Deep slate report 20260907

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_20260907.json	277792
forensic deep audit	loaded	probability, provenance, starter and decision contracts	/app/output/reports/PREDICTION_SYSTEM_DEEP_AUDIT_20260907.json	180869
continuous learning	loaded	pipeline-stage quality, calibration, probability lineage, score distribution, and official starter comparisons	/app/output/learning/continuous_learning_daily_20260907.json	1791725
detailed report	loaded	full game and simulation rendering	/app/output/reports/detailed_20260907.html	259218
ultra detailed player audit	loaded	player-level prediction trace	/app/output/reports/20260907_ultra_detailed_player_audit.html	22292012
structural audit	loaded	all invariant findings	/app/output/reports/audit_report_20260907.html	1318983
integrity revision audit	missing	post-Luna remediation proof for immutable revision artifacts	/app/output/reports/audit_report_20260907_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 (11)

Severity	Domain	Finding and evidence	Affected	Action & guardrail
CRITICAL	probability_lineage	Published probability differs materially without a recorded transformation 4 game(s) have an unexplained ≥2pp calibrated-to-published delta	ARI@KC, CIN@LAD, LAA@BOS, NYM@MIA	Do thisBlock any promotion until the missing probability transformation is recorded and reconciled. Don't concludeDo not assume the probability governor was bypassed until you have traced the pick-side math.
HIGH	calibration	Published confidence buckets were poorly aligned on this slate weighted absolute calibration error 0.2611 across 11 games	ARI@KC, ATL@PHI, CHC@MIL, CIN@LAD, CLE@BAL, LAA@BOS, MIN@DET, NYM@MIA, STL@SF, TOR@OAK, WAS@SD	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 = 0.0124/0.0205/0.0184	ARI@KC, ATL@PHI, CHC@MIL, CIN@LAD, CLE@BAL, LAA@BOS, MIN@DET, NYM@MIA, STL@SF, TOR@OAK, WAS@SD	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 11/11 games structurally audited	ARI@KC, ATL@PHI, CHC@MIL, CIN@LAD, CLE@BAL, LAA@BOS, MIN@DET, NYM@MIA, STL@SF, TOR@OAK, WAS@SD	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	simulation	Simulation produced no winner-direction intervention on this slate flip candidates 0; published flips 0	ARI@KC, ATL@PHI, CHC@MIL, CIN@LAD, CLE@BAL, LAA@BOS, MIN@DET, NYM@MIA, STL@SF, TOR@OAK, WAS@SD	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	0	4/11	no	code/diagnostic
score_margin_compression	0	3/11	no	param
simulation_inert_on_slate	11	11/11	yes	code/diagnostic
single_slate_calibration_instability	1	6/11	yes	param
weak_run_model_signal	11	11/11	yes	code/diagnostic

Metric	Direction	Series (old → new)
winner_accuracy	flat	[0.4667, 0.9412, 0.5, 0.6667, 0.4667, 0.3333, 0.5556, 0.625, 0.5333, 0.6, 0.5455]
weighted_calibration_error	regressing	[0.2179, 0.3961, 0.0693, 0.1661, 0.2476, 0.21, 0.1474, 0.1316, 0.1, 0.0937, 0.2611]
margin_bias	improving	[-1.333, -1.882, -2.786, -3.5, -2.533, -3.067, -1.778, -0.688, -1.467, -1.467, -0.182]
team_run_error_on_losses	improving	[5.0, 3.0, 7.0, 8.25, 7.125, 6.4, 6.75, 6.0, 5.286, 5.333, 3.8]

Stage	Winner hits	Accuracy	Team-run MAE	Total-run MAE	Margin MAE
historical_ml	6/11	54.5%	1.578	2.342	2.195
variance_applied	5/11	45.5%	1.591	2.454	2.091
enhanced_prediction	6/11	54.5%	1.818	2.909	2
embedded_savant_run_environment	6/11	54.5%	1.909	3.091	2.182
fundamental_edge_target_review	6/11	54.5%	1.909	3.091	2.182
capped_target	6/11	54.5%	1.909	3.091	2.182
simulation_result	6/11	54.5%	1.909	3.091	2.182
published_final	6/11	54.5%	1.909	3.091	2.182

Transition	Δ accuracy	Δ hits	Δ team-run MAE	Δ margin MAE	Games whose winner/tie changed
historical_ml → variance_applied	-0.091	-1	0.0127	-0.1041	CIN@LAD: CIN → TIE; CLE@BAL: BAL → TIE; MIN@DET: MIN → TIE
variance_applied → enhanced_prediction	0.091	1	0.2273	-0.0909	CIN@LAD: TIE → LAD; CLE@BAL: TIE → BAL; MIN@DET: TIE → MIN; STL@SF: SF → STL
enhanced_prediction → embedded_savant_run_environment	0	0	0.0909	0.1818
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@KC	miss	0.5978	0.618	0.618	0.618	0.5796	no recorded transformation explains the publication delta
ATL@PHI	hit	0.5995	0.5565	0.5565	0.5565	0.53	score coherence followed by matchup-recency guardrail
CHC@MIL	hit	0.504	0.508	0.508	0.508	0.5068	published probability matches the calibrated/governed value
CIN@LAD	hit	0.5322	0.5751	0.5751	0.5751	0.5326	no recorded transformation explains the publication delta
CLE@BAL	hit	0.5133	0.5107	0.5107	0.5107	0.5078	published probability matches the calibrated/governed value
LAA@BOS	hit	0.6371	0.6539	0.6539	0.6539	0.6231	no recorded transformation explains the publication delta
MIN@DET	miss	0.5258	0.5675	0.5675	0.5675	0.53	score coherence followed by matchup-recency guardrail
NYM@MIA	miss	0.582	0.5477	0.5477	0.5477	0.5208	no recorded transformation explains the publication delta
STL@SF	miss	0.5409	0.6268	0.6268	0.6268	0.6268	published probability matches the calibrated/governed value
TOR@OAK	miss	0.7068	0.6996	0.6996	0.6996	0.6996	published probability matches the calibrated/governed value
WAS@SD	hit	0.6299	0.5724	0.5724	0.5724	0.5297	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
22	5	4	0	13	-0.86	-0.36

Slate	Games	Winner acc	Mean starter ERA error	Mean lineup runs vs proj	Sides broke on pitching	Sides broke on offense
20260906	15	0.600	0.74	-0.30	6	10
20260905	15	0.533	-1.17	-0.07	2	6
20260904	16	0.625	-0.23	-0.44	5	12
20260903	9	0.556	0.24	-0.83	4	6
20260902	15	0.333	0.72	0.30	7	6
20260901	15	0.467	1.48	1.07	9	3
20260831	12	0.667	0.83	0.08	2	7
20260830	14	0.500	-0.36	-0.29	3	7
20260829	17	0.941	—	—
20260828	15	0.467	—	—

Game	Pred margin	Actual margin	Pred − actual
ARI@KC	2	1	1
ATL@PHI	1	1	0
CHC@MIL	2	1	1
CIN@LAD	1	3	-2
CLE@BAL	1	2	-1
LAA@BOS	2	3	-1
MIN@DET	1	1	0
NYM@MIA	1	5	-4
STL@SF	2	1	1
TOR@OAK	4	1	3
WAS@SD	1	1	0

Team	Starter	Projected	Actual	ER err
ARI	Jose Cabrera	2 ER / 14 outs	0 ER / 6 outs	-2
KC	Noah Cameron	1 ER / 18 outs	0 ER / 18 outs	-1

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
KC	Noah Cameron (Starter)	18 / 18	0 / 1.5	0 / 1	3 / 2	0 / 0	2 / 1	7 / 4	venue home: 4.95 vs 3.62 ERA (worse here) · opp run factor 1.05 (average) · 27 prior appearances this season
KC	Craig Kimbrel (Middle Reliever (High Leverage))	3 / 3	36 / 18 (+18.00)	4 / 2	5 / 4	0 / 0	1 / 0	1 / 0
KC	Nate Pearson (Setup Man)	6 / 3	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	3 / 1
KC	Steven Cruz (Closer)	3 / 3	0 / 0	0 / 0	2 / 0	0 / 0	0 / 0	1 / 2

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
KC	Salvador Perez (DH)	5 / 5	3 / 1	1 / 0	0 / 0	0 / 0	3 / 1	0 / 0	0 / 0	1 / 1	1.25	0.309 average	no signal
ARI	Corbin Carroll (RF)	2 / 5	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	2 / 1	-1.11	0.344 elite	bad day
KC	Bobby Witt Jr. (SS)	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	1 / 1	-1.11	0.385 elite	bad day
ARI	Nolan Arenado (3B)	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	1 / 0	1 / 1	0.8	0.283 weak	good day
ARI	Gabriel Moreno (C)	5 / 5	2 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	0 / 1	-0.79	0.365 elite	bad day
ARI	Jordan Lawlar (CF)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.309 average	—
ARI	Tim Tawa (1B)	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0.47	0.303 below average	—
KC	Vinnie Pasquantino (1B)	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 0	0.47	0.328 above average	—
KC	Nick Loftin (3B)	5 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.297 below average	—
KC	Michael Massey (2B)	5 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	-0.47	0.306 average	—
ARI	Ketel Marte (DH)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 0	1 / 1	0.33	0.349 elite	—
KC	Isaac Collins (LF)	1 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	1 / 1	-0.33	0.296 below average	—
KC	Carter Jensen (C)	4 / 4	1 / 1	1 / 0	0 / 0	0 / 1	1 / 1	1 / 1	1 / 0	1 / 1	-0.29	0.304 below average	—
KC	Kyle Isbel (CF)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	1 / 1	-0.14	0.262 weak	—
ARI	Geraldo Perdomo (SS)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	1 / 1	1 / 0	0	0.328 above average	—

Signal	Home-away edge	Meaning
Season quality	-0.403	positive favors home; negative favors away
Recent form	0.204	positive favors home; negative favors away
Starting-pitcher matchup	-0.061	positive favors home; negative favors away
Composite	0.056	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.38-5.08	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-5	0.58	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	3-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-5	n/a	Capped at max 12 realistic runs
simulation_result	3-5	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	4.38-5.08	4-5	0.303	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	3-5	1	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	3-5	3-5	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-5	3-5	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-5	3-5	0	Capped at max 12 realistic runs
simulation_result	3-5	3-5	0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
KC	Nick Loftin	batter / 3B	better	predicted_H 2.06
KC	Jac Caglianone	batter / RF	worse	predicted_SO 1.992
ARI	Ildemaro Vargas	batter / 2B	worse	predicted_SO 1.622
KC	Bobby Witt Jr.	batter / SS	worse	predicted_AB -1.42
KC	Carter Jensen	batter / C	worse	predicted_AB -1.38
KC	Kyle Isbel	batter / CF	worse	predicted_H 1.32
ARI	Nolan Arenado	batter / 3B	worse	predicted_AB -1.27
ARI	Jordan Lawlar	batter / CF	worse	predicted_SO -1.038

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	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	ARI model-prediction batting totals differ from final simulation box score (AB: current=36, sim=33; H: current=8, sim=7; 2B: current=1, sim=2; HR: current=1, sim=0; RBI: current=4, sim=3; R: current=4, sim=3; BB: current=3, sim=1; 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: Ketel Marte: H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Gabriel Moreno: AB current=5 box=4, H current=1 box=2, 2B current=0 box=1, HR current=1 box=0; Ildemaro Vargas: RBI current=1 box=0, SO current=0 box=2; Nolan Arenado: AB current=4 box=3, H current=1 box=0, SO current=1 box=2, SF current=0 box=1; Corbin Carroll: AB current=5 box=4, H current=1 box=2, R current=1 box=0, BB current=1 box=0; Geraldo Perdomo: AB current=4 box=3, H current=1 box=0; Lourdes Gurriel Jr.: RBI current=0 box=1, R current=0 box=1; Jordan Lawlar: SO current=1 box=0	ARI
info	CURRENT_FINALIZED_MATCH_BOX	KC model-prediction batting totals differ from final simulation box score (AB: current=35, sim=33; H: current=8, sim=9; 2B: current=2, sim=3; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=2; HBP: current=0, sim=1).	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=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Jac Caglianone: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Salvador Perez: AB current=5 box=4, 2B current=0 box=1, R current=0 box=1, SO current=1 box=0; Bobby Witt Jr.: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, R current=1 box=0; Vinnie Pasquantino: 2B current=0 box=1, BB current=1 box=0, SO current=0 box=1; Nick Loftin: H current=1 box=3, HR current=0 box=1, RBI current=0 box=1, R current=0 box=1; Isaac Collins: AB current=3 box=4, H current=0 box=1, 2B current=0 box=1, BB current=1 box=0; Michael Massey: RBI current=0 box=1, R current=0 box=1	KC
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
ATL	Grant Holmes	2 ER / 14 outs	0 ER / 18 outs	-2
PHI	Jesús Luzardo	2 ER / 19 outs	0 ER / 27 outs	-2

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
ATL	Grant Holmes (Starter)	18 / 14	0 / 3.86 (-3.86)	0 / 2	3 / 2	0 / 0	2 / 1	6 / 6	venue away: 2.83 vs 4.8 ERA (ok) · opp run factor 0.98 (average) · 28 prior appearances this season
PHI	Jesús Luzardo (Starter)	27 / 19	0 / 2.84	0 / 2	2 / 5	0 / 1	1 / 0	12 / 5	venue home: 4.52 vs 1.76 ERA (worse here) · opp run factor 1.09 (strong) · 28 prior appearances this season

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
ATL	Michael Harris II (CF)	3 / 5	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-0.93	0.342 elite	bad day
ATL	Lane Thomas (RF)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	-0.8	0.314 average	no signal
ATL	Drake Baldwin (C)	4 / 6	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.8	0.365 elite	bad day
PHI	Bryson Stott (3B)	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	1 / 1	-0.8	0.327 above average	bad day
ATL	Ozzie Albies (2B)	4 / 5	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	-0.78	0.284 below average	projection too high
ATL	Ronald Acuña Jr. (DH)	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	-0.47	0.364 elite	—
ATL	Ha-Seong Kim (SS)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.216 weak	—
ATL	Austin Riley (3B)	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.47	0.3 below average	—
ATL	Mauricio Dubón (LF)	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	-0.47	0.302 below average	—
PHI	Bryce Harper (RF)	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	-0.47	0.392 elite	—
PHI	Luis Arraez (2B)	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.306 average	—
PHI	J.T. Realmuto (C)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	3 / 1	-0.47	0.322 above average	—
PHI	Alec Bohm (1B)	2 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	-0.45	0.322 above average	—
ATL	Matt Olson (1B)	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	-0.31	0.353 elite	—
PHI	Kyle Schwarber (DH)	4 / 4	2 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	1 / 1	0.14	0.367 elite	—
PHI	Brandon Marsh (LF)	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.306 average	—
PHI	Trea Turner (SS)	4 / 5	1 / 1	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0	0.293 below average	—
PHI	Justin Crawford (CF)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0	0.282 weak	—

