 2.36 vs 5.22 ERA (ok) · opp run factor 0.96 (weak) · 29 prior appearances this season

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Dominic Canzone (DH)	SEA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-1.4	0.356 elite	bad day
Julio Rodríguez (CF)	SEA	4 / 4	2 / 1	0 / 0	0 / 0	1 / 0	2 / 1	1 / 1	0 / 0	1 / 1	1.4	0.334 above average	projection too low
Josh Lowe (RF)	LAA	3 / 3	2 / 0	1 / 0	0 / 0	0 / 0	0 / 0	3 / 0	0 / 0	0 / 1	1.25	0.285 below average	good day
Christian Moore (2B)	LAA	4 / 3	2 / 1	1 / 0	0 / 0	0 / 0	3 / 0	1 / 0	0 / 0	1 / 2	0.78	0.231 weak	good day
Zach Neto (SS)	LAA	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	1 / 0	0 / 1	1 / 0	2 / 2	-0.6	0.313 average	no signal
Vaughn Grissom (1B)	LAA	5 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.311 average	—
Bryce Teodosio (CF)	LAA	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	2 / 0	0 / 0	0 / 0	1 / 1	0.47	0.203 weak	—
J.P. Crawford (SS)	SEA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	-0.47	0.32 average	—
Lazaro Montes (RF)	SEA	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.47	0.319 average	—
Leo Rivas (3B)	SEA	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0 / 1	0.47	0.25 weak	—
Moisés Ballesteros (C)	LAA	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	3 / 1	0.33	0.293 below average	—
Randy Arozarena (LF)	SEA	4 / 4	1 / 1	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.33	0.351 elite	—
Cole Young (2B)	SEA	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0.33	0.312 average	—
Wade Meckler (LF)	LAA	4 / 4	2 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	2 / 1	0.16	0.314 average	—
Mike Trout (DH)	LAA	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	1 / 1	-0.14	0.38 elite	—
Denzer Guzman (3B)	LAA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 0	2 / 1	-0.14	0.268 weak	—
Cal Raleigh (C)	SEA	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	2 / 1	0 / 1	-0.14	0.31 average	—
Josh Naylor (1B)	SEA	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 1	-0.14	0.312 average	—

Signal	Home-away edge	Meaning
Season quality	0.404	positive favors home; negative favors away
Recent form	0.614	positive favors home; negative favors away
Starting-pitcher matchup	0.139	positive favors home; negative favors away
Composite	0.863	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	3.53-5.1	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	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

Stage	Before	After	Toward final pick	What changed
variance_applied	3.53-5.1	4-5	-0.575	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

Team	Player	Role	Net effect	Largest change
SEA	Josh Naylor	batter / 1B	better	predicted_RBI 2.57
LAA	Ben Joyce	pitcher / Closer	better	predicted_IP 1.9
SEA	Randy Arozarena	batter / LF	worse	predicted_AB -1.7
SEA	Lazaro Montes	batter / RF	better	predicted_RBI 1.7
LAA	Zach Neto	batter / SS	deboosted	predicted_RBI 1.64
LAA	Bryce Teodosio	batter / CF	worse	predicted_AB 1.5
SEA	Dominic Canzone	batter / DH	worse	predicted_H -1.42
LAA	Moisés Ballesteros	batter / C	better	predicted_H 1.33

