Deep slate report 20260923

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_20260923.json	404001
forensic deep audit	loaded	probability, provenance, starter and decision contracts	/app/output/reports/PREDICTION_SYSTEM_DEEP_AUDIT_20260923.json	256114
continuous learning	loaded	pipeline-stage quality, calibration, probability lineage, score distribution, and official starter comparisons	/app/output/learning/continuous_learning_daily_20260923.json	2651580
detailed report	loaded	full game and simulation rendering	/app/output/reports/detailed_20260923.html	382569
ultra detailed player audit	loaded	player-level prediction trace	/app/output/reports/20260923_ultra_detailed_player_audit.html	35142661
structural audit	loaded	all invariant findings	/app/output/reports/audit_report_20260923.html	2035176
integrity revision audit	missing	post-Luna remediation proof for immutable revision artifacts	/app/output/reports/audit_report_20260923_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 (16)

Severity	Domain	Finding and evidence	Affected	Action & guardrail
CRITICAL	probability_lineage	Published probability differs materially without a recorded transformation 1 game(s) have an unexplained ≥2pp calibrated-to-published delta	NYM@TEX	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	artifact_integrity	Canonical artifact warning requires reconciliation structural audit totals: {'critical': 0, 'error': 0, 'warning': 2, 'info': 119}	slate-wide	Do thisFix the flagged stat or presentation contract before treating this slate as clean evidence. Don't concludeA warning flags a display or contract issue; it does not prove the model picked the wrong side.
HIGH	model_validation	Validated run models explain very little holdout variance home/away/total R² = 0.0040/0.0349/0.0031	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@SEA, LAA@OAK, MIA@CHC, MIL@PHI, MIN@SF, NYM@TEX, SD@LAD, STL@PIT, TB@NYY, TOR@BAL, TOR@BAL, WAS@DET	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 16/16 games structurally audited	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@SEA, LAA@OAK, MIA@CHC, MIL@PHI, MIN@SF, NYM@TEX, SD@LAD, STL@PIT, TB@NYY, TOR@BAL, TOR@BAL, WAS@DET	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@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@SEA, LAA@OAK, MIA@CHC, MIL@PHI, MIN@SF, NYM@TEX, SD@LAD, STL@PIT, TB@NYY, TOR@BAL, TOR@BAL, WAS@DET	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	5/11	no	code/diagnostic
simulation_inert_on_slate	11	11/11	yes	code/diagnostic
single_slate_calibration_instability	0	7/11	no	param
weak_run_model_signal	11	11/11	yes	code/diagnostic

Metric	Direction	Series (old → new)
winner_accuracy	regressing	[0.6, 0.7, 0.4, 0.6667, 0.4444, 0.6667, 0.7333, 0.6, 0.3333, 0.4, 0.5]
weighted_calibration_error	improving	[0.0827, 0.2812, 0.2795, 0.1661, 0.1016, 0.1247, 0.3458, 0.3184, 0.2028, 0.1881, 0.1108]
margin_bias	improving	[-1.8, -0.4, -1.667, -1.133, -2.111, -2.0, -0.667, -1.533, -2.667, -1.867, -1.188]
team_run_error_on_losses	regressing	[5.0, 6.333, 5.556, 6.4, 3.6, 5.2, 6.5, 6.5, 3.0, 6.222, 5.75]

Stage	Winner hits	Accuracy	Team-run MAE	Total-run MAE	Margin MAE
historical_ml	8/16	50%	2.053	3.025	2.729
variance_applied	6/16	37.5%	2.062	3.125	2.875
enhanced_prediction	8/16	50%	2.219	3.688	2.438
embedded_savant_run_environment	8/16	50%	2.281	3.812	2.438
fundamental_edge_target_review	8/16	50%	2.281	3.812	2.438
capped_target	8/16	50%	2.281	3.812	2.438
simulation_result	8/16	50%	2.281	3.812	2.438
published_final	8/16	50%	2.281	3.812	2.438