Signal	Home-away edge	Meaning
Season quality	-0.39	positive favors home; negative favors away
Recent form	0.03	positive favors home; negative favors away
Starting-pitcher matchup	-0.113	positive favors home; negative favors away
Composite	-0.275	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	3.61-5.64	n/a	Base ML historical model prediction
variance_applied	4-6	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-4	0.53	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	3-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	n/a	Capped at max 12 realistic runs
simulation_result	3-4	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	3.61-5.64	4-6	-0.033	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-6	3-4	-1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	3-4	3-4	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	3-4	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	3-4	0	Capped at max 12 realistic runs
simulation_result	3-4	3-4	0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
ATL	Lane Thomas	batter / RF	worse	predicted_RBI 1.642
ATL	Drake Baldwin	batter / C	worse	predicted_AB -1.63
PHI	Trea Turner	batter / SS	worse	predicted_RBI 1.61
PHI	Bryson Stott	batter / 3B	worse	predicted_AB -1.481
PHI	Alec Bohm	batter / 1B	better	predicted_RBI 1.41
PHI	Luis Arraez	batter / 2B	worse	predicted_AB -1.41
PHI	J.T. Realmuto	batter / C	worse	predicted_SO 1.202
ATL	Matt Olson	batter / 1B	worse	predicted_SO -1.113

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	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	ATL model pitcher projections differ from final simulation pitching box score (HR: current=0, sim=1).	ATL
info	CURRENT_FINALIZED_MATCH_BOX	ATL model-prediction batting totals differ from final simulation box score (AB: current=38, sim=35; H: current=9, sim=8; RBI: current=4, sim=3; R: current=5, sim=3; BB: current=3, sim=0; SO: current=8, sim=5).	ATL
info	CURRENT_PLAYER_MATCH_BOX	ATL current finalized hitter lines are model predictions and differ from the simulation box score: Ronald Acuña Jr.: RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Lane Thomas: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Drake Baldwin: AB current=6 box=4, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Ozzie Albies: AB current=5 box=4, H current=1 box=2, 2B current=1 box=0, SO current=1 box=0; Matt Olson: H current=1 box=0, 2B current=1 box=0, R current=1 box=0, SO current=1 box=0; Michael Harris II: AB current=5 box=4, 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0; Ha-Seong Kim: SO current=1 box=0; Austin Riley: RBI current=1 box=0	ATL
info	CURRENT_FINALIZED_MATCH_BOX	PHI model-prediction batting totals differ from final simulation box score (AB: current=35, sim=29; H: current=8, sim=5; BB: current=3, sim=1; SO: current=8, sim=7).	PHI
info	CURRENT_PLAYER_MATCH_BOX	PHI current finalized hitter lines are model predictions and differ from the simulation box score: Bryce Harper: RBI current=1 box=0, BB current=1 box=0; Alec Bohm: AB current=4 box=3, RBI current=1 box=2, R current=0 box=1, SO current=1 box=0; Bryson Stott: AB current=4 box=2, H current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Brandon Marsh: AB current=4 box=3, H current=1 box=0; Trea Turner: AB current=5 box=4, 2B current=1 box=0, HR current=0 box=1, RBI current=0 box=2; Luis Arraez: AB current=4 box=3, 2B current=0 box=1, SO current=1 box=0; Kyle Schwarber: HR current=1 box=0, RBI current=1 box=0, BB current=1 box=0; J.T. Realmuto: H current=1 box=0, R current=1 box=0, SO current=1 box=2	PHI
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
CHC	Matthew Boyd	4 ER / 17 outs	3 ER / 21 outs	-1
MIL	Robert Gasser	4 ER / 15 outs	1 ER / 14 outs	-3

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
CHC	Matthew Boyd (Starter)	21 / 17	3.86 / 6.35	3 / 4	5 / 7	0 / 1	2 / 2	7 / 4	venue away: 2.83 vs 5.4 ERA (ok) · opp run factor 1.1 (strong) · 18 prior appearances this season
CHC	Jacob Webb (Closer)	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1
CHC	Ryan Zeferjahn (Middle Reliever (High Leverage))	1 / 1	27 / 81 (-54.00)	1 / 3	1 / 5	0 / 1	1 / 1	1 / 1
MIL	Robert Gasser (Starter)	14 / 15	1.93 / 7.2 (-5.27)	1 / 4	3 / 5	1 / 1	2 / 0	5 / 5	venue home: 4.05 vs 5.44 ERA (ok) · opp run factor 1.03 (strong) · 17 prior appearances this season
MIL	Aaron Ashby (Middle Reliever (High Leverage))	3 / 3	9 / 0 (+9.00)	1 / 0	2 / 1	0 / 0	1 / 0	0 / 1
MIL	Abner Uribe (Setup Man)	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 2	0 / 0	0 / 1	0 / 0
MIL	Trevor Megill (Closer)	3 / 6	9 / 0 (+9.00)	1 / 0	1 / 0	1 / 0	0 / 0	0 / 0

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
MIL	William Contreras (C)	4 / 5	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	2 / 1	-1.4	0.333 above average	bad day
CHC	Pedro Ramirez (2B)	4 / 3	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	0 / 1	0.93	0.289 below average	good day
MIL	Jackson Chourio (LF)	4 / 5	1 / 1	0 / 1	0 / 0	0 / 0	2 / 1	1 / 1	0 / 1	0 / 1	-0.64	0.333 above average	bad day
CHC	Nico Hoerner (SS)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	-0.47	0.324 above average	—
CHC	Alex Bregman (3B)	3 / 4	2 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	1 / 1	0 / 1	0.47	0.318 average	—
CHC	BJ Murray Jr. (1B)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.099 weak	—
CHC	Carson Kelly (C)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.47	0.31 average	—
MIL	Christian Yelich (DH)	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	2 / 1	-0.47	0.317 average	—
CHC	Seiya Suzuki (DH)	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	1 / 1	-0.45	0.339 elite	—
MIL	Brice Turang (2B)	2 / 5	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	1 / 1	-0.45	0.336 above average	—
MIL	Andrew Vaughn (1B)	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 0	-0.33	0.337 above average	—
MIL	Garrett Mitchell (CF)	2 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	0 / 1	0.33	0.35 elite	—
MIL	Joey Ortiz (3B)	3 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0.31	0.295 below average	—
CHC	Pete Crow-Armstrong (CF)	4 / 5	2 / 1	0 / 0	0 / 0	1 / 1	1 / 1	2 / 1	0 / 1	0 / 1	0.14	0.371 elite	—
CHC	Tyrone Taylor (LF)	2 / 3	1 / 1	1 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 0	0	0.318 average	—
CHC	Matt Shaw (RF)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0	0.307 average	—
MIL	Luis Lara (RF)	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	1 / 1	0	0.313 average	—
MIL	Cooper Pratt (SS)	2 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	0	0.313 average	—

Signal	Home-away edge	Meaning
Season quality	0.113	positive favors home; negative favors away
Recent form	-0.346	positive favors home; negative favors away
Starting-pitcher matchup	0.079	positive favors home; negative favors away
Composite	0.508	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.48-4.53	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-6	0.507	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	4.48-4.53	4-5	0.955	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	5-6	0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by 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
CHC	Tyrone Taylor	batter / LF	better	predicted_RBI 2.37
MIL	Garrett Mitchell	batter / CF	better	predicted_H 2.05
MIL	Jackson Chourio	batter / LF	better	predicted_H 1.91
CHC	Matt Shaw	batter / RF	worse	predicted_H 1.8
CHC	BJ Murray Jr.	batter / 1B	better	predicted_AB 1.5
MIL	Brice Turang	batter / 2B	worse	predicted_H -1.4
CHC	Pedro Ramirez	batter / 2B	worse	predicted_SO 1.393
CHC	Carson Kelly	batter / C	worse	predicted_SO 1.36

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	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_FINALIZED_MATCH_BOX	CHC model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; 2B: current=2, sim=3; BB: current=3, sim=1; SO: current=7, sim=6).	CHC
info	CURRENT_PLAYER_MATCH_BOX	CHC current finalized hitter lines are model predictions and differ from the simulation box score: Seiya Suzuki: 2B current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Nico Hoerner: H current=1 box=0; Alex Bregman: AB current=4 box=3, SO current=1 box=0; Tyrone Taylor: AB current=3 box=4, H current=1 box=3, HR current=0 box=1, RBI current=1 box=3; BJ Murray Jr.: AB current=3 box=4, H current=1 box=0; Matt Shaw: AB current=3 box=4, H current=0 box=2, 2B current=0 box=1, SO current=1 box=0; Pete Crow-Armstrong: AB current=5 box=4, 2B current=0 box=1, HR current=1 box=0, BB current=1 box=0; Carson Kelly: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=2	CHC
info	CURRENT_FINALIZED_MATCH_BOX	MIL model-prediction batting totals differ from final simulation box score (AB: current=34, sim=36; H: current=9, sim=12; 2B: current=2, sim=3; 3B: current=0, sim=1; HR: current=1, sim=2; RBI: current=5, sim=7; R: current=5, sim=7; 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: H current=1 box=2, HR current=1 box=0, SO current=1 box=0; Brice Turang: AB current=5 box=4, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Luis Lara: AB current=3 box=4, 2B current=0 box=1, RBI current=1 box=0, R current=0 box=1; Jackson Chourio: H current=1 box=3, BB current=1 box=0, SO current=1 box=0; Andrew Vaughn: 2B current=0 box=1, RBI current=0 box=1, R current=0 box=1, SO current=0 box=1; Joey Ortiz: AB current=3 box=4, RBI current=0 box=1, R current=0 box=1; Cooper Pratt: AB current=3 box=4, HR current=0 box=1, RBI current=1 box=2, R current=0 box=1; Christian Yelich: AB current=4 box=3, H current=1 box=0, R current=1 box=0, BB current=0 box=1	MIL
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
CIN	Chase Burns	3 ER / 17 outs	0 ER / 9 outs	-3
LAD	Emmet Sheehan	2 ER / 15 outs	1 ER / 17 outs	-1

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
CIN	Chase Burns (Starter)	9 / 17	0 / 4.76 (-4.76)	0 / 3	0 / 5	0 / 0	1 / 3	5 / 8	venue away: 3.5 vs 2.09 ERA (worse here) · opp run factor 1.03 (strong) · 26 prior appearances this season

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
LAD	Teoscar Hernández (LF)	3 / 4	2 / 1	0 / 1	0 / 0	1 / 0	3 / 1	2 / 0	1 / 0	1 / 1	1.42	0.325 above average	projection too low
LAD	Shohei Ohtani (DH)	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	2 / 1	-1.4	0.401 elite	bad day
CIN	Tyler Stephenson (C)	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	1 / 1	-1.11	0.335 above average	bad day
CIN	Eugenio Suárez (DH)	4 / 3	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	2 / 1	0.93	0.29 below average	good day
CIN	JJ Bleday (LF)	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	2 / 1	-0.8	0.346 elite	bad day
LAD	Kyle Tucker (RF)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	-0.8	0.326 above average	bad day
LAD	Freddie Freeman (1B)	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	0 / 1	-0.78	0.372 elite	bad day
LAD	Max Muncy (3B)	2 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	2 / 1	2 / 1	0 / 1	0.64	0.349 elite	projection too low
CIN	Ke'Bryan Hayes (3B)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.309 average	—
CIN	Carlos Jorge (CF)	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.47	0.181 weak	—
LAD	Tommy Edman (2B)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.47	0.316 average	—
LAD	Alex Freeland (SS)	1 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.287 below average	—
CIN	Héctor Rodríguez (RF)	4 / 4	2 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 0	1 / 1	-0.46	0.277 weak	—
CIN	Elly De La Cruz (SS)	4 / 5	2 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	2 / 2	0.14	0.368 elite	—
CIN	Sal Stewart (1B)	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 0	2 / 1	-0.14	0.351 elite	—
LAD	Alex Call (CF)	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	-0.14	0.279 weak	—
CIN	Matt McLain (2B)	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	0	0.307 average	—
LAD	Hunter Feduccia (C)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	3 / 1	0	0.32 average	—

Signal	Home-away edge	Meaning
Season quality	1.246	positive favors home; negative favors away
Recent form	-0.108	positive favors home; negative favors away
Starting-pitcher matchup	0.076	positive favors home; negative favors away
Composite	1.006	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.24-3.66	n/a	Base ML historical model prediction
variance_applied	4-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-6	0.533	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, starter_quality_favours_the_other_side, projected_margin_is_one_run, season_quality_guardrail)
embedded_savant_run_environment	5-6	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-6	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-6	n/a	Capped at max 12 realistic runs
simulation_result	5-6	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	4.24-3.66	4-4	0.574	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-4	5-6	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, starter_quality_favours_the_other_side, projected_margin_is_one_run, season_quality_guardrail)
embedded_savant_run_environment	5-6	5-6	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-6	5-6	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-6	5-6	0	Capped at max 12 realistic runs
simulation_result	5-6	5-6	0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
CIN	Carlos Jorge	batter / CF	better	predicted_SO 2
LAD	Hunter Feduccia	batter / C	worse	predicted_HR 1.96
CIN	Elly De La Cruz	batter / SS	better	predicted_RBI 1.78
LAD	Tommy Edman	batter / 2B	worse	predicted_BB 1.758
LAD	Alex Call	batter / CF	worse	predicted_AB 1.5
LAD	Max Muncy	batter / 3B	better	predicted_AB 1.48
LAD	Alex Freeland	batter / SS	worse	predicted_H 1.395
LAD	Freddie Freeman	batter / 1B	worse	predicted_SO 1.292