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=34; 2B: current=1, sim=2; HR: current=1, sim=0; RBI: current=3, sim=4; R: current=3, sim=4; BB: current=2, sim=1; HBP: current=0, sim=1).	LAA
info	CURRENT_PLAYER_MATCH_BOX	LAA current finalized hitter lines are model predictions and differ from the simulation box score: Wade Meckler: H current=1 box=0, 2B current=1 box=0, BB current=1 box=0; Mike Trout: AB current=4 box=3, H current=1 box=0, R current=1 box=0, BB current=1 box=0; Vaughn Grissom: RBI current=1 box=0; Moisés Ballesteros: H current=1 box=2, 2B current=0 box=1, R current=0 box=1, SO current=1 box=0; Josh Lowe: H current=0 box=1, RBI current=0 box=1, R current=0 box=1, BB current=0 box=1; Bryce Teodosio: AB current=3 box=4, SO current=1 box=2; Zach Neto: HR current=1 box=0, RBI current=0 box=2, SO current=2 box=1; Christian Moore: AB current=3 box=4, H current=1 box=0	LAA
info	CURRENT_FINALIZED_MATCH_BOX	SEA model-prediction batting totals differ from final simulation box score (AB: current=34, sim=29; H: current=8, sim=5; 2B: current=1, sim=0; HR: current=1, sim=2; RBI: current=4, sim=5; R: current=4, sim=5; SO: current=9, sim=6).	SEA
info	CURRENT_PLAYER_MATCH_BOX	SEA current finalized hitter lines are model predictions and differ from the simulation box score: J.P. Crawford: H current=1 box=0, R current=1 box=0, BB 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: AB current=4 box=2, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Cal Raleigh: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0; Cole Young: AB current=4 box=3, R current=0 box=1; Julio Rodríguez: RBI current=1 box=0, SO current=1 box=0; Lazaro Montes: HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Josh Naylor: AB current=4 box=3, HR current=0 box=1, RBI current=0 box=3, R current=0 box=1	SEA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
MIA	Tyler Phillips	1 ER / 15 outs	0 ER / 12 outs	-1
CHC	Matthew Boyd	3 ER / 17 outs	1 ER / 21 outs	-2

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIA	Tyler Phillips (Starter)	12 / 15	0 / 1.8	0 / 1	3 / 4	0 / 0	4 / 5	1.77	2 / 3	3.61	venue away: 5.26 vs 3.17 ERA (worse here) · opp run factor 1.04 (strong) · 36 prior appearances this season
MIA	Victor Vodnik (Middle Reliever (High Leverage))	3 / 2	0 / 13.5 (-13.50)	0 / 1	0 / 1	0 / 1	0 / 1	0.79	1 / 1	0.99
CHC	Matthew Boyd (Starter)	21 / 17	1.29 / 4.76 (-3.47)	1 / 3	5 / 4	0 / 0	2 / 1	2.04	3 / 1	4.28	venue home: 5.53 vs 2.86 ERA (worse here) · opp run factor 0.96 (average) · 21 prior appearances this season
CHC	Daniel Palencia (Middle Reliever (High Leverage))	3 / 5	0 / 5.4 (-5.40)	0 / 1	0 / 3	0 / 0	1 / 0	0.37	0 / 1	0.98
CHC	Ryan Zeferjahn (Setup Man)	3 / 1	0 / 0	0 / 0	0 / 2	0 / 0	0 / 1	0.65	0 / 1	1.36

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Heriberto Hernández (LF)	MIA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	1 / 1	-1.73	0.348 elite	bad day
Seiya Suzuki (RF)	CHC	3 / 4	2 / 1	0 / 1	0 / 0	1 / 0	1 / 1	1 / 1	1 / 1	1 / 1	1.09	0.337 above average	projection too low
Pete Crow-Armstrong (CF)	CHC	3 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	1 / 1	-0.93	0.376 elite	bad day
Michael Busch (1B)	CHC	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	-0.8	0.333 above average	bad day
Brian Navarreto (C)	MIA	3 / 3	1 / 0	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0.78	0.176 weak	good day
Ian Happ (LF)	CHC	3 / 4	1 / 1	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	1 / 1	0 / 1	0.62	0.315 average	no signal
Deyvison De Los Santos (1B)	MIA	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.173 weak	—
Jared Serna (2B)	MIA	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	-0.47	0.214 weak	—
Carson Kelly (C)	CHC	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 0	-0.47	0.313 average	—
Agustín Ramírez (DH)	MIA	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.33	0.284 below average	—
Otto Lopez (SS)	MIA	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 0	0 / 0	0.33	0.327 above average	—
Pedro Ramirez (3B)	CHC	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	0.33	0.282 weak	—
Javier Sanoja (3B)	MIA	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 0	-0.31	0.294 below average	—
Esteury Ruiz (RF)	MIA	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	-0.14	0.276 weak	—
Michael Conforto (DH)	CHC	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0 / 1	-0.14	0.336 above average	—
Jakob Marsee (CF)	MIA	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	1 / 1	0 / 1	0	0.31 average	—
Nico Hoerner (2B)	CHC	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0 / 0	0	0.323 above average	—
Dansby Swanson (SS)	CHC	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0	0.305 average	—