Transition	Δ accuracy	Δ hits	Δ team-run MAE	Δ margin MAE	Games whose winner/tie changed
historical_ml → variance_applied	-0.125	-2	0.0095	0.1457	ARI@COL: ARI → TIE; HOU@SEA: SEA → TIE; MIA@CHC: CHC → TIE
variance_applied → enhanced_prediction	0.125	2	0.1563	-0.4375	ARI@COL: TIE → ARI; CLE@BOS: BOS → CLE; HOU@SEA: TIE → SEA; MIA@CHC: TIE → CHC; TB@NYY: NYY → TB; TOR@BAL: TOR → BAL; WAS@DET: DET → WAS
enhanced_prediction → embedded_savant_run_environment	0	0	0.0624	0
embedded_savant_run_environment → fundamental_edge_target_review	0	0	0	0
fundamental_edge_target_review → capped_target	0	0	0	0
capped_target → simulation_result	0	0	0	0
simulation_result → published_final	0	0	0	0

Game	Outcome	Raw model	Calibrated	Governor	Score coherence	Published	Explanation
ARI@COL	hit	0.5896	0.6329	0.6329	0.6329	0.56	score coherence followed by matchup-recency guardrail
CHW@KC	miss	0.6938	0.6519	0.6519	0.613	0.56	score coherence followed by matchup-recency guardrail
CIN@ATL	hit	0.7023	0.6488	0.6488	0.6114	0.56	score coherence followed by matchup-recency guardrail
CLE@BOS	miss	0.5	0.5073	0.5073	0.5073	0.5053	published probability matches the calibrated/governed value
HOU@SEA	hit	0.5518	0.5314	0.5314	0.5314	0.5153	score coherence followed by matchup-recency guardrail
LAA@OAK	miss	0.5613	0.5821	0.5821	0.5781	0.54	score coherence followed by matchup-recency guardrail
MIA@CHC	miss	0.5725	0.5686	0.5686	0.5686	0.53	score coherence followed by matchup-recency guardrail
MIL@PHI	hit	0.6273	0.6418	0.6418	0.6418	0.56	score coherence followed by matchup-recency guardrail
MIN@SF	hit	0.544	0.5271	0.5271	0.5271	0.5111	score coherence followed by matchup-recency guardrail
NYM@TEX	hit	0.6402	0.6218	0.6218	0.6218	0.5954	no recorded transformation explains the publication delta
SD@LAD	miss	0.5606	0.5414	0.5414	0.5414	0.5352	score coherence followed by matchup-recency guardrail
STL@PIT	miss	0.6152	0.581	0.581	0.5775	0.556	score coherence followed by matchup-recency guardrail
TB@NYY	miss	0.5	0.5073	0.5073	0.5073	0.5053	published probability matches the calibrated/governed value
TOR@BAL	hit	0.5059	0.5073	0.5073	0.5073	0.505	published probability matches the calibrated/governed value
TOR@BAL	miss	0.5415	0.5258	0.5258	0.5258	0.5117	score coherence followed by matchup-recency guardrail
WAS@DET	hit	0.5	0.5073	0.5073	0.5073	0.505	published probability matches the calibrated/governed value

Sides analyzed	Broke on pitching	Broke on offense	Both	Variance	Mean starter ERA error	Mean lineup runs vs proj
32	4	11	1	16	-0.49	-0.91

Game	Trained p(home)	Blended p(home)	Blend favoured	Published p(home)	Published side	Artifact says winner changed	Outcome
CLE@BOS	0.677	0.587	home	0.495	away	yes	miss
TB@NYY	0.71	0.606	home	0.495	away	yes	miss
WAS@DET	0.616	0.555	home	0.495	away	yes	hit

Slate	Games	Winner acc	Mean starter ERA error	Mean lineup runs vs proj	Sides broke on pitching	Sides broke on offense
20260922	15	0.400	-0.32	-0.97	3	11
20260921	3	0.333	-1.08	-0.50
20260920	15	0.600	-0.89	-1.00	2	10
20260919	15	0.733	1.36	-0.10	8	7
20260918	15	0.667	3.95	0.30	6	5
20260917	9	0.444	2.16	-0.22	2	3
20260916	15	0.667	2.09	-0.23	4	5
20260915	15	0.400	-1.58	-0.70	1	12
20260914	10	0.700	-0.48	-0.20	2	6
20260913	15	0.600	-0.07	0.03	2	8
20260912	15	0.533	1.35	0.57	6	7
20260911	15	0.667	0.99	0.00	4	9