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	CIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CIN
info	CURRENT_PITCHER_DUPLICATION	LAD pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	LAD
info	CURRENT_FINALIZED_MATCH_BOX	CIN model pitcher projections differ from final simulation pitching box score (H: current=8, sim=10; HR: current=2, sim=4).	CIN
info	CURRENT_FINALIZED_MATCH_BOX	CIN model-prediction batting totals differ from final simulation box score (H: current=8, sim=6; 2B: current=1, sim=0; 3B: current=0, sim=1; HR: current=2, sim=3; R: current=4, sim=5; BB: current=3, sim=2; SO: current=10, sim=8).	CIN
info	CURRENT_PLAYER_MATCH_BOX	CIN current finalized hitter lines are model predictions and differ from the simulation box score: Héctor Rodríguez: AB current=4 box=3, H current=1 box=2, BB current=0 box=1, SO current=1 box=0; Elly De La Cruz: AB current=5 box=4, H current=1 box=2, 3B current=0 box=1, RBI current=1 box=3; JJ Bleday: H current=1 box=0, RBI current=1 box=0, BB current=1 box=0, SO current=1 box=0; Tyler Stephenson: 2B current=1 box=0, RBI current=1 box=0, R current=0 box=1, BB current=1 box=0; Sal Stewart: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Eugenio Suárez: AB current=3 box=4, H current=1 box=0; Matt McLain: AB current=3 box=4, R current=1 box=0, SO current=1 box=2; Ke'Bryan Hayes: H current=1 box=0, SO current=1 box=0	CIN
info	CURRENT_FINALIZED_MATCH_BOX	LAD model-prediction batting totals differ from final simulation box score (AB: current=33, sim=36; H: current=8, sim=10; 2B: current=2, sim=1; 3B: current=0, sim=1; HR: current=1, sim=4; RBI: current=4, sim=6; R: current=5, sim=6; SO: current=8, sim=10).	LAD
info	CURRENT_PLAYER_MATCH_BOX	LAD current finalized hitter lines are model predictions and differ from the simulation box score: Shohei Ohtani: 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0; Tommy Edman: AB current=4 box=3, 3B current=0 box=1, RBI current=1 box=2, BB current=0 box=2; Freddie Freeman: 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=2; Teoscar Hernández: 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Alex Call: AB current=3 box=4, H current=1 box=0, SO current=0 box=1; Alex Freeland: AB current=3 box=4, H current=1 box=2, R current=0 box=1; Max Muncy: AB current=4 box=5, HR current=0 box=1, RBI current=0 box=1, BB current=1 box=0; Kyle Tucker: HR current=0 box=1, RBI current=0 box=1, BB current=1 box=0, SO current=1 box=2	LAD
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
CLE	Joey Cantillo	2 ER / 14 outs	4 ER / 7 outs	2
BAL	Trevor Rogers	3 ER / 17 outs	1 ER / 22 outs	-2

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
CLE	Joey Cantillo (Starter)	7 / 14	15.43 / 3.86 (+11.57)	4 / 2	3 / 2	1 / 1	5 / 0	3 / 6	venue away: 3.81 vs 3.82 ERA (ok) · opp run factor 0.99 (average) · 29 prior appearances this season
CLE	Daniel Espino (Reliever)	3 / 0	0 / n/a	0 / 2	2 / 1	0 / 1	0 / 0	2 / 1
BAL	Trevor Rogers (Starter)	22 / 17	1.23 / 4.76 (-3.53)	1 / 3	6 / 6	0 / 1	0 / 3	6 / 3	venue home: 4.05 vs 4.34 ERA (ok) · opp run factor 1.01 (weak) · 26 prior appearances this season
BAL	Andrew Kittredge (Closer)	2 / 3	40.5 / 0 (+40.50)	3 / 0	3 / 0	1 / 0	0 / 0	0 / 1
BAL	Josh Walker (Middle Reliever (High Leverage))	1 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 2
BAL	Rico Garcia (Setup Man)	2 / 4	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
BAL	Coby Mayo (3B)	5 / 3	2 / 1	0 / 0	0 / 0	2 / 0	3 / 1	2 / 1	0 / 0	2 / 1	2.33	0.317 average	no signal
BAL	Rece Hinds (CF)	2 / 3	2 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	0 / 2	1.6	0.158 weak	good day
BAL	Pete Alonso (1B)	2 / 4	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	2 / 1	0 / 1	-1.07	0.392 elite	bad day
CLE	Brayan Rocchio (SS)	3 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 0	0 / 1	-0.93	0.291 below average	projection too high
CLE	David Fry (1B)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	-0.8	0.259 weak	projection too high
CLE	Jo Adell (DH)	4 / 4	2 / 1	1 / 0	0 / 0	0 / 0	1 / 1	1 / 0	0 / 0	1 / 1	0.78	0.324 above average	projection too low
BAL	Dylan Beavers (LF)	4 / 3	1 / 0	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	3 / 1	0.78	0.303 below average	good day
BAL	Gunnar Henderson (SS)	5 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	-0.64	0.315 average	no signal
CLE	Angel Genao (2B)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	-0.47	0.29 below average	—
BAL	Christian Encarnacion-Strand (DH)	4 / 3	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	1 / 1	-0.45	0.345 elite	—
CLE	José Ramírez (3B)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 0	-0.33	0.335 above average	—
CLE	Angel Martínez (RF)	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.31	0.303 below average	—
BAL	Blaze Alexander (RF)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	2 / 0	2 / 1	0.19	0.352 elite	—
CLE	Steven Kwan (LF)	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0.14	0.322 above average	—
BAL	Samuel Basallo (C)	3 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	2 / 0	0 / 1	1 / 1	0.14	0.322 above average	—
CLE	Petey Halpin (CF)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.283 weak	—
CLE	Austin Hedges (C)	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.275 weak	—
BAL	Jackson Holliday (2B)	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	1 / 1	0	0.303 below average	—

Signal	Home-away edge	Meaning
Season quality	-0.116	positive favors home; negative favors away
Recent form	-0.235	positive favors home; negative favors away
Starting-pitcher matchup	-0.025	positive favors home; negative favors away
Composite	0.004	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.21-4.24	n/a	Base ML historical model prediction
variance_applied	4-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-4	0.508	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	3-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	n/a	Capped at max 12 realistic runs
simulation_result	3-4	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	4.21-4.24	4-4	-0.032	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-4	3-4	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	3-4	3-4	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	3-4	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	3-4	0	Capped at max 12 realistic runs
simulation_result	3-4	3-4	0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
CLE	Daniel Espino	pitcher / Reliever	worse	predicted_ER 2
BAL	Trevor Rogers	pitcher / Starter	neutral	predicted_SO 2
BAL	Pete Alonso	batter / 1B	boosted	predicted_SO 1.929
BAL	Blaze Alexander	batter / RF	worse	predicted_AB 1.835
BAL	Dylan Beavers	batter / LF	worse	predicted_H 1.77
BAL	Jackson Holliday	batter / 2B	better	predicted_RBI 1.68
CLE	Austin Hedges	batter / C	boosted	predicted_RBI 1.66
CLE	David Fry	batter / 1B	worse	predicted_AB 1.5

STL@SFMISS

Predicted: STL 7-5 (conf -0.25, strong)Actual: SF 4-5Brier 0.393total bias +3

Root cause: Backed the wrong side with real conviction

Evidence-backed diagnosis

published system edge 0.25 (strong); this is not the half-scale distance from 50%
team-run absolute error 3; total bias +3
pregame home-minus-away evidence: season -0.516, recent -0.372, pitcher -0.103, composite -0.963
confidence pipeline reached 0.627 before later publication/recency controls; published home probability was 0.373
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 3 (12 vs 9)
STL offense over-projected by 3
Logan Webb (SF) allowed 3 fewer ER than projected (1 vs 4)
committed to STL at 0.25 confidence and it lost

Team	Starter	Projected	Actual	ER err
STL	Michael McGreevy	2 ER / 16 outs	2 ER / 20 outs	0
SF	Logan Webb	4 ER / 16 outs	1 ER / 18 outs	-3

Decision internals: run factors home×0.975 / away×1.078; 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:

Shay Whitcomb (SF 3B): 2-for-5, 0 HR, 1 RBI (projected 0.0 H, 0.0 HR) [run impact 0.94]
Joshua Baez (STL LF): 1-for-5, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Turner Hill (SF LF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]
Alec Burleson (STL 1B): 1-for-5, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.62]

Starter miss: Logan Webb (SF) posted 1.50 ERA vs projected 6.75 (1 ER in 18 outs); he faced a strong offense (opponent run factor 1.08, bucket average).

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

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

Michael McGreevy (STL) threw 20 outs, 2 ER = 2.7 ERA against a projected 3.38 ERA (-0.68 ERA error)

Bullpen behind him: 0 ER over 7 outs.

The HOME lineup he faced produced 4 runs on 10 hits and 0 HR, against a projected 4 runs / 8 hits (0 runs vs projection).

Who did the damage to him:

Shay Whitcomb: 2-for-5, 0 HR, 1 RBI [+0.94 runs vs projection] season xwOBA 0.251, weak - a good day, not a model error
Rafael Devers: 2-for-3, 0 HR, 1 RBI [+0.49 runs vs projection] season xwOBA 0.339, elite - a good day, not a model error
Christian Koss: 1-for-3, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.236, weak - a good day, not a model error

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

HOME pitching

Logan Webb (SF) threw 18 outs, 1 ER = 1.5 ERA against a projected 6.75 ERA (-5.25 ERA error)

Bullpen behind him: 1 ER over 14 outs.

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

Who did the damage to him:

Alec Burleson: 1-for-5, 1 HR, 1 RBI [+0.62 runs vs projection] season xwOBA 0.373, elite - real projection error
Bryan Torres: 1-for-4, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.298, below average - a good day, not a model error
Nathan Church: 1-for-5, 0 HR, 1 RBI [+0.31 runs vs projection] season xwOBA 0.282, weak - a good day, not a model error

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

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)

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
STL	Michael McGreevy (Starter)	20 / 16	2.7 / 3.38	2 / 2	6 / 3	0 / 2	2 / 2	4 / 1	venue away: 3.89 vs 4.06 ERA (ok) · opp run factor 0.98 (weak) · 27 prior appearances this season
STL	George Soriano (Setup Man)	1 / 3	0 / 18 (-18.00)	0 / 2	1 / 1	0 / 0	0 / 2	0 / 1
STL	Riley O'Brien (Closer)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	2 / 1	0 / 2
STL	Ryne Stanek (Middle Reliever (High Leverage))	3 / 5	0 / 5.4 (-5.40)	0 / 1	2 / 2	0 / 0	1 / 1	1 / 1
SF	Logan Webb (Starter)	18 / 16	1.5 / 6.75 (-5.25)	1 / 4	3 / 6	1 / 2	2 / 2	8 / 5	venue home: 3.39 vs 5.29 ERA (ok) · opp run factor 1.08 (average) · 25 prior appearances this season
SF	Trent Harris (Setup Man)	3 / 4	0 / 0	0 / 0	1 / 1	0 / 0	1 / 0	2 / 2
SF	Dylan Smith (Closer)	5 / 3	5.4 / 0 (+5.40)	1 / 0	4 / 0	0 / 0	1 / 0	0 / 0
SF	Jason Foley (Middle Reliever (High Leverage))	6 / 1	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0 / 0

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

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
SF	Shay Whitcomb (3B)	5 / 3	2 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0.94	0.251 weak	good day
STL	Joshua Baez (LF)	5 / 3	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 0	2 / 1	-0.93	0.337 above average	bad day
SF	Turner Hill (LF)	4 / 3	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	2 / 0	1 / 1	0 / 1	-0.78	0.346 elite	bad day
STL	Alec Burleson (1B)	5 / 4	1 / 1	0 / 1	0 / 0	1 / 0	1 / 1	2 / 0	0 / 0	2 / 1	0.62	0.373 elite	projection too low
SF	Rafael Devers (DH)	3 / 4	2 / 1	2 / 0	0 / 0	0 / 1	1 / 1	1 / 1	2 / 1	0 / 1	0.49	0.339 elite	—
SF	Andrew Knizner (C)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.271 weak	—
STL	Bryan Torres (2B)	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0.33	0.298 below average	—
STL	Leonardo Bernal (DH)	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	1 / 1	-0.33	0.365 elite	—
SF	Christian Koss (SS)	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 0	0 / 1	0.33	0.236 weak	—
SF	Jung Hoo Lee (RF)	4 / 5	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0 / 1	0.33	0.316 average	—
STL	Nathan Church (CF)	5 / 3	1 / 1	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 0	0.31	0.282 weak	—
STL	Thomas Saggese (SS)	5 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0.14	0.313 average	—
STL	Nolan Gorman (3B)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	1 / 1	-0.14	0.276 weak	—
SF	Bryce Eldridge (1B)	4 / 4	2 / 1	1 / 1	0 / 0	0 / 0	3 / 1	0 / 1	0 / 1	1 / 1	0.14	0.356 elite	—
STL	Jordan Walker (RF)	4 / 5	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	2 / 1	0.02	0.347 elite	—
STL	Iván Herrera (C)	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	0	0.332 above average	—
SF	Osleivis Basabe (2B)	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.273 weak	—
SF	Jonah Cox (CF)	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	0	0.273 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.516	positive favors home; negative favors away
Recent form	-0.372	positive favors home; negative favors away
Starting-pitcher matchup	-0.103	positive favors home; negative favors away
Composite	-0.963	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.26-4.78	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	7-5	0.627	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	7-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	7-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	7-5	n/a	Capped at max 12 realistic runs
simulation_result	7-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.26-4.78	4-5	-0.485	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	7-5	3	Enhanced ML with recent form, H2H, low-score analysis
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

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
STL	Bryan Torres	batter / 2B	better	predicted_AB 2.5
STL	Leonardo Bernal	batter / DH	better	predicted_RBI 2
SF	Turner Hill	batter / LF	boosted	predicted_AB 1.941
STL	Iván Herrera	batter / C	better	predicted_BB 1.543
SF	Rafael Devers	batter / DH	better	predicted_BB 1.538
SF	Andrew Knizner	batter / C	worse	predicted_AB 1.5
SF	Jonah Cox	batter / CF	worse	predicted_AB 1.5
SF	Jung Hoo Lee	batter / RF	worse	predicted_RBI 1.39

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	SF pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	SF
info	CURRENT_PITCHER_DUPLICATION	STL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	STL
info	CURRENT_FINALIZED_MATCH_BOX	SF model pitcher projections differ from final simulation pitching box score (BB: current=2, sim=3).	SF
info	CURRENT_FINALIZED_MATCH_BOX	SF model-prediction batting totals differ from final simulation box score (AB: current=31, sim=33; H: current=8, sim=6; 3B: current=0, sim=1; HR: current=1, sim=2; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=6; SO: current=9, sim=5).	SF
info	CURRENT_PLAYER_MATCH_BOX	SF current finalized hitter lines are model predictions and differ from the simulation box score: Turner Hill: AB current=3 box=5, H current=1 box=2, 2B current=1 box=2, R current=0 box=1; Rafael Devers: AB current=4 box=3, BB current=1 box=2, SO current=1 box=0; Bryce Eldridge: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Osleivis Basabe: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=1, R current=0 box=1; Andrew Knizner: AB current=3 box=4, H current=1 box=0; Christian Koss: BB current=0 box=1, SO current=1 box=0; Jung Hoo Lee: AB current=5 box=4, 3B current=0 box=1, RBI current=1 box=2, R current=0 box=1; Jonah Cox: AB current=3 box=4, H current=1 box=0, R current=1 box=0, BB current=0 box=1	SF
info	CURRENT_FINALIZED_MATCH_BOX	STL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=37; H: current=9, sim=10; 2B: current=2, sim=1; HR: current=1, sim=3; RBI: current=5, sim=7; R: current=5, sim=7; HBP: current=0, sim=1).	STL
info	CURRENT_PLAYER_MATCH_BOX	STL current finalized hitter lines are model predictions and differ from the simulation box score: Iván Herrera: AB current=4 box=3, RBI current=1 box=0, BB current=1 box=2; Bryan Torres: AB current=3 box=5, HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Leonardo Bernal: AB current=4 box=5, HR current=0 box=1, RBI current=1 box=3, BB current=1 box=0; Joshua Baez: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Thomas Saggese: AB current=3 box=4, 2B current=0 box=1, R current=1 box=0, BB current=1 box=0; Nolan Gorman: RBI current=1 box=0, R current=0 box=1, SO current=1 box=0; Alec Burleson: H current=1 box=2, 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=2; Nathan Church: AB current=3 box=4, H current=1 box=2, R current=0 box=1, SO current=0 box=1	STL
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Cross-game lever tension

Each candidate fix, and exactly which games it would help versus hurt. A lever that repairs one game and breaks two is visible here before it is proposed.