Signal	Home-away edge	Meaning
Season quality	0.626	positive favors home; negative favors away
Recent form	-0.098	positive favors home; negative favors away
Starting-pitcher matchup	0.081	positive favors home; negative favors away
Composite	0.677	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.61-4.84	n/a	Base ML historical model prediction
variance_applied	5-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.519	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

Stage	Before	After	Toward final pick	What changed
variance_applied	4.61-4.84	5-5	-0.231	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-5	4-5	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	4-5	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	4-5	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	4-5	0	Capped at max 12 realistic runs
simulation_result	4-5	4-5	0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
MIA	Esteury Ruiz	batter / RF	worse	predicted_AB 2.5
CHC	Michael Conforto	batter / DH	better	predicted_H 2.24
CHC	Pedro Ramirez	batter / 3B	worse	predicted_AB 1.877
CHC	Seiya Suzuki	batter / RF	better	predicted_AB -1.78
CHC	Carson Kelly	batter / C	worse	predicted_AB 1.69
MIA	Javier Sanoja	batter / 3B	better	predicted_H 1.65
CHC	Dansby Swanson	batter / SS	better	predicted_AB 1.555
MIA	Deyvison De Los Santos	batter / 1B	better	predicted_H 1.537

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	CHC pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CHC
info	CURRENT_PITCHER_DUPLICATION	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	CHC model-prediction batting totals differ from final simulation box score (AB: current=33, sim=38; HR: current=1, sim=2; BB: current=4, sim=6; SO: current=7, sim=5; 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: Michael Conforto: AB current=3 box=4, H current=1 box=3, R current=0 box=1, BB current=0 box=1; Seiya Suzuki: AB current=4 box=3, 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=2; Michael Busch: SO current=1 box=0; Pete Crow-Armstrong: H current=1 box=2; Pedro Ramirez: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0; Ian Happ: AB current=4 box=3, H current=1 box=0, R current=1 box=0, BB current=1 box=2; Nico Hoerner: AB current=4 box=5, H current=1 box=0; Dansby Swanson: AB current=3 box=5, H current=0 box=1, 2B current=0 box=1, R current=1 box=0	CHC
info	CURRENT_FINALIZED_MATCH_BOX	MIA model pitcher projections differ from final simulation pitching box score (H: current=9, sim=8).	MIA
info	CURRENT_FINALIZED_MATCH_BOX	MIA model-prediction batting totals differ from final simulation box score (AB: current=32, sim=40; H: current=8, sim=10; 2B: current=1, sim=2; 3B: current=0, sim=1; HR: current=1, sim=0; R: current=5, sim=4; BB: current=3, sim=2; SO: current=7, sim=5).	MIA
info	CURRENT_PLAYER_MATCH_BOX	MIA current finalized hitter lines are model predictions and differ from the simulation box score: Heriberto Hernández: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Javier Sanoja: AB current=4 box=5, H current=1 box=3, RBI current=1 box=0, R current=0 box=1; Agustín Ramírez: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=0; Deyvison De Los Santos: AB current=3 box=4, H current=1 box=2, RBI current=0 box=1, R current=0 box=1; Jared Serna: H current=1 box=2, 2B current=0 box=1, R current=1 box=0, BB current=0 box=1; Esteury Ruiz: AB current=3 box=5, H current=1 box=0; Otto Lopez: AB current=4 box=5, RBI current=1 box=2, SO current=0 box=1; Brian Navarreto: AB current=3 box=4, H current=0 box=1, 3B current=0 box=1, RBI current=0 box=1	MIA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
MIL	Shane Drohan	1 ER / 14 outs	1 ER / 8 outs	0
PHI	Andrew Painter	5 ER / 15 outs	1 ER / 18 outs	-4

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIL	Shane Drohan (Starter)	8 / 14	3.38 / 1.93	1 / 1	3 / 3	0 / 0	2 / 2	2.36	2 / 3	4.75	venue away: 4.61 vs 5.54 ERA (ok) · opp run factor 1.01 (average) · 31 prior appearances this season
MIL	Aaron Ashby (Middle Reliever (High Leverage))	6 / 3	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0.54	2 / 0	1.1
PHI	Andrew Painter (Starter)	18 / 15	1.5 / 9 (-7.50)	1 / 5	5 / 7	0 / 0	3 / 2	1.97	8 / 3	4.37	venue home: 6.56 vs 4.88 ERA (worse here) · opp run factor 1 (strong) · 24 prior appearances this season
PHI	Tim Mayza (Middle Reliever (High Leverage))	3 / 3	9 / 0 (+9.00)	1 / 0	2 / 0	1 / 0	0 / 0	0.28	0 / 1	0.73