Game	Pred margin	Actual margin	Pred − actual
ARI@COL	2	2	0
CHW@KC	2	1	1
CIN@ATL	1	1	0
CLE@BOS	1	1	0
HOU@SEA	1	1	0
LAA@OAK	1	4	-3
MIA@CHC	1	1	0
MIL@PHI	2	3	-1
MIN@SF	1	2	-1
NYM@TEX	3	5	-2
SD@LAD	2	4	-2
STL@PIT	1	4	-3
TB@NYY	1	7	-6
TOR@BAL	1	2	-1
TOR@BAL	1	2	-1
WAS@DET	2	2	0

ARI@COLHIT

Predicted: ARI 8-6 (conf -0.12, moderate)Actual: ARI 5-3Brier 0.194total bias +6

Root cause: Right winner, wrong score

Evidence-backed diagnosis

published system edge 0.12 (moderate); this is not the half-scale distance from 50%
team-run absolute error 6; total bias +6
pregame home-minus-away evidence: season -1.174, recent -0.762, pitcher -0.070, composite -1.350
confidence pipeline reached 0.633 before later publication/recency controls; published home probability was 0.440
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 6 (14 vs 8)
ARI offense over-projected by 3
COL offense over-projected by 3
Merrill Kelly (ARI) allowed 3 fewer ER than projected (2 vs 5)
Mason Adams (COL) allowed 5 fewer ER than projected (0 vs 5)

Team	Starter	Projected	Actual	ER err
ARI	Merrill Kelly	5 ER / 16 outs	2 ER / 16 outs	-3
COL	Mason Adams	5 ER / 13 outs	0 ER / 18 outs	-5

Decision internals: run factors home×0.951 / away×1.021; 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:

Ezequiel Tovar (COL SS): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Cole Carrigg (COL CF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Tim Tawa (ARI CF): 1-for-3, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]
Nolan Arenado (ARI 3B): 1-for-4, 0 HR, 1 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Lars Nootbaar (ARI LF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Starter miss: Mason Adams (COL) posted 0.00 ERA vs projected 10.38 (0 ER in 18 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

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

Pitcher versus batter: what actually broke this game

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

AWAY pitching

No settled pitching line for this club. The forecast named Merrill Kelly (ARI) at a projected 8.44 ERA, but the per-player feed carries no pitching row for him in this game, so the actual-versus-projected comparison cannot be made on this side. That is a source-coverage gap in the settled box, not a forecasting failure.

Bullpen behind him: 0 ER over 0 outs.

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

Who did the damage to him:

Ezequiel Tovar: 2-for-4, 1 HR, 1 RBI [+1.4 runs vs projection] season xwOBA 0.279, weak - a good day, not a model error
Adael Amador: 1-for-4, 0 HR, 1 RBI [+0.31 runs vs projection] season xwOBA 0.263, weak - a good day, not a model error

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

HOME pitching

Mason Adams (COL) threw 18 outs, 0 ER = 0 ERA against a projected 10.38 ERA (-10.38 ERA error) only 5 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 0 ER over 3 outs.

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

Who did the damage to him:

Tim Tawa: 1-for-3, 1 HR, 1 RBI [+0.93 runs vs projection] season xwOBA 0.31, average - a good day, not a model error
Corbin Carroll: 1-for-4, 0 HR, 2 RBI [+0.62 runs vs projection] season xwOBA 0.345, elite - real projection 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. exp BB and exp SO are the continuous rate expectations. The integer beside each is a simulation draw reconciled to the opposing simulated batting box, so it is not the line that ranks pitchers: on the 2026-09-07 slate the integers published 39 walks and 121 strikeouts against 51 and 130 real, while these expectations landed on 50.4 and 132.6.)