Day over day: did it improve or worsen

Compared against packet 20260906. Last recorded change: v4.26.0 (20260906) The nightly self-improvement cycle runs at 05:00 America/New_York instead of 02:00.

The adjustment the system made in response

Luna deep post-mortem (full)

1. Evidence receipt and gaps

Sources received

All seven source_manifest entries were checked against the supplied evidence:

Source	Status	Purpose
`prediction_review_20260907.json`	Loaded	Official outcomes and scoring errors
`PREDICTION_SYSTEM_DEEP_AUDIT_20260907.json`	Loaded	Probability, provenance, starter, and decision contracts
`continuous_learning_daily_20260907.json`	Loaded	Pipeline-stage quality, calibration, probability lineage, score distribution, and starter comparisons
`detailed_20260907.html`	Loaded	Full game and simulation rendering
`20260907_ultra_detailed_player_audit.html`	Loaded	Player-level prediction trace
`audit_report_20260907.html`	Loaded	Invariant findings
`audit_report_20260907_integrity_revision.html`	Missing	Post-Luna remediation proof for immutable revision artifacts

The missing integrity-revision audit is the only explicitly missing source. It prevents independent confirmation of immutable revision-artifact remediation, although the supplied structural and per-game audits report no critical or error findings.

Slate coverage

All 11 expected games were reviewed:

ARI@KC, ATL@PHI, CHC@MIL, CIN@LAD, CLE@BAL, LAA@BOS, MIN@DET, NYM@MIA, STL@SF, TOR@OAK, and WAS@SD.

Coverage is internally complete for the principal slate artifacts:

11 expected games
11 canonical artifacts loaded
11 continuous-learning games
11 forensic deep-audit games
11 structural audits
11 starter comparisons
11 games with structural QA

This agrees with coverage_matrix.all_games_structurally_audited=true.

Gaps and contradictions

Missing integrity revision audit.

integrity revision audit is marked missing; no bytes are supplied.

Four unexplained probability-publication deltas.

The causal register identifies ARI@KC, CIN@LAD, LAA@BOS, and NYM@MIA as having unexplained calibrated-to-published deltas of at least two percentage points. This is a critical evidence/contract failure until traced.

Probability-lineage representation is not uniformly intuitive.

Some records show a calibrated probability above 0.50 while the published pick probability is below 0.50 because the selected side changes from the home side to the away side. That is not necessarily a defect, but the lineage must clearly identify whether each probability is:

probability of the home team,
probability of the selected side, or
probability after side inversion.

ARI starter identity inconsistency.

The game-level starter line lists away starter Jose Cabrera, with identity_match=false, actual 0.00 ERA over 6 outs, and no settled pitching line. The player-forensics block instead identifies Noah Cameron as the relevant Kansas City starter and explicitly reports the Arizona starter as missing. This is a real data-reconciliation issue, even though that game’s structural audit reports zero findings.

Audit finding names are not fully supplied.

The structural summary reports 0 critical, 0 error, 0 warning, and 81 info, but the individual 81 informational finding texts are not present in the supplied object. They therefore cannot be enumerated by title without inventing evidence.

No market cross-check.

No source-labelled moneyline was available for any game. Consequently, market agreement or disagreement cannot be evaluated.

No actual player results may be inferred beyond supplied lines.

The analysis below uses only the listed player, starter, lineup, and run-deviation records.

---

2. Executive diagnosis

Overall result

The slate finished 6–5, or 54.55% winner accuracy. That is below the supplied walk-forward baseline accuracy of 55.53% and below the historical validation winner accuracy of 59.16%.

The system’s own headline is therefore directionally correct:

Slate winner accuracy: 0.5455
Walk-forward baseline: 0.555307
Historical model validation accuracy: 0.591647
Slate Brier score is not explicitly aggregated, but individual Brier values show substantial dispersion, including 0.489381 for TOR@OAK and 0.392833 for STL@SF
High-conviction games: 4
High-conviction misses: 3
High-conviction miss rate: 75%

The slate should not be described as a good prediction merely because it got six winners correct.

Winner accuracy

Winner performance was weak but not catastrophically below chance:

Correct winners: 6 of 11
Incorrect winners: 5 of 11
Correct-winner/wrong-score games: ATL@PHI and WAS@SD
Toss-up calls that happened to win: CHC@MIL and CLE@BAL
Strong calls that won: LAA@BOS
Strong calls that lost: STL@SF and TOR@OAK

The five misses were:

ARI@KC, MIN@DET, NYM@MIA, STL@SF, and TOR@OAK.

Calibration quality

Calibration was poor on this slate:

Weighted absolute calibration error: 0.2611
The 0.65–0.70 bucket had average confidence 0.6996, zero hits, and observed hit rate 0.0
The 0.55–0.60 bucket had average confidence 0.5796, zero hits, and observed hit rate 0.0
The 0.60–0.65 bucket had average confidence 0.625, one hit in two games
The 0.50–0.55 bucket had five hits in seven games, observed hit rate 0.7143, and was underconfident by the stated definition

The current slate’s confidence was anti-predictive at the upper end: strong convictions lost three of four times. This is material calibration evidence, but the causal register correctly warns not to fit a stable calibration curve from one slate.

The high-raw cluster was specifically:

TOR@OAK
Calibrated probability: 0.6996
Empirical historical hit rate for the relevant cluster: 0.5614
Outcome: miss

That is a warning against treating raw or governed confidence near 0.70 as reliable.

Score accuracy

Score accuracy was materially worse than winner accuracy:

Holdout run-model R²:
Away runs: 0.020536
Home runs: 0.012404
Total runs: 0.018362
Pipeline final team-run MAE: 1.9091
Pipeline final total-run MAE: 3.0909
Pipeline final winner accuracy: 0.5455
Actual margins of at least four runs: 1 of 11
Predicted margins of two runs or less: 10 of 11
Mean final margin bias: -0.364
Margin trend across slates is reported as improving, but the current model still compressed most games toward close outcomes.

The exact-score system is therefore low-confidence. A correct winner frequently coexisted with a poor score forecast.

Starter accuracy

Starter projections were highly volatile:

Five of 20 listed starters exceeded projected ERA by at least 3.00 in the matchup-pattern analysis.
Two starters had fewer than 12 prior appearances in the feed: Jonah Tong with 3 and Nick Pivetta with 4.
Starter variance accounted for many large game-level deviations, but not all winner misses.
The strongest recurring starter-specific evidence was the credible venue split for Chase Burns in CIN@LAD; the supplied diagnosis recommends increasing context.venue_adjustment_weight in shadow validation.
The diagnosis does not support blindly tuning starters on every outlier. It explicitly classifies many of the largest deviations as variance.

Artifact integrity

Artifact integrity was generally good, but not clean in the broadest contractual sense:

Every game’s structural audit reported PASS.
Every game reported zero critical, error, and warning findings.
The aggregate structural summary reported:
Critical: 0
Error: 0
Warning: 0
Informational: 81
However, four probability-lineage records have unexplained material publication deltas.
The integrity-revision audit is missing.
ARI contains a starter identity/missing-line inconsistency.

Therefore:

Presentation/file integrity: generally passed.
Probability contract integrity: failed for four games pending reconciliation.
Starter-data integrity: at least one game, ARI, requires reconciliation.
Model quality: weak run signal and poor slate calibration are independent of artifact cleanliness.

Causal issue register reconciliation

Critical: `unexplained_probability_publication_delta`

Severity: Critical
Evidence: Four games have unexplained calibrated-to-published deltas of at least two percentage points.
Affected games: ARI@KC, CIN@LAD, LAA@BOS, NYM@MIA
Recommended action: Block promotion until the missing probability transformation is recorded and reconciled.
Guardrail: Do not assume the probability governor was bypassed until the pick-side math is traced.

This is a contract failure, not merely a calibration problem.

High: `single_slate_calibration_instability`

Severity: High
Evidence: Weighted absolute calibration error 0.2611 across 11 games
Affected games: all 11
Recommended action: Observe the same buckets over the full rolling window before changing calibration weights.
Guardrail: Do not treat this slate as sufficient to establish a stable calibration curve.

High: `weak_run_model_signal`

Severity: High
Evidence: Home/away/total run R² of 0.0124/0.0205/0.0184
Affected games: all 11
Recommended action: Treat exact scores as low confidence and investigate features/targets.
Guardrail: Do not infer that small positive R² makes exact scores reliable.

Info: `complete_slate_audit_coverage`

Severity: Informational
Evidence: 11/11 structurally audited
Affected games: all 11
Recommended action: Retain the per-game audit receipt.
Guardrail: Full coverage does not cancel the findings within audited games.

Info: `simulation_inert_on_slate`

Severity: Informational on this record, but escalated to a persistent failure in cross-slate evidence
Evidence: 0 flip candidates and 0 published flips
Affected games: all 11
Recommended action: Measure simulation contribution over rolling slates.
Guardrail: Zero winner flips is not itself a defect and does not prove that no score reconciliation occurred.

---

3. Every game

The probability below is the published probability for the selected side unless otherwise noted. A positive confidence edge means the home side was favored; a negative edge means the away side was favored.

`ARI@KC` — miss; wrong winner and wrong score

Predicted: ARI 3, KC 5; winner KC
Actual: ARI 5, KC 4; winner ARI
Published selected-side probability: KC 0.579623
System confidence edge: home-minus-away +0.1592
Bucket: Moderate
Brier: 0.335963
Team-run absolute error: 3
Total-run bias: -1; predicted total 8 versus actual total 9
Primary cause: Wrong conviction side
Secondary cause: None

Edges and controls

Season home-minus-away edge: -0.4026
Recent edge: +0.204
Pitcher edge: -0.0608
Composite edge: +0.0558
Starter-experience shrink applied: yes
KC starter history: 27 career games in feed
Opposing starter record was thin/missing in the settled feed
Score-volatility action: no_change
Published probability was lower than the pre-publication pipeline value: governed 0.618, published 0.579623

Starter projection versus actual

ARI listed starter Jose Cabrera: projected ERA 3.86, projected 2 ER/14 outs; listed actual 0.00 ERA, 0 ER/6 outs; identity mismatch and no settled pitching line
KC Noah Cameron: projected ERA 1.5, projected 1 ER/18 outs; actual 0.00 ERA, 0 ER/18 outs
Forensic block separately records KC starter Noah Cameron with projected 1 ER and actual 0 ER, and identifies the ARI starter line as missing

Score path and player adjustments

Historical ML: ARI 4.38, KC 5.08
Variance-applied: 4–5
Enhanced: 3–5
Savant/environment: 3–5
Final simulation: 3–5
Important player adjustments included Nick Loftin, Jac Caglianone, Ildemaro Vargas, Bobby Witt Jr., Carter Jensen, Kyle Isbel, Nolan Arenado, and Jordan Lawlar.
Player-level result highlights included Salvador Pérez outperforming projection and Corbin Carroll/Bobby Witt Jr. underperforming it.

Audit and causal assessment

Structural audit: PASS; zero critical/error/warning findings
Forecast artifact: reported clean
The miss is best classified as a moderate-confidence directional error with starter-feed inconsistency and player variance. The supplied forensic verdict says neither side broke materially and calls the result closer to variance.

---

`ATL@PHI` — correct winner, wrong score

Predicted: ATL 3, PHI 4; winner PHI
Actual: ATL 0, PHI 1; winner PHI
Published probability: PHI 0.53
Confidence edge: +0.06
Bucket: Lean
Brier: 0.2209
Team-run absolute error: 6
Total-run bias: +6; predicted 7 versus actual 1
Winner quality: Right winner, wrong score
Primary/secondary causes: None assigned

Edges and controls

Season edge: -0.390
Recent edge: +0.030
Pitcher edge: -0.1125
Composite edge: -0.2745
Recency guardrail: applied
Starter-experience shrink: not applied
Large Savant run-environment swing flag: present
Score-volatility action: no_change

Starters

ATL Grant Holmes: projected 2 ER/14 outs, projected ERA 3.86; actual 0 ER/18 outs, 0.00 ERA
PHI Jesús Luzardo: projected 2 ER/19 outs, projected ERA 2.84; actual 0 ER/27 outs, 0.00 ERA
Both starters materially beat projections.

Score path and player adjustments

Published score: 3–4
The supplied score-path details are less extensive than for several other games, but the actual result was a one-run, one-total-run game.
Major player-level deviations included Michael Harris II, Lane Thomas, Drake Baldwin, Bryson Stott, and Ozzie Albies.
Both offenses scored well below projection.

Audit and causal assessment

Structural audit: PASS
This is a correct-winner/wrong-score game. The score miss was driven by simultaneous low offensive output and unexpectedly dominant starting pitching, not by a wrong side.

---

`CHC@MIL` — correct winner, poor score

Predicted: CHC 5, MIL 7; winner MIL
Actual: CHC 3, MIL 4; winner MIL
Published probability: MIL 0.506818
Confidence edge: +0.0136
Bucket: Toss-up
Brier: 0.243228
Team-run absolute error: 5
Total-run bias: +5; predicted 12 versus actual 7
Winner quality: Toss-up went our way

Edges and controls

Season edge: +0.1133
Recent edge: -0.346
Pitcher edge: +0.0785
Composite edge: +0.5085
Recency guardrail: applied
Starter-experience shrink: not applied
Score-volatility action: no_change

Starters

CHC Matthew Boyd: projected 4 ER/17 outs, ERA 6.35; actual 3 ER/21 outs, ERA 3.86
MIL Robert Gasser: projected 4 ER/15 outs, ERA 7.20; actual 1 ER/14 outs, ERA 1.93
Gasser was the largest starter miss: 3 fewer ER than projected.

Score path and player adjustments

The exact score path is not included in the game-level block, but the final prediction was 5–7.
Player-level contributors included William Contreras, Pedro Ramirez, and Jackson Chourio.
CHC offense scored below projection; MIL was classified as not materially broken on the offensive side.

Audit and causal assessment

Structural audit: PASS
This was a near-coin-flip call. The winner was correct, but the score model overestimated scoring by five runs. Starter variance, particularly Gasser’s unexpected performance, was central.