Deep slate report — 20260924

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 (12)

ARI@COLHIT

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@KCHIT

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

CIN@ATLMISS

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

CLE@BOSHIT

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

HOU@OAKHIT

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

LAA@SEAMISS

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@CHCHIT

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@PHIHIT

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

NYM@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

SD@LADMISS

Player-level breakdown (projected vs actual)

`pipeline_winner_instability` — High

`single_slate_calibration_instability` — High

`weak_run_model_signal` — High

`complete_slate_audit_coverage` — Info

`probability_publication_delta_explained` — Info

`simulation_inert_on_slate` — Info on this slate, escalated persistent failure across slates

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Sal Frelick (LF)	MIL	2 / 3	1 / 1	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	2 / 0	0 / 0	0.97	0.279 weak	good day
Cooper Pratt (SS)	MIL	3 / 3	2 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0 / 1	0.8	0.314 average	no signal
Trea Turner (SS)	PHI	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	-0.78	0.294 below average	projection too high
Brice Turang (2B)	MIL	5 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.64	0.344 elite	bad day
Jake Bauers (1B)	MIL	5 / 4	2 / 1	1 / 0	0 / 0	0 / 1	1 / 1	1 / 1	0 / 1	2 / 1	-0.48	0.376 elite	—
Kyle Schwarber (DH)	PHI	4 / 4	2 / 1	1 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	0 / 1	-0.48	0.364 elite	—
Gary Sánchez (C)	MIL	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 0	1 / 1	2 / 1	0.47	0.339 elite	—
Bryson Stott (3B)	PHI	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	-0.47	0.328 above average	—
Luis Arraez (2B)	PHI	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	-0.47	0.308 average	—
Justin Crawford (CF)	PHI	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.284 below average	—
J.T. Realmuto (C)	PHI	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	0.33	0.317 average	—
William Contreras (DH)	MIL	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 0	0 / 1	-0.14	0.333 above average	—
Bryce Harper (RF)	PHI	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	0 / 1	0.14	0.38 elite	—
Alec Bohm (1B)	PHI	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	-0.14	0.33 above average	—
Bryan De La Cruz (LF)	PHI	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	-0.14	0.288 below average	—
Luis Lara (RF)	MIL	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	0	0.294 below average	—
Joey Ortiz (3B)	MIL	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	2 / 1	0	0.295 below average	—
Garrett Mitchell (CF)	MIL	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	4 / 1	0	0.352 elite	—

Signal	Home-away edge	Meaning
Season quality	-0.955	positive favors home; negative favors away
Recent form	-0.53	positive favors home; negative favors away
Starting-pitcher matchup	0.006	positive favors home; negative favors away
Composite	-0.825	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.56-3.87	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.539	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

Stage	Before	After	Toward final pick	What changed
variance_applied	4.56-3.87	5-4	0.312	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

Team	Player	Role	Net effect	Largest change
PHI	Luis Arraez	batter / 2B	deboosted	predicted_RBI 2.52
PHI	Trea Turner	batter / SS	worse	predicted_AB -2.11
MIL	Jake Bauers	batter / 1B	better	predicted_RBI 1.57
PHI	Bryan De La Cruz	batter / LF	worse	predicted_AB 1.481
PHI	Kyle Schwarber	batter / DH	better	predicted_SO -1.453
PHI	Justin Crawford	batter / CF	worse	predicted_SO 1.432
MIL	Joey Ortiz	batter / 3B	worse	predicted_SO 1.298
MIL	Garrett Mitchell	batter / CF	worse	predicted_SO -1.18