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
COL	Mason Adams (Starter)	18 / 13	0 / 10.38 (-10.38)	0 / 5	5 / 6	0 / 2	0 / 4	2.4	8 / 4	4.75	venue home: 9 vs 2.57 ERA (ok) · opp run factor 1.02 (average) · 5 prior appearances this season
COL	Jaden Hill (Middle Reliever (High Leverage))	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.58	1 / 1	0.96

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Ezequiel Tovar (SS)	COL	4 / 3	2 / 1	0 / 0	0 / 0	1 / 0	1 / 0	2 / 0	0 / 0	0 / 1	1.4	0.279 weak	good day
Cole Carrigg (CF)	COL	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-1.11	0.311 average	no signal
Tim Tawa (CF)	ARI	3 / 3	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	0 / 0	1 / 1	0.93	0.31 average	no signal
Nolan Arenado (3B)	ARI	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	0 / 0	2 / 1	-0.93	0.296 below average	projection too high
Lars Nootbaar (LF)	ARI	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	-0.8	0.341 elite	bad day
Geraldo Perdomo (SS)	ARI	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 0	-0.8	0.328 above average	bad day
TJ Rumfield (1B)	COL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	-0.8	0.334 above average	bad day
Corbin Carroll (RF)	ARI	4 / 4	1 / 1	0 / 0	1 / 0	0 / 0	2 / 1	0 / 1	0 / 0	0 / 1	0.62	0.345 elite	projection too low
Mickey Moniak (RF)	COL	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	-0.47	0.277 weak	—
Zac Veen (LF)	COL	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.274 weak	—
Hunter Goodman (C)	COL	3 / 4	2 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 0	0 / 1	-0.46	0.319 average	—
Ketel Marte (2B)	ARI	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	0 / 0	-0.31	0.343 elite	—
Jake McCarthy (DH)	COL	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.31	0.324 above average	—
Adael Amador (2B)	COL	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 0	0 / 0	0.31	0.263 weak	—
Gabriel Moreno (C)	ARI	4 / 4	3 / 1	0 / 1	0 / 0	0 / 0	1 / 1	1 / 1	0 / 1	0 / 0	0.3	0.362 elite	—
Pavin Smith (1B)	ARI	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0	0.274 weak	—
Kyle Karros (3B)	COL	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	1 / 1	0	0.338 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-1.174	positive favors home; negative favors away
Recent form	-0.762	positive favors home; negative favors away
Starting-pitcher matchup	-0.07	positive favors home; negative favors away
Composite	-1.35	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.41-5.12	n/a	Base ML historical model prediction
variance_applied	5-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	8-6	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	8-6	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	8-6	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	8-6	n/a	Capped at max 12 realistic runs
simulation_result	8-6	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	5.41-5.12	5-5	-0.294	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-5	8-6	2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	8-6	8-6	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	8-6	8-6	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	8-6	8-6	-0	Capped at max 12 realistic runs
simulation_result	8-6	8-6	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
COL	Jake McCarthy	batter / DH	better	predicted_H 2.853
ARI	Jesús Sánchez	batter / DH	better	predicted_RBI 2.84
ARI	Ketel Marte	batter / 2B	better	predicted_RBI 2.52
COL	Mason Adams	pitcher / Starter	worse	predicted_H 2
COL	Brennan Bernardino	pitcher / Setup Man	better	predicted_H -2
COL	Adael Amador	batter / 2B	better	predicted_AB 1.81
COL	Zac Veen	batter / LF	worse	predicted_H 1.59
ARI	Geraldo Perdomo	batter / SS	worse	predicted_SO 1.457