---

`CIN@LAD` — correct winner, wrong score

Predicted: CIN 5, LAD 6; winner LAD
Actual: CIN 3, LAD 6; winner LAD
Published probability: LAD 0.532561
Confidence edge: +0.0651
Bucket: Lean
Brier: 0.218499
Team-run absolute error: 2
Total-run bias: +2
Winner quality: Solid correct call

Edges and controls

Season edge: +1.2463
Recent edge: -0.108
Pitcher edge: +0.076
Composite edge: +1.0057
Season-quality margin adjustment: +0.5234
Score-volatility action: supported blowout-margin expansion
No starter-experience shrink

Starters

CIN Chase Burns: projected 3 ER/17 outs, ERA 4.76; actual 0 ER/9 outs, 0.00 ERA
LAD Emmet Sheehan: projected 2 ER/15 outs, ERA 3.60; actual 1 ER/17 outs, ERA 1.59
Burns was three ER better than projected; the forensic assessment notes a credible away venue weakness that was underweighted.

Score path and player adjustments

Final score path ended at CIN 5, LAD 6.
Important players: Teoscar Hernández exceeded projection by 1.42 runs; Shohei Ohtani underperformed by 1.40; Tyler Stephenson and other Cincinnati hitters also deviated.
The supplied forensic recommendation is to enable or raise context.venue_adjustment_weight in shadow validation.

Audit and causal assessment

Structural audit: PASS
Correct winner, somewhat inflated Cincinnati score, and a starter projection miss. The venue-split signal is more actionable than the one-game player deviations.

---

`CLE@BAL` — correct winner, wrong score

Predicted: CLE 3, BAL 4; winner BAL
Actual: CLE 4, BAL 6; winner BAL
Published probability: BAL 0.507765
Confidence edge: +0.0155
Bucket: Toss-up
Brier: 0.242295
Team-run absolute error: 3
Total-run bias: -3; predicted 7 versus actual 10
Winner quality: Toss-up went our way

Edges and controls

Season edge: -0.1158
Recent edge: -0.235
Pitcher edge: -0.0245
Composite edge: +0.0037
Recency guardrail: applied
Large Savant run-environment swing flag: present
Score-volatility action: no_change

Starters

CLE Joey Cantillo: projected 2 ER/14 outs, ERA 3.86; actual 4 ER/7 outs, ERA 15.43
BAL Trevor Rogers: projected 3 ER/17 outs, ERA 4.76; actual 1 ER/22 outs, ERA 1.23
Cantillo was the major negative starter miss, while Rogers outperformed projection.

Score path and player adjustments

Player deviations included Coby Mayo’s two-homer performance, Rece Hinds exceeding projection, and below-projection results from Brayan Rocchio and David Fry.
The model underpredicted the ten-run environment by three runs.

Audit and causal assessment

Structural audit: PASS
Correct winner at essentially even odds. Score error was driven by highly variable starter and player outcomes rather than a clear directional model failure.

---

`LAA@BOS` — strong correct winner, close score

Predicted: LAA 3, BOS 5; winner BOS
Actual: LAA 2, BOS 5; winner BOS
Published probability: BOS 0.62315
Confidence edge: +0.2463
Bucket: Strong
Brier: 0.142016
Team-run absolute error: 1
Total-run bias: +1
Winner quality: Strong correct call

Edges and controls

Season edge: +1.0127
Recent edge: +0.658
Pitcher edge: +0.0198
Composite edge: +1.125
Season-quality margin adjustment: +0.4253
Score-volatility action: no_change

Starters

LAA Grayson Rodriguez: projected 3 ER/15 outs, ERA 5.40; actual 5 ER/12 outs, ERA 11.25
BOS Brayan Bello: projected 2 ER/13 outs, ERA 4.15; actual 0 ER/15 outs, ERA 0.00
Both starter outcomes favored Boston, despite Rodriguez being worse than projected and Bello much better.

Score path and player adjustments

Player-level effects included Moisés Ballesteros, Jarren Duran, Mike Trout, Isiah Kiner-Falefa, and Josh Lowe.
The score was close to the actual result despite starter-level deviations.

Audit and causal assessment

Structural audit: PASS
This is evidence against blindly widening every run distribution. The variance lever that could help several high-confidence misses would also soften this strong correct call.

---

`MIN@DET` — miss; wrong winner and wrong score

Predicted: MIN 5, DET 4; winner MIN
Actual: MIN 4, DET 5; winner DET
Published probability: selected away side MIN 0.53; home probability 0.470
Confidence edge: home-minus-away -0.06
Bucket: Lean
Brier: 0.2809
Team-run absolute error: 2
Total-run bias: 0
Primary cause: Mixed
Secondary cause: None

Edges and controls

Season edge: +0.257
Recent edge: -0.550
Pitcher edge: -0.046
Composite edge: -0.2773
Recency guardrail: applied
Supported blowout-margin expansion was applied
Head-to-head history contained several high-scoring and blowout games

Starter projections

MIN Joe Ryan: projected 4 ER/17 outs, ERA 6.35; actual 2 ER/12 outs, ERA 4.50
DET Troy Melton: projected 2 ER/18 outs, ERA 3.00; actual 4 ER/17 outs, ERA 6.35
DET’s starter was worse than projected by 2 ER; MIN’s starter was better by 2 ER. The starter results actually supported the MIN side, but not by enough to offset other factors.

Score path and player adjustments

Historical ML: approximately 4.38–4.38
Variance-applied: 4–4
Enhanced prediction: 5–4, moving the winner toward MIN
Savant/environment: approximately 5–4
Final simulation: 5–4
The earliest winner-direction regression was the variance_applied → enhanced_prediction transition.
Player adjustments included Colt Keith, Troy Melton, Luke Keaschall, Brooks Lee, Kody Clemens, Zach McKinstry, Riley Greene, and Ryan Jeffers.

Audit and causal assessment

Structural audit: PASS
This is a wrong-winner/wrong-score game. The model’s enhanced stage converted a tie into a MIN prediction, despite recent form favoring MIN but season quality favoring DET. The evidence supports a weighting/transition issue more than a pure starter miss.

---

`NYM@MIA` — miss; wrong winner and wrong score

Predicted: NYM 4, MIA 5; winner MIA
Actual: NYM 9, MIA 4; winner NYM
Published probability: MIA 0.520771
Confidence edge: +0.0415
Bucket: Lean
Brier: 0.271202
Team-run absolute error: 6
Total-run bias: -4; predicted 9 versus actual 13
Primary cause: Lean-side miss
Secondary cause: None

Edges and controls

Season edge: +0.291
Recent edge: -0.042
Pitcher edge: +0.0386
Composite edge: +0.4145
Inexperienced-starter shrink applied; shrink factor 0.5125
The selected side remained MIA after the shrink
Score-volatility action: no_change

Starters

NYM Jonah Tong: projected 3 ER/10 outs, ERA 8.10; actual 4 ER/15 outs, ERA 7.20
MIA Eury Pérez: projected 3 ER/16 outs, ERA 5.06; actual 5 ER/12 outs, ERA 11.25
Pérez was worse than projected by 2 ER and was the main starter-side deterioration.

Score path and player adjustments

Historical ML: NYM 3.50, MIA 4.81
Variance-applied: 3–5
Enhanced: 4–5
Savant/environment: 4–5
Final: 4–5
NYM offense exceeded projection by 5 runs.
Important deviations included Juan Soto, Nick Morabito, A.J. Ewing, Graham Pauley, Bo Bichette, and others.
Several of the positive batter deviations were not necessarily supported by strong prior profiles; the audit identifies only 18 of 72 material player deviations across the slate as learnable.

Audit and causal assessment

Structural audit: PASS
This was a small-edge loss, not a near-certain miss. A confidence edge of approximately 0.04 is a lean. The score model also missed the offensive breakout.

---

`STL@SF` — strong miss; wrong winner and wrong score

Predicted: STL 7, SF 5; winner STL
Actual: STL 4, SF 5; winner SF
Published probability: STL 0.626764
Confidence edge: home-minus-away -0.2535
Bucket: Strong
Brier: 0.392833
Team-run absolute error: 3
Total-run bias: +3; predicted 12 versus actual 9
Primary cause: Wrong conviction side
Secondary cause: None

Edges and controls

Season edge: -0.5161
Recent edge: -0.372
Pitcher edge: -0.1027
Composite edge: -0.9628
Season-quality margin adjustment: -0.2168
Supported blowout-margin expansion applied
Final away margin was two runs, with the model selecting STL strongly

Starters

STL Michael McGreevy: projected 2 ER/16 outs, ERA 3.38; actual 2 ER/20 outs, ERA 2.70
SF Logan Webb: projected 4 ER/16 outs, ERA 6.75; actual 1 ER/18 outs, ERA 1.50
Webb beat projection by 3 ER and was the central starter miss.

Score path and player adjustments

Historical ML: 4.26–4.78, favoring SF
Variance-applied: 4–5, still SF
Enhanced prediction: 7–5, flipping to STL
Savant/environment: approximately 7–5
Final: 7–5
The earliest winner-direction regression occurred at variance_applied → enhanced_prediction.
Player adjustments included Bryan Torres, Leonardo Bernal, Turner Hill, Iván Herrera, Andrew Knizner, Jonah Cox, and Jung Hoo Lee.

Audit and causal assessment

Structural audit: PASS
This is one of the clearest process/model risks. The model moved from an initially SF-favorable prediction to a strong STL conviction without a corresponding realized result. The starter miss was real, but the high conviction magnified its cost.

---

`TOR@OAK` — strongest miss; wrong winner and wrong score

Predicted: TOR 6, OAK 2; winner TOR
Actual: TOR 5, OAK 6; winner OAK
Published probability: TOR 0.699558
Confidence edge: home-minus-away -0.3991
Bucket: Strong
Brier: 0.489381
Team-run absolute error: 5
Total-run bias: -3; predicted 8 versus actual 11
Primary cause: Wrong conviction side
Secondary cause: Lean-side miss

Edges and controls

Season edge: -0.9732
Recent edge: -0.906
Pitcher edge: -0.1731
Composite edge: -1.8626
High-margin-pick diagnostic: present
Large Savant run-environment swing: present
Supported blowout-margin expansion applied
High-raw cluster member; historical empirical hit rate for the cluster was 0.5614

Starters

TOR Dylan Cease: projected 2 ER/18 outs, ERA 3.00; actual 4 ER/15 outs, ERA 7.20
OAK Jacob Lopez: projected 5 ER/15 outs, ERA 9.00; actual 4 ER/18 outs, ERA 6.00
Cease was worse than projected by 2 ER; Lopez was better than projected by 1 ER.

Score path and player adjustments

Historical ML: TOR 5.90, OAK 3.64
Variance-applied: 6–4
Enhanced: 6–3
Savant/environment: 6–2
Final simulation: 6–2
The score path repeatedly reinforced the already incorrect TOR conviction.
Player adjustments included Dylan Cease, Nathan Lukes, Zack Gelof, Alejandro Kirk, Carlos Cortes, Myles Straw, Vladimir Guerrero Jr., and Jonah Heim.
OAK offense was underprojected by four runs.

Audit and causal assessment

Structural audit: PASS
This is the most damaging individual miss:
Highest published confidence
Highest Brier error among the slate
High-raw cluster failure
Large margin error
It supports lowering confidence caps and widening tails, but not a one-game reversal rule for Toronto/Oakland.

---

`WAS@SD` — correct winner, wrong score

Predicted: WAS 5, SD 6; winner SD
Actual: WAS 2, SD 3; winner SD
Published probability: SD 0.529654
Confidence edge: +0.0593
Bucket: Lean
Brier: 0.221225
Team-run absolute error: 6
Total-run bias: +6; predicted 11 versus actual 5
Winner quality: Right winner, wrong score

Edges and controls

Season edge: -0.0369
Recent edge: -0.3847
Pitcher edge: -0.0934
Composite edge: -0.1419
Starter-experience/recency guardrail applied
SD starter had only four prior appearances in the feed
Large Savant run-environment swing flag present
Score-volatility action: no_change

Starters

WAS Jake Irvin: projected 3 ER/14 outs, ERA 5.79; actual 3 ER/17 outs, ERA 4.76
SD Nick Pivetta: projected 4 ER/15 outs, ERA 7.20; actual 0 ER/15 outs, ERA 0.00
Pivetta was four ER better than projected; the thin four-appearance history explains high projection uncertainty.

Score path and player adjustments

Important player deviations included CJ Abrams, Dylan Crews, Jackson Merrill, Luis Campusano, and Keibert Ruiz.
Both offenses scored below projection.
Merrill’s positive deviation was identified as one of the more learnable player-level deviations because of above-average xwOBA.

Audit and causal assessment

Structural audit: PASS
Correct winner/wrong score. The supplied recommendation is to reduce confidence for low-history starters, prefer priors, and validate run-parameter changes in shadow.

---

4. Miss reconstruction

Earliest wrong-direction stages

Loss	Earliest stage moving toward wrong result	Reconstruction
`ARI@KC`	Historical ML already favored KC	The wrong side was present before later controls; subsequent stages maintained KC. Starter identity/feed inconsistency is a data-quality concern, but no later winner-direction flip is documented.
`MIN@DET`	`enhanced_prediction`	Historical ML was approximately tied; variance stayed tied; enhanced prediction changed `TIE → MIN`, which was wrong.
`NYM@MIA`	Historical ML already favored MIA	Variance, enhanced, Savant, and simulation all maintained MIA. This was a small-edge miss rather than a later regression.
`STL@SF`	`enhanced_prediction`	Historical ML and variance favored SF; enhanced prediction changed `SF → STL`, and later stages preserved the wrong side.
`TOR@OAK`	Historical ML already favored TOR	All stages reinforced TOR, with Savant and score-volatility adjustments widening the wrong margin.

A confidence edge of 0.12 would be moderate, not a near coin flip. More generally, the system’s own edges must not be interpreted as half-scale distances from 50%. The ARI edge was approximately 0.16 and still lost; STL and TOR had much stronger edges and also lost. Conversely, several 0.01–0.06 edges won because they were genuinely uncertain calls.

Loss-type classification

Bad input

ARI has the clearest input/contract issue:
Away starter identity mismatch
Missing settled pitching line for the listed away starter
No other game has a reported structural input error.

Bad weighting or transition

MIN@DET: enhanced_prediction changed a tie into the wrong MIN direction.
STL@SF: enhanced_prediction changed an SF direction into the wrong STL direction.
TOR@OAK: strong season/recent/compound edges were allowed to produce a four-run predicted margin, despite the broader slate showing severe margin compression and high uncertainty.

Calibration

The most consequential calibration miss was TOR@OAK at published probability 0.699558.
STL@SF also received a strong 0.626764 probability and lost.
Three of four high-conviction selections lost.
The appropriate conclusion is overconfidence at the upper end, not that every moderate-confidence call should be flattened.