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	PHI pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	PHI
info	CURRENT_FINALIZED_MATCH_BOX	MIL model-prediction batting totals differ from final simulation box score (AB: current=32, sim=34; H: current=8, sim=7; 2B: current=1, sim=2; HR: current=1, sim=0; SO: current=8, sim=6).	MIL
info	CURRENT_PLAYER_MATCH_BOX	MIL current finalized hitter lines are model predictions and differ from the simulation box score: William Contreras: RBI current=1 box=0, R current=1 box=0, SO current=1 box=0; Jake Bauers: AB current=4 box=3, 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=2; Luis Lara: 2B current=0 box=1, R current=0 box=1; Brice Turang: 2B current=1 box=0, BB current=1 box=0; Sal Frelick: H current=1 box=0, BB current=0 box=1; Joey Ortiz: AB current=3 box=4, H current=1 box=0, R current=1 box=0, SO current=1 box=2; Garrett Mitchell: AB current=4 box=5, SO current=1 box=0; Gary Sánchez: RBI current=1 box=0	MIL
info	CURRENT_FINALIZED_MATCH_BOX	PHI model-prediction batting totals differ from final simulation box score (AB: current=32, sim=33; H: current=8, sim=6; HR: current=1, sim=2; BB: current=3, sim=4; SO: current=8, sim=4).	PHI
info	CURRENT_PLAYER_MATCH_BOX	PHI current finalized hitter lines are model predictions and differ from the simulation box score: Bryce Harper: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Alec Bohm: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Bryan De La Cruz: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Bryson Stott: AB current=3 box=4, 2B current=0 box=1, BB current=1 box=0; Trea Turner: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, R current=1 box=0; J.T. Realmuto: AB current=3 box=4, H current=0 box=1, HR current=0 box=1, RBI current=0 box=1; Luis Arraez: H current=1 box=2, HR current=0 box=1, RBI current=0 box=3, R current=0 box=1; Kyle Schwarber: HR current=1 box=0, RBI current=1 box=0, SO current=1 box=0	PHI
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
NYM	Zac Thornton	2 ER / 17 outs	2 ER / 15 outs	0
TEX	Kumar Rocker	4 ER / 14 outs	1 ER / 15 outs	-3

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
NYM	Zac Thornton (Starter)	15 / 17	3.6 / 3.18	2 / 2	4 / 5	1 / 0	1 / 0	0.85	4 / 2	4.17	venue away: 2.86 vs 3.71 ERA (ok) · opp run factor 0.95 (average) · 14 prior appearances this season
TEX	Kumar Rocker (Starter)	15 / 14	1.8 / 7.71 (-5.91)	1 / 4	3 / 3	0 / 1	2 / 5	2.26	4 / 4	5.11	venue home: 4.61 vs 4.57 ERA (ok) · opp run factor 1.05 (average) · 29 prior appearances this season
TEX	Chase Silseth (Middle Reliever (High Leverage))	3 / 3	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0.35	2 / 1	0.97

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Josh Jung (3B)	TEX	3 / 4	3 / 1	1 / 1	0 / 0	1 / 0	2 / 0	1 / 1	1 / 0	0 / 0	2.2	0.343 elite	projection too low
Juan Soto (DH)	NYM	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	0 / 0	-1.4	0.413 elite	bad day
A.J. Ewing (CF)	NYM	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	-0.8	0.318 average	no signal
Nick Morabito (LF)	NYM	3 / 3	1 / 0	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0.78	0.31 average	no signal
Luis Torrens (C)	NYM	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.47	0.292 below average	—
Bo Bichette (3B)	NYM	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.47	0.328 above average	—
Jake Burger (1B)	TEX	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	-0.47	0.317 average	—
Ezequiel Duran (CF)	TEX	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	2 / 1	-0.47	0.303 average	—
Elias Díaz (C)	TEX	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.266 weak	—
Brandon Nimmo (RF)	TEX	2 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	-0.45	0.342 elite	—
Francisco Lindor (SS)	NYM	4 / 4	1 / 1	1 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	1 / 1	-0.33	0.332 above average	—
Corey Seager (SS)	TEX	4 / 4	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	2 / 1	-0.33	0.346 elite	—
Marcus Semien (2B)	NYM	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	-0.14	0.303 average	—
Wyatt Langford (LF)	TEX	3 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0.14	0.322 above average	—
Carson Benge (RF)	NYM	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	1 / 1	0	0.339 elite	—
Brett Baty (1B)	NYM	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.32 average	—
Justin Foscue (2B)	TEX	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	0 / 1	0	0.327 above average	—
Danny Jansen (DH)	TEX	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.268 weak	—

Signal	Home-away edge	Meaning
Season quality	0.11	positive favors home; negative favors away
Recent form	-0.018	positive favors home; negative favors away
Starting-pitcher matchup	-0.1	positive favors home; negative favors away
Composite	-0.344	positive favors home; negative favors away