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	ARI pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	ARI
info	CURRENT_PITCHER_DUPLICATION	COL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	COL
info	CURRENT_FINALIZED_MATCH_BOX	ARI model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=9, sim=8; 2B: current=2, sim=1; HR: current=1, sim=3; RBI: current=5, sim=8; R: current=4, sim=8; BB: current=3, sim=5; SO: current=6, sim=8).	ARI
info	CURRENT_PLAYER_MATCH_BOX	ARI current finalized hitter lines are model predictions and differ from the simulation box score: Lars Nootbaar: 2B current=0 box=1, RBI current=0 box=1, R current=0 box=1, SO current=1 box=2; Gabriel Moreno: AB current=4 box=5, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Pavin Smith: H current=1 box=0, R current=0 box=1, BB current=0 box=1; Geraldo Perdomo: AB current=4 box=5, H current=1 box=0, BB current=1 box=0, SO current=0 box=2; Ketel Marte: H current=1 box=2, 2B current=1 box=0, HR current=0 box=2, RBI current=1 box=3; Tim Tawa: RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Nolan Arenado: HR current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Corbin Carroll: H current=1 box=2	ARI
info	CURRENT_FINALIZED_MATCH_BOX	COL model pitcher projections differ from final simulation pitching box score (ER: current=7, sim=8).	COL
info	CURRENT_FINALIZED_MATCH_BOX	COL model-prediction batting totals differ from final simulation box score (AB: current=34, sim=39; H: current=9, sim=13; RBI: current=5, sim=6; R: current=4, sim=6; SO: current=7, sim=5; SF: current=0, sim=1).	COL
info	CURRENT_PLAYER_MATCH_BOX	COL current finalized hitter lines are model predictions and differ from the simulation box score: Jake McCarthy: AB current=4 box=5, H current=1 box=4, RBI current=1 box=2, SO current=1 box=0; Cole Carrigg: 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Adael Amador: AB current=4 box=5, HR current=0 box=1, RBI current=1 box=2, SO current=0 box=1; Mickey Moniak: RBI current=1 box=0, BB current=0 box=1; Hunter Goodman: H current=1 box=0, HR current=1 box=0, SF current=0 box=1; Kyle Karros: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, R current=0 box=1; TJ Rumfield: H current=1 box=0, SO current=0 box=1; Zac Veen: AB current=3 box=4, H current=1 box=2, RBI current=0 box=1, R current=0 box=1	COL
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

CHW@KCMISS

Predicted: CHW 7-5 (conf -0.12, moderate)Actual: KC 4-5Brier 0.314total bias +3

Root cause: Backed the wrong side with real conviction

Evidence-backed diagnosis

published system edge 0.12 (moderate); this is not the half-scale distance from 50%
team-run absolute error 3; total bias +3
pregame home-minus-away evidence: season -0.991, recent -1.018, pitcher -0.039, composite -1.245
confidence pipeline reached 0.652 before later publication/recency controls; published home probability was 0.440
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 3 (12 vs 9)
CHW offense over-projected by 3
committed to CHW at 0.12 confidence and it lost

Team	Starter	Projected	Actual	ER err
CHW	Bryan Hudson	2 ER / 15 outs	3 ER / 1 outs	1
KC	Seth Lugo	3 ER / 14 outs	4 ER / 10 outs	1

Decision internals: run factors home×0.988 / away×1.027; 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:

Maikel Garcia (KC 3B): 3-for-3, 0 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 1.27]
Braden Montgomery (CHW RF): 3-for-4, 0 HR, 0 RBI (projected 0.0 H, 0.0 HR) [run impact 1.08]
Munetaka Murakami (CHW 1B): 1-for-4, 0 HR, 1 RBI (projected 1.0 H, 1.0 HR) [run impact -0.95]
Tristan Peters (CHW CF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]
Bobby Witt Jr. (KC SS): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.64]

Starter miss: Seth Lugo (KC) posted 10.80 ERA vs projected 5.79 (4 ER in 10 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

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

Pitcher versus batter: what actually broke this game

AWAY pitching

No settled pitching line for this club. The forecast named Bryan Hudson (CHW) at a projected 3.6 ERA, but the per-player feed carries no pitching row for him in this game, so the actual-versus-projected comparison cannot be made on this side. That is a source-coverage gap in the settled box, not a forecasting failure.

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 5 runs on 12 hits and 0 HR, against a projected 4 runs / 9 hits (1 runs vs projection).

Who did the damage to him:

Maikel Garcia: 3-for-3, 0 HR, 3 RBI [+1.27 runs vs projection] season xwOBA 0.313, average - a good day, not a model error
Michael Massey: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.305, average - a good day, not a model error
Vinnie Pasquantino: 1-for-4, 0 HR, 1 RBI [+0.31 runs vs projection] season xwOBA 0.336, above average - a good day, not a model error

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

HOME pitching

Seth Lugo (KC) threw 10 outs, 4 ER = 10.8 ERA against a projected 5.79 ERA (5.01 ERA error)

Bullpen behind him: 0 ER over 9 outs.

The AWAY lineup he faced produced 4 runs on 9 hits and 0 HR, against a projected 5 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

Braden Montgomery: 3-for-4, 0 HR, 0 RBI [+1.08 runs vs projection] season xwOBA 0.325, above average - real projection error
Chase