Starter error

STL@SF: Logan Webb, actual ERA 1.50 versus projected 6.75.
TOR@OAK: Dylan Cease, actual ERA 7.20 versus projected 3.00.
MIN@DET: Troy Melton actual ERA 6.35 versus projected 3.00.
NYM@MIA: Eury Pérez actual ERA 11.25 versus projected 5.06.
ARI@KC: the listed away starter is unresolved; KC’s Noah Cameron beat projection modestly.
Starter error explains substantial score and directional pressure, but these are mostly outlier outcomes rather than proven systematic starter-model failures.

Offense/bullpen miss

NYM@MIA: NYM offense exceeded projection by 5 runs.
TOR@OAK: OAK offense exceeded projection by 4 runs relative to the matchup breakdown.
STL@SF: STL offense scored three fewer than projected.
MIN@DET: opposing offensive and starter effects were mixed.
Bullpen effects were material in some games, such as KC’s four ER from Craig Kimbrel and BAL’s three ER from Andrew Kittredge, but those did not independently explain all winner misses.

Run-environment miss

The slate overprojected total scoring in:

ATL@PHI: +6
CHC@MIL: +5
WAS@SD: +6

It underprojected:

NYM@MIA: -4
CLE@BAL: -3
TOR@OAK: -3

This supports the persistent weak-run-signal diagnosis rather than a single consistent high- or low-scoring bias.

Irreducible variance

The player forensics identify:

72 material player deviations
Only 18 supported by the hitter’s own season contact quality
Learnable share: 25%
Variance: 54 deviations

Several major starter outcomes were also explicitly labelled outliers. A model change based on every individual over- or underperformance would learn noise.

Pipeline-stage quality across all stages

The stored stage metrics were:

Stage	Winner accuracy	Team-run MAE	Total-run MAE	Margin MAE
`historical_ml`	6/11 = 0.5455	1.5782	2.3421	2.1950
`variance_applied`	5/11 = 0.4545	1.5909	2.4545	2.0909
`enhanced_prediction`	6/11 = 0.5455	1.8182	2.9091	2.0000
`embedded_savant_run_environment`	6/11 = 0.5455	1.9091	3.0909	2.1818
`fundamental_edge_target_review`	6/11 = 0.5455	1.9091	3.0909	2.1818
`capped_target`	6/11 = 0.5455	1.9091	3.0909	2.1818
`simulation_result`	6/11 = 0.5455	1.9091	3.0909	2.1818
`published_final`	6/11 = 0.5455	1.9091	3.0909	2.1818

Transition audit

Historical ML → variance applied

Winner accuracy fell from 6 to 5.
CIN@LAD, CLE@BAL, and MIN@DET moved toward ties.
CLE@BAL is explicitly listed as regressed.
Margin MAE improved by 0.1041, but winner accuracy worsened by 0.091.

Variance applied → enhanced prediction

Winner accuracy recovered from 5 to 6.
CIN@LAD and CLE@BAL improved.
STL@SF regressed from SF to STL.
Team-run MAE increased by 0.2273.
This transition contains a documented intermediate winner-direction regression; the later aggregate recovery does not erase it.

Enhanced prediction → embedded Savant run environment

No winner-direction changes.
Team-run MAE worsened by 0.0909.
Total-run MAE worsened to 3.0909.
Margin MAE worsened by 0.1818.

All later transitions

No winner-direction changes.
No measurable score changes.
simulation_result → published_final made no change to any recorded metric.

The simulation therefore did not intervene in winner direction on this slate, and this same inertness has persisted for 11 consecutive slates.

---

5. Cross-game counterfactuals

Lever tension: widen run distributions

Lever: score.variance_multiplier / score.variance_floor

Target metrics: Brier score and calibration error

Games helped

ARI@KC: moderate-confidence miss; wider tails reduce overconfidence.
STL@SF: strong-confidence miss; wider tails reduce overconfidence.
TOR@OAK: strongest miss; wider tails reduce overconfidence.

Game hurt

LAA@BOS: strong correct call; widening the distribution would soften a good edge.

Net evidence

3 games helped versus 1 hurt.
The lever tension supports a bounded shadow experiment, not an unrestricted increase.
This lever is primarily about uncertainty and score tails; it should not be presented as a guaranteed winner-accuracy fix.

Lever tension: pitcher form weights and prior-game shrinkage

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

Target metrics: starter ER MAE and starter outs MAE

Games helped

CHC@MIL: Gasser was three ER better than projected.
CIN@LAD: Burns was three ER better than projected.
STL@SF: Webb was three ER better than projected.
WAS@SD: Pivetta was four ER better than projected.

Games hurt

None listed.

Net evidence

4 helped, 0 hurt in the supplied slate-level tension table.
However, the same document says many starter/player deviations are variance and that persistent cross-slate starter ERA bias is not separable from noise.
A bounded shadow test is reasonable; direct promotion is not.

Lever tension: recent/season run-mean weighting

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

Target metrics: team-run MAE, total-run MAE, accuracy

Games helped

ATL@PHI
CHC@MIL
NYM@MIA
TOR@OAK
WAS@SD

Games hurt

CIN@LAD
LAA@BOS
MIN@DET

Net evidence

5 helped versus 3 hurt.
The net count is mildly favorable, but the persistent cross-slate evidence says lineup-run bias is not separable from noise: mean -0.06, z-score -0.3, nonpersistent.
Do not make a large direct change from this slate.

Tempting one-game fixes that must not be made

Do not reverse the STL pick rule.

STL@SF was a severe miss, but a single game does not justify suppressing strong negative composite edges.

Do not permanently downgrade Toronto or Oakland.

TOR@OAK was the strongest miss, but the evidence supports broader confidence/variance control, not a team-specific rule.

Do not tune to individual player box scores.

Only 18 of 72 material player deviations were supported by player contact-quality evidence; the other 54 are classified as variance.

Do not remove starter adjustments entirely.

Several games show useful starter information, and the pitcher-form lever helped four games with no listed hurts.

Do not use one slate to alter calibration weights aggressively.

The causal register explicitly says the slate is too small for a stable confidence curve.

Do not assume the probability governor was bypassed.

Trace pick-side math for the four unexplained deltas first.

---

6. Data and contract defects

Critical findings

1. Unexplained published probability transformations

Affected games:

ARI@KC: calibrated 0.618; published 0.579623; delta -0.038377
CIN@LAD: calibrated 0.5751; published 0.532561; delta -0.042539
LAA@BOS: calibrated 0.6539; published 0.62315; delta -0.03075
NYM@MIA: calibrated 0.5477; published 0.520771; delta -0.026929

The lineage says “no recorded transformation explains the publication delta” for all four. This is a critical contract failure because a material probability transformation is missing from evidence.

Error findings

No error findings were reported in the supplied structural audits:

Aggregate structural audit: error: 0
Every per-game structural audit: error: 0

This does not mean the four probability-lineage issues are resolved; they are raised in a separate causal contract register.

Warning findings

No warning findings were reported:

Aggregate: warning: 0
Every per-game structural audit: warning: 0

Informational findings

Aggregate informational findings: 81
Per-game informational counts range from 7 to 8.
The supplied object does not include the individual informational finding texts, so the 81 entries cannot be enumerated by name without fabricating details.
Known informational conditions include:
Simulation produced zero winner flips
Full slate structural coverage
Score-volatility no_change decisions
Starter-recency and experience context
Forecast-file informational notes

Prediction-model defects versus artifact bugs

Prediction-model defects or risks

Near-zero run-model signal:
Total R² 0.018362
Home R² 0.012404
Away R² 0.020536
Upper-tail confidence overstatement:
High-conviction miss rate 75%
TOR high-raw cluster miss
Persistent simulation inertness
Intermediate winner-direction regressions at the enhanced-prediction stage
Margin compression and poor score reliability

Presentation/artifact/contract bugs

Missing integrity revision audit
Four unexplained calibrated-to-published probability deltas
ARI starter identity/missing-line inconsistency
No market cross-check because no source-labelled prices were supplied

The per-game forecast files themselves were reported clean, so the primary defect is not a rendering or saved-file corruption issue.

Complete probability lineage audit

The required conceptual path is:

raw model → calibrated → governor → score coherence/recency → published

Correctly recorded transformations

Examples:

CHC@MIL, CLE@BAL, STL@SF, TOR@OAK: published probability approximately matches calibrated/governed value.
ATL@PHI, MIN@DET, WAS@SD: lineage records score coherence followed by matchup-recency guardrail.
ARI@KC, NYM@MIA: starter-experience shrink is recorded in the game trace.

Material unresolved transformations

The following have no recorded transformation explaining the published delta:

ARI@KC
CIN@LAD
LAA@BOS
NYM@MIA

These are bypass-like symptoms, but they must not be called a governor bypass until pick-side math is traced. The causal register’s guardrail is appropriate.

No bypass conclusion for other games

No material unexplained delta is present for:

CHC@MIL
CLE@BAL
STL@SF
TOR@OAK

For those games, the published value matches the calibrated/governed or documented post-governor value within rounding.

Coverage reconciliation

Coverage is complete at the game level:

11/11 expected games structurally audited
11/11 with starter comparisons
11/11 with continuous-learning records
11/11 with forensic deep audits

The slate is therefore structurally covered, but not contractually clean because coverage does not resolve the four probability deltas, the missing integrity revision audit, or ARI’s starter inconsistency.

---

7. Cross-slate persistence

The supplied cross-slate window contains 11 slates, with 10 prior slates loaded.

Escalated findings — established failures

These are not single-slate noise.

`simulation_inert_on_slate`

Consecutive-slate count: 11
Incidence in window: 11 of 11
Present now: yes
Metric group: run model
Target lever listed: simulation.allow_score
Target metrics: team_run_mae, total_run_mae
Established direction of fix: confirm the simulation module is executing and not silently no-oping
Required action: code/diagnostic investigation, not a parameter promotion

The self-assessment provides a specific suspected cause: the simulation distribution is built from the same draws used to search for a target-realizing play-by-play, and the loop can terminate after approximately two selected draws. That makes the estimated win rate non-probabilistic and explains zero flip candidates over 11 slates.

`weak_run_model_signal`

Consecutive-slate count: 11
Incidence in window: 11 of 11
Present now: yes
Metric group: run model
Target levers listed: training.xgb.n_estimators, training.xgb.max_depth, training.xgb.learning_rate
Target metrics: team_run_mae, total_run_mae
Established direction of fix: investigate features and target through an isolated training backtest
Required action: code/data/training investigation, not a nightly knob adjustment

The raw evidence remains poor:

Away R² 0.020536
Home R² 0.012404
Total R² 0.018362

The self-assessment specifically recommends adding, where valid and source-labelled:

No-vig market probability
Park/weather run environment
Bullpen rest/fatigue
Platoon-adjusted lineup quality

It also requires a model schema-version bump when feature semantics change.

Every finding streak

Finding	Consecutive count	Incidence	Present now?	Interpretation
`high_raw_probability_cluster`	0	3/11	No	Historical recurrence, not current
`pipeline_winner_instability`	0	4/11	No	Historical recurrence, not current
`score_margin_compression`	0	3/11	No	Historical recurrence, not current
`simulation_inert_on_slate`	11	11/11	Yes	Established persistent failure
`single_slate_calibration_instability`	1	6/11	Yes	Current finding, not yet persistent by consecutive-slate count
`weak_run_model_signal`	11	11/11	Yes	Established persistent failure

The fact that single_slate_calibration_instability occurred in 6 of 11 slates but has only a one-slate consecutive streak is important: its incidence is notable, but it has not reached the escalation status of the two 11-slate failures.

Metric trends

Each reported direction is:

Margin bias: improving

Series ends at -0.182, better than earlier values as low as -3.5.

Mean team-run error on losses: improving

Current value 3.8, compared with earlier values up to 8.25.

Weighted calibration error: regressing

Current value 0.2611, worse than many preceding values, including 0.0937 and 0.1000.

Winner accuracy: flat

Current slate 0.5455, rolling mean 0.5667, with substantial slate-to-slate noise.

Top-confidence-pick loss streak

The current top-confidence pick has a loss streak of 2:

2026-09-06: miss
2026-09-07: miss

The full sequence includes several earlier misses, but the supplied explicit streak is two consecutive losses. This is concerning when combined with the current high-conviction miss rate of 75%, though still weaker evidence than the 11-slate escalations.

Cross-slate bias findings

The two listed cross-slate bias findings are not actionable yet:

Starter ERA error positive

Mean: +0.281
Same-sign share: 0.625
5 of 8 slates
z = 0.95
Status: hold
No change to context.opponent_strength_weight

Lineup run delta negative

Mean: -0.06
Same-sign share: 0.625
z = -0.3
Status: hold
No change to score.variance_floor

The source correctly treats these as not separable from slate-to-slate noise.

---

8. Adjustments for today

Recommendations are ranked by persistence and scope. Persistent issues receive code/diagnostic actions rather than forced parameter moves, because the supplied escalation records explicitly state that they are not parameter-actionable.

1. Simulation inertness — highest persistence and scope

Finding: simulation_inert_on_slate
Persistence: 11 consecutive slates
Registered lever: simulation.allow_score
Parameter move: Abstain from changing the parameter.
Bounded action: Inspect and repair the simulation execution path:
Separate target-realization search from independent win-probability simulation.
Use an independent unsteered batch of approximately 200–500 draws or an equivalent closed-form distribution.
Remove or revise the early-exit behavior around MIN_DRAWS_BEFORE_CLOSE_EXIT.
Register the batch size as a parameter.
Feed the independent win rate into _distribution_agreement.
Expected metric: team_run_mae and total_run_mae should become measurable and, if the module is contributing valid information, improve; winner-direction intervention should become nonzero when justified.
Helps: Broadly addresses all games, especially wrong-margin cases such as TOR@OAK, STL@SF, ATL@PHI, and WAS@SD.
Hurts/regression risk: An independent simulation may alter some correct calls, including LAA@BOS; rollback must monitor winner accuracy, Brier score, and score MAE.
Evidence strength: Highest, because the failure persisted 11 consecutive slates and has a concrete code-level diagnosis.

2. Weak run-model signal — persistent feature/target failure

Finding: weak_run_model_signal
Persistence: 11 consecutive slates
Registered levers: training.xgb.n_estimators, training.xgb.max_depth, training.xgb.learning_rate
Parameter move: Do not tune these nightly parameters directly.
Bounded action: Run the registered isolated training backtest after adding or validating:
No-vig market probability where source-labelled
Park/weather run environment
Bullpen fatigue/rest
Platoon-adjusted lineup quality
Expected metric: Improve holdout home/away/total R² and reduce team-run/total-run MAE.
Helps: Broad score failures such as ATL@PHI, NYM@MIA, TOR@OAK, and WAS@SD.
Hurts/regression risk: Feature changes could disturb close correct score calls such as LAA@BOS and CIN@LAD; require same-fingerprint holdout comparison and rollback.
Evidence strength: Highest, because all three run R² values are near zero and the pattern persisted 11 slates.

3. Upper-tail confidence overstatement — strongest current single-slate driver

Finding: Current high-confidence performance is anti-predictive; top-pick loss streak is 2 and high-conviction miss rate is 75%.
Registered parameter: confidence.base_cap
Bounded change: Reduce confidence.base_cap by a small amount, approximately 0.03, in shadow only.
Expected metric: Lower weighted calibration error and improve Brier score by reducing overconfident losses.
Helps: ARI@KC, STL@SF, and especially TOR@OAK, all of which lost with meaningful conviction.
Hurts: LAA@BOS, a strong correct call whose published probability would be softened.
Do not simultaneously make a large moderate_cap and variance-floor change: isolate this confidence-cap experiment so the effect is identifiable.
Evidence strength: Strong for the current slate, but weaker than the 11-slate code failures because the calibration instability streak is only one slate.

4. Score-distribution shadow experiment

Registered parameters: score.variance_multiplier or score.variance_floor
Bounded change: Test a small increase in score.variance_multiplier, approximately +0.05, in shadow.
Expected metric: Lower Brier score, calibration error, and margin overconfidence; possibly improve total-run error.
Helps: ARI@KC, STL@SF, TOR@OAK
Hurts: LAA@BOS; potentially other strong correct calls.
Evidence strength: Moderate. The lever tension is 3 helped versus 1 hurt, but cross-slate lineup-run bias does not support a direct permanent change. This should remain a shadow experiment with rollback monitoring.

5. Starter venue/form experiment

Registered parameter: context.venue_adjustment_weight
Bounded change: Increase modestly in shadow, only for starters with a credible venue split.
Expected metric: Lower starter ERA MAE and potentially reduce total-run MAE.
Helps most directly: CIN@LAD, where the supplied forensic assessment explicitly says the venue weakness was underweighted.
Potential harms: Other games where venue splits are noisy or insufficiently sampled.
Evidence strength: Moderate and game-specific; do not promote from this slate alone.

6. No direct change to `context.opponent_strength_weight`

The cross-slate evidence says the positive starter ERA bias is not separable from noise:

Mean +0.281
z = 0.95
Status: hold

This lever should remain unchanged pending more evidence.

---

9. Bottom line

Known

The system went 6–5, below the supplied walk-forward baseline.
Winner accuracy was materially better than score accuracy.
Two games had correct winners but incorrect scores: ATL@PHI and WAS@SD; other correct winners also had substantial run errors.
Three of four high-conviction calls lost: STL@SF, TOR@OAK, and one of the four strong/upper-confidence selections as counted in the supplied high-conviction set.
The worst individual miss was TOR@OAK, with published probability 0.699558 and Brier score 0.489381.
Run-model holdout signal is near zero and has persisted across 11 slates.
Simulation winner-direction intervention has been inert for 11 consecutive slates.
The enhanced-prediction transition introduced documented wrong-direction regressions in MIN@DET and STL@SF.
Four probability publications have unexplained material deltas.
One starter record in ARI has an identity/missing-line inconsistency.
Structural game coverage was complete, with zero reported critical/error/warning findings in the structural audit layer.

Uncertain

Whether the four probability deltas represent an actual governor bypass, a side-orientation transformation, or an undocumented but valid post-processing step.
Whether the observed high-confidence failures reflect persistent calibration bias or a bad single slate; the upper-tail result is concerning, but only one current consecutive slate is recorded for calibration instability.
Whether pitcher-form or venue adjustments will improve future results rather than simply react to outliers.
How much of the run error is model deficiency versus irreducible player and starter variance.
Whether the missing integrity-revision artifact contains additional remediation evidence.

Safest next action

Repair and independently validate the simulation probability path before promoting any model or confidence change. The simulation inertness is persistent, has a concrete code-level explanation, and can be tested without committing to a risky one-game or one-slate parameter adjustment. In parallel, keep confidence-cap and score-variance changes in shadow mode only, with Brier score, calibration error, winner accuracy, team-run MAE, and total-run MAE monitored for rollback.

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	BAL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	BAL
info	CURRENT_PITCHER_DUPLICATION	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	BAL model pitcher projections differ from final simulation pitching box score (H: current=6, sim=7).	BAL
info	CURRENT_FINALIZED_MATCH_BOX	BAL model-prediction batting totals differ from final simulation box score (AB: current=31, sim=39; H: current=7, sim=9; 2B: current=2, sim=0; HR: current=1, sim=2; BB: current=3, sim=0; SO: current=10, sim=12).	BAL
info	CURRENT_PLAYER_MATCH_BOX	BAL current finalized hitter lines are model predictions and differ from the simulation box score: Blaze Alexander: AB current=3 box=5, R current=0 box=1; Jackson Holliday: AB current=4 box=5, HR current=0 box=1, RBI current=0 box=2; Christian Encarnacion-Strand: AB current=3 box=4, 2B current=1 box=0, RBI current=1 box=0; Pete Alonso: AB current=4 box=5, H current=1 box=2, RBI current=1 box=2, BB current=1 box=0; Coby Mayo: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Rece Hinds: AB current=3 box=4, H current=0 box=1, SO current=2 box=1; Gunnar Henderson: H current=1 box=0, 2B current=1 box=0, R current=1 box=0, BB current=1 box=0; Samuel Basallo: RBI current=1 box=0, BB current=1 box=0, SO current=1 box=2	BAL
info	CURRENT_FINALIZED_MATCH_BOX	CLE model-prediction batting totals differ from final simulation box score (AB: current=33, sim=37; H: current=8, sim=7; R: current=4, sim=3; BB: current=3, sim=4; SO: current=7, sim=8; HBP: current=0, sim=1).	CLE
info	CURRENT_PLAYER_MATCH_BOX	CLE current finalized hitter lines are model predictions and differ from the simulation box score: Jo Adell: H current=1 box=2, RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Angel Martínez: AB current=4 box=5, 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=2; David Fry: AB current=3 box=4, H current=1 box=0; Angel Genao: AB current=4 box=5, 2B current=0 box=1, R current=1 box=0; José Ramírez: H current=1 box=0, R current=1 box=0, SO current=0 box=1; Petey Halpin: SO current=1 box=0, HBP current=0 box=1; Austin Hedges: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Steven Kwan: AB current=4 box=5, H current=1 box=2, RBI current=0 box=1, BB current=1 box=0	CLE
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
LAA	Grayson Rodriguez	3 ER / 15 outs	5 ER / 12 outs	2
BOS	Brayan Bello	2 ER / 13 outs	0 ER / 15 outs	-2

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
LAA	Grayson Rodriguez (Starter)	12 / 15	11.25 / 5.4 (+5.85)	5 / 3	6 / 4	2 / 2	4 / 0	3 / 5	venue away: 3.9 vs 7.54 ERA (ok) · opp run factor 1.01 (average) · 16 prior appearances this season
LAA	Ben Joyce (Closer)	3 / 7	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0 / 2
BOS	Brayan Bello (Starter)	15 / 13	0 / 4.15 (-4.15)	0 / 2	4 / 4	0 / 1	2 / 2	7 / 4	venue home: 10.95 vs 10.03 ERA (worse here) · opp run factor 0.99 (weak) · 22 prior appearances this season
BOS	Garrett Whitlock (Middle Reliever (High Leverage))	3 / 4	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 0

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
LAA	Moisés Ballesteros (1B)	4 / 3	3 / 0	1 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	0 / 1	2.65	0.289 below average	good day
BOS	Jarren Duran (LF)	4 / 4	2 / 1	1 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	1 / 1	1.71	0.296 below average	good day
LAA	Mike Trout (DH)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	3 / 1	-1.26	0.375 elite	bad day
BOS	Isiah Kiner-Falefa (2B)	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	2 / 0	0 / 0	1.13	0.307 average	no signal
LAA	Josh Lowe (RF)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	-0.8	0.291 below average	projection too high
LAA	Zach Neto (SS)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.8	0.305 average	no signal
BOS	Trevor Story (SS)	4 / 5	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	-0.78	0.252 weak	projection too high
BOS	Caleb Durbin (3B)	4 / 3	2 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0.78	0.289 below average	good day
BOS	Wilyer Abreu (RF)	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	-0.78	0.347 elite	bad day
LAA	Wade Meckler (LF)	2 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	2 / 0	0 / 1	0.66	0.311 average	no signal
LAA	Kyren Paris (CF)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	3 / 1	-0.47	0.698 elite	—
BOS	Mickey Gasper (1B)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	-0.47	0.377 elite	—
BOS	Eli White (CF)	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	1 / 1	0.47	0.275 weak	—
LAA	Vaughn Grissom (3B)	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	3 / 1	-0.45	0.314 average	—
LAA	Christian Moore (2B)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	2 / 1	0	0.249 weak	—
LAA	Tyler Heineman (C)	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.243 weak	—
BOS	Roman Anthony (DH)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0	0.356 elite	—
BOS	Adley Rutschman (C)	4 / 5	1 / 1	0 / 0	0 / 0	1 / 1	3 / 1	1 / 1	0 / 0	1 / 1	0	0.334 above average	—

Signal	Home-away edge	Meaning
Season quality	1.013	positive favors home; negative favors away
Recent form	0.658	positive favors home; negative favors away
Starting-pitcher matchup	0.02	positive favors home; negative favors away
Composite	1.125	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.26-4.91	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-5	0.623	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average)
embedded_savant_run_environment	3-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-5	n/a	Capped at max 12 realistic runs
simulation_result	3-5	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	4.26-4.91	4-5	0.348	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	3-5	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average)
embedded_savant_run_environment	3-5	3-5	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-5	3-5	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-5	3-5	0	Capped at max 12 realistic runs
simulation_result	3-5	3-5	0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
BOS	Mickey Gasper	batter / 1B	better	predicted_RBI 3.569
BOS	Roman Anthony	batter / DH	worse	predicted_SO 1.842
BOS	Trevor Story	batter / SS	worse	predicted_AB -1.64
BOS	Wilyer Abreu	batter / RF	worse	predicted_AB -1.5
LAA	Josh Lowe	batter / RF	better	predicted_H 1.391
LAA	Zach Neto	batter / SS	worse	predicted_SO -1.384
LAA	Christian Moore	batter / 2B	better	predicted_H 1.38
LAA	Moisés Ballesteros	batter / 1B	worse	predicted_H 1.368

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	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_FINALIZED_MATCH_BOX	BOS model-prediction batting totals differ from final simulation box score (AB: current=34, sim=30; H: current=8, sim=6; 2B: current=2, sim=0; HR: current=1, sim=3; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=0; SO: current=7, sim=8).	BOS
info	CURRENT_PLAYER_MATCH_BOX	BOS current finalized hitter lines are model predictions and differ from the simulation box score: Roman Anthony: BB current=1 box=0, SO current=1 box=3; Mickey Gasper: AB current=3 box=4, H current=1 box=2, HR current=0 box=2, RBI current=1 box=4; Trevor Story: AB current=5 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Adley Rutschman: AB current=5 box=4; Caleb Durbin: H current=1 box=0, SO current=0 box=1; Eli White: RBI current=1 box=0, R current=0 box=1; Wilyer Abreu: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, R current=1 box=0; Jarren Duran: AB current=4 box=3, R current=0 box=1, SO current=1 box=0	BOS
info	CURRENT_FINALIZED_MATCH_BOX	LAA model-prediction batting totals differ from final simulation box score (AB: current=31, sim=35; H: current=7, sim=9; 2B: current=1, sim=2; 3B: current=0, sim=2; BB: current=3, sim=2; SO: current=9, sim=6; SF: 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: AB current=3 box=4, H current=1 box=0, R current=1 box=0, SO current=1 box=2; Vaughn Grissom: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Mike Trout: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Moisés Ballesteros: AB current=3 box=4, H current=0 box=2, 2B current=0 box=1, SO current=1 box=2; Josh Lowe: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, HR current=0 box=1; Kyren Paris: 3B current=0 box=1, RBI current=0 box=1, R current=0 box=1, BB current=0 box=1; Zach Neto: RBI current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=1 box=0; Christian Moore: H current=1 box=2, 3B current=0 box=1, RBI current=0 box=1, R current=0 box=1	LAA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
MIN	Joe Ryan	4 ER / 17 outs	2 ER / 12 outs	-2
DET	Troy Melton	2 ER / 18 outs	4 ER / 17 outs	2

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
MIN	Joe Ryan (Starter)	12 / 17	4.5 / 6.35	2 / 4	3 / 7	1 / 0	1 / 1	7 / 6	venue away: 3.88 vs 3.44 ERA (ok) · opp run factor 0.95 (average) · 23 prior appearances this season
DET	Troy Melton (Starter)	17 / 18	6.35 / 3 (+3.35)	4 / 2	7 / 3	1 / 1	1 / 2	3 / 5	venue home: 2.37 vs 2.06 ERA (ok) · opp run factor 0.99 (average) · 17 prior appearances this season
DET	Kenley Jansen (Closer)	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 1	0 / 0	1 / 0	0 / 0

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
MIN	Trevor Larnach (LF)	3 / 3	2 / 1	0 / 0	0 / 0	1 / 0	1 / 0	2 / 0	0 / 0	1 / 1	1.4	0.327 above average	projection too low
MIN	Ryan Jeffers (C)	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	1 / 1	-1.11	0.357 elite	bad day
DET	Zach McKinstry (RF)	3 / 4	1 / 1	0 / 0	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.62	0.282 weak	good day
DET	Riley Greene (LF)	4 / 4	3 / 1	0 / 0	0 / 0	1 / 1	2 / 1	1 / 1	0 / 1	0 / 1	0.61	0.369 elite	projection too low
MIN	Kody Clemens (2B)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	2 / 1	-0.6	0.339 elite	bad day
DET	Kevin McGonigle (3B)	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 1	0 / 1	0 / 0	0.6	0.35 elite	projection too low
MIN	Royce Lewis (1B)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.315 average	—
DET	Colt Keith (DH)	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.335 above average	—
DET	Dillon Dingler (C)	4 / 5	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	3 / 1	-0.47	0.343 elite	—
MIN	Walker Jenkins (CF)	3 / 3	1 / 0	1 / 0	0 / 0	0 / 0	2 / 0	0 / 0	0 / 1	0 / 1	0.45	0.24 weak	—
MIN	Luke Keaschall (RF)	2 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	2 / 1	0 / 0	0.33	0.314 average	—
MIN	Brooks Lee (3B)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0 / 0	0.33	0.276 weak	—
DET	Gleyber Torres (2B)	4 / 5	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	2 / 1	-0.33	0.327 above average	—
DET	John Peck (SS)	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.31	0.325 above average	—
MIN	Kaelen Culpepper (SS)	4 / 4	2 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	0 / 1	0.16	0.313 average	—
MIN	Josh Bell (DH)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 1	-0.14	0.331 above average	—
DET	Max Clark (CF)	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	2 / 1	0.02	0.311 average	—
DET	Spencer Torkelson (1B)	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.31 average	—

Signal	Home-away edge	Meaning
Season quality	0.257	positive favors home; negative favors away
Recent form	-0.55	positive favors home; negative favors away
Starting-pitcher matchup	-0.046	positive favors home; negative favors away
Composite	-0.277	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	4.38-4.38	n/a	Base ML historical model prediction
variance_applied	4-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.53	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	n/a	Capped at max 12 realistic runs
simulation_result	5-4	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	4.38-4.38	4-4	-0.003	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-4	5-4	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	5-4	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	5-4	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	5-4	-0	Capped at max 12 realistic runs
simulation_result	5-4	5-4	-0	Final inning-by-inning simulation result

Team	Player	Role	Net effect	Largest change
DET	Colt Keith	batter / DH	better	predicted_RBI 3.756
DET	Troy Melton	pitcher / Starter	better	predicted_ER -3
MIN	Luke Keaschall	batter / RF	worse	predicted_RBI 1.669
MIN	Brooks Lee	batter / 3B	worse	predicted_AB -1.39
MIN	Kody Clemens	batter / 2B	better	predicted_AB -1.35
DET	Zach McKinstry	batter / RF	better	predicted_H 1.35
DET	Riley Greene	batter / LF	worse	predicted_AB -1.33
MIN	Ryan Jeffers	batter / C	worse	predicted_SO 1.316

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	DET pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	DET
info	CURRENT_PITCHER_DUPLICATION	MIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	MIN
info	CURRENT_FINALIZED_MATCH_BOX	DET model-prediction batting totals differ from final simulation box score (AB: current=36, sim=34; H: current=8, sim=7; 2B: current=2, sim=3; HR: current=1, sim=0; BB: current=3, sim=2; SO: current=8, sim=9; HBP: current=0, sim=1).	DET
info	CURRENT_PLAYER_MATCH_BOX	DET current finalized hitter lines are model predictions and differ from the simulation box score: Colt Keith: AB current=3 box=4, 2B current=0 box=1, RBI current=0 box=4, R current=0 box=1; Kevin McGonigle: BB current=1 box=0; Riley Greene: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Gleyber Torres: AB current=5 box=4, R current=1 box=0, BB current=1 box=0, SO current=1 box=0; Zach McKinstry: H current=1 box=2, 2B current=0 box=1, R current=0 box=1, SO current=1 box=0; Dillon Dingler: AB current=5 box=4, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; John Peck: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=2; Max Clark: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0	DET
info	CURRENT_FINALIZED_MATCH_BOX	MIN model-prediction batting totals differ from final simulation box score (AB: current=34, sim=33; H: current=8, sim=6; 2B: current=2, sim=0; SO: current=7, sim=5).	MIN
info	CURRENT_PLAYER_MATCH_BOX	MIN current finalized hitter lines are model predictions and differ from the simulation box score: Luke Keaschall: RBI current=0 box=2, BB current=1 box=0, SO current=0 box=1; Kody Clemens: AB current=4 box=3, HR current=1 box=0, BB current=0 box=1, SO current=1 box=0; Brooks Lee: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Royce Lewis: H current=1 box=2, R current=0 box=1, SO current=1 box=0; Josh Bell: H current=1 box=0, R current=1 box=0, SO current=1 box=0; Ryan Jeffers: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Trevor Larnach: AB current=3 box=4, R current=0 box=1, SO current=1 box=0; Kaelen Culpepper: 2B current=1 box=0, HR current=0 box=1, SO current=1 box=2	MIN
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
NYM	Jonah Tong	3 ER / 10 outs	4 ER / 15 outs	1
MIA	Eury Pérez	3 ER / 16 outs	5 ER / 12 outs	2

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
NYM	Jonah Tong (Starter)	15 / 10	7.2 / 8.1	4 / 3	4 / 8	1 / 1	1 / 1	7 / 5	opp run factor 1 (average) · thin sample: 3 prior appearances this season (high uncertainty)
NYM	Daniel Duarte (Setup Man)	3 / 6	0 / 0	0 / 0	0 / 0	0 / 0	0 / 3	0 / 0
MIA	Eury Pérez (Starter)	12 / 16	11.25 / 5.06 (+6.19)	5 / 3	10 / 7	0 / 1	3 / 1	5 / 1	venue home: 2.81 vs 4.54 ERA (ok) · opp run factor 0.96 (weak) · 25 prior appearances this season

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
MIA	Javier Sanoja (2B)	3 / 3	2 / 1	0 / 0	1 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	1.42	0.288 below average	good day
MIA	Heriberto Hernández (LF)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	3 / 1	-1.4	0.355 elite	bad day
MIA	Graham Pauley (3B)	3 / 3	1 / 0	0 / 0	0 / 0	1 / 0	3 / 0	1 / 0	0 / 0	2 / 1	1.4	0.231 weak	good day
NYM	Nick Morabito (LF)	4 / 3	2 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0 / 1	1.27	0.294 below average	good day
NYM	A.J. Ewing (CF)	5 / 4	3 / 1	1 / 0	0 / 0	0 / 0	3 / 0	0 / 0	0 / 0	0 / 1	1.25	0.322 above average	projection too low
MIA	Kyle Stowers (DH)	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 2	-1.11	0.336 above average	bad day
NYM	Bo Bichette (3B)	5 / 5	2 / 1	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	1 / 0	1 / 1	0.8	0.326 above average	projection too low
MIA	Griffin Conine (1B)	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	-0.8	0.37 elite	bad day
NYM	Marcus Semien (2B)	5 / 4	2 / 1	2 / 0	0 / 0	0 / 0	1 / 0	2 / 0	0 / 1	0 / 1	0.76	0.306 average	no signal
MIA	Joe Mack (C)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.29 below average	—
NYM	Juan Soto (DH)	5 / 4	2 / 1	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	1 / 1	1 / 1	-0.46	0.416 elite	—
NYM	Carson Benge (RF)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	1 / 0	2 / 1	0.33	0.336 above average	—
NYM	Brett Baty (1B)	4 / 3	1 / 1	1 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	1 / 1	0.33	0.308 average	—
NYM	Francisco Alvarez (C)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	1 / 0	2 / 1	0.33	0.326 above average	—
MIA	Otto Lopez (SS)	4 / 5	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.31	0.324 above average	—
NYM	Francisco Lindor (SS)	6 / 5	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	3 / 1	0.14	0.338 above average	—
MIA	Jakob Marsee (CF)	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	0.14	0.307 average	—
MIA	Owen Caissie (RF)	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	2 / 1	-0.14	0.294 below average	—

Deep slate report — 20260907

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 (11)

ARI@KCMISS

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

ATL@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

CHC@MILHIT

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@LADHIT

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@BALHIT

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@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

MIN@DETMISS

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@MIAMISS

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

STL@SFMISS

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

TOR@OAKMISS

Player-level breakdown (projected vs actual)

Critical: `unexplained_probability_publication_delta`

High: `single_slate_calibration_instability`

High: `weak_run_model_signal`

Info: `complete_slate_audit_coverage`

Info: `simulation_inert_on_slate`

`ARI@KC` — miss; wrong winner and wrong score

`ATL@PHI` — correct winner, wrong score

`CHC@MIL` — correct winner, poor score

`CIN@LAD` — correct winner, wrong score

`CLE@BAL` — correct winner, wrong score

`LAA@BOS` — strong correct winner, close score

`MIN@DET` — miss; wrong winner and wrong score

Signal	Home-away edge	Meaning
Season quality	0.291	positive favors home; negative favors away
Recent form	-0.042	positive favors home; negative favors away
Starting-pitcher matchup	0.039	positive favors home; negative favors away
Composite	0.414	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	3.5-4.81	n/a	Base ML historical model prediction
variance_applied	3-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.521	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run)
embedded_savant_run_environment	4-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	n/a	Capped at max 12 realistic runs
simulation_result	4-5	n/a	Final inning-by-inning simulation result

Stage	Before	After	Toward final pick	What changed
variance_applied	3.5-4.81	3-5	0.686	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-5	4-5	-1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run)
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
NYM	Juan Soto	batter / DH	better	predicted_RBI 3.378
MIA	Griffin Conine	batter / 1B	worse	predicted_AB 1.958
NYM	Nick Morabito	batter / LF	better	predicted_H 1.737
MIA	Owen Caissie	batter / RF	better	predicted_RBI 1.56
MIA	Joe Mack	batter / C	better	predicted_RBI 1.51
MIA	Kyle Stowers	batter / DH	better	predicted_SO -1.349
MIA	Graham Pauley	batter / 3B	worse	predicted_AB 1.333
NYM	Bo Bichette	batter / 3B	worse	predicted_H -1.21

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	MIA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	MIA
info	CURRENT_PITCHER_DUPLICATION	NYM pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	NYM
info	CURRENT_FINALIZED_MATCH_BOX	MIA model-prediction batting totals differ from final simulation box score (AB: current=32, sim=35; H: current=8, sim=11; HR: current=1, sim=2; BB: current=3, sim=5; SO: current=9, sim=5).	MIA
info	CURRENT_PLAYER_MATCH_BOX	MIA current finalized hitter lines are model predictions and differ from the simulation box score: Kyle Stowers: AB current=4 box=5, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Griffin Conine: AB current=3 box=5, H current=1 box=2, BB current=1 box=0; Javier Sanoja: H current=1 box=0, BB current=0 box=1, SO current=0 box=1; Heriberto Hernández: HR current=1 box=0, RBI current=1 box=0, BB current=0 box=1, SO current=1 box=2; Otto Lopez: H current=1 box=2, 2B current=1 box=0, SO current=1 box=0; Jakob Marsee: AB current=4 box=3, 2B current=0 box=1, R current=1 box=0, SO current=1 box=0; Owen Caissie: H current=1 box=2, HR current=0 box=1, RBI current=1 box=2, R current=0 box=1; Graham Pauley: AB current=3 box=4	MIA
info	CURRENT_FINALIZED_MATCH_BOX	NYM model-prediction batting totals differ from final simulation box score (H: current=8, sim=9; HR: current=1, sim=2; RBI: current=5, sim=4; BB: current=3, sim=1; SO: current=9, sim=2).	NYM
info	CURRENT_PLAYER_MATCH_BOX	NYM current finalized hitter lines are model predictions and differ from the simulation box score: Juan Soto: AB current=4 box=3, H current=1 box=2, HR current=1 box=2, RBI current=1 box=4; Carson Benge: H current=1 box=0, R current=1 box=0, SO current=1 box=0; A.J. Ewing: H current=1 box=2, 2B current=0 box=1, R current=0 box=1; Marcus Semien: H current=1 box=0, BB current=1 box=0, SO current=1 box=0; Brett Baty: AB current=3 box=4, 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Nick Morabito: AB current=3 box=4, H current=0 box=2, R current=0 box=1; Francisco Alvarez: RBI current=1 box=0, SO current=1 box=0; Bo Bichette: AB current=5 box=4, H current=1 box=0, RBI current=1 box=0, R current=1 box=0	NYM
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Team	Starter	Projected	Actual	ER err
TOR	Dylan Cease	2 ER / 18 outs	4 ER / 15 outs	2
OAK	Jacob Lopez	5 ER / 15 outs	4 ER / 18 outs	-1

Team	Pitcher	Outs	ERA	ER	H	HR	BB	SO	Context (as-of)
TOR	Dylan Cease (Starter)	15 / 18	7.2 / 3 (+4.20)	4 / 2	4 / 3	1 / 0	2 / 1	6 / 7	venue away: 1.54 vs 2.91 ERA (ok) · opp run factor 0.92 (weak) · 26 prior appearances this season
TOR	Louis Varland (Closer)	2 / 3	13.5 / 0 (+13.50)	1 / 0	2 / 0	0 / 0	2 / 0	0 / 0
TOR	Tyler Rogers (Setup Man)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0
OAK	Jacob Lopez (Starter)	18 / 15	6 / 9 (-3.00)	4 / 5	7 / 6	2 / 1	1 / 2	1 / 8	venue home: 5.11 vs 4.02 ERA (worse here) · opp run factor 1.1 (weak) · 23 prior appearances this season
OAK	Luis Medina (Middle Reliever (High Leverage))	5 / 4	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	2 / 0
OAK	Hogan Harris (Setup Man)	4 / 5	6.75 / 5.4	1 / 1	1 / 1	1 / 1	1 / 0	1 / 2

Team	Batter	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
TOR	Andrés Giménez (SS)	4 / 3	2 / 1	1 / 0	0 / 0	1 / 0	2 / 0	1 / 0	0 / 0	0 / 1	1.71	0.269 weak	good day
OAK	Henry Bolte (CF)	2 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	3 / 1	3 / 1	0 / 1	1.59	0.311 average	no signal
TOR	Alejandro Kirk (C)	4 / 5	2 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	0 / 0	0 / 1	1.4	0.302 below average	good day
TOR	George Springer (DH)	5 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	-1.11	0.306 average	no signal
OAK	Lawrence Butler (RF)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	3 / 1	-0.8	0.322 above average	bad day
OAK	Alika Williams (SS)	4 / 3	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.78	0.243 weak	projection too high
OAK	Jonah Heim (C)	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.78	0.292 below average	projection too high
TOR	Kazuma Okamoto (3B)	2 / 4	1 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	2 / 0	0 / 1	0.66	0.317 average	no signal
OAK	Zack Gelof (3B)	4 / 4	1 / 1	1 / 0	0 / 0	0 / 1	2 / 1	2 / 1	0 / 0	0 / 1	-0.62	0.313 average	no signal
TOR	Brett Bateman (CF)	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	-0.47	0.3 below average	—
TOR	Myles Straw (LF)	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	-0.47	0.3 below average	—
TOR	Nathan Lukes (RF)	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	0 / 1	0.47	0.289 below average	—
OAK	Carlos Cortes (LF)	2 / 3	2 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0 / 1	0.47	0.304 below average	—
TOR	Ernie Clement (2B)	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	-0.31	0.265 weak	—
TOR	Vladimir Guerrero Jr. (1B)	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 0	0	0.332 above average	—
OAK	Jeff McNeil (1B)	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 0	0 / 0	0	0.306 average	—
OAK	Max Muncy (DH)	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	3 / 1	0	0.291 below average	—
OAK	Donovan Walton (2B)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	0 / 1	0	0.333 above average	—

Signal	Home-away edge	Meaning
Season quality	-0.973	positive favors home; negative favors away
Recent form	-0.906	positive favors home; negative favors away
Starting-pitcher matchup	-0.173	positive favors home; negative favors away
Composite	-1.863	positive favors home; negative favors away

Stage	Away-home runs	Confidence	What changed
historical_ml	5.9-3.64	n/a	Base ML historical model prediction
variance_applied	6-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-3	0.7	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	6-2	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-2	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-2	n/a	Capped at max 12 realistic runs
simulation_result	6-2	n/a	Final inning-by-inning simulation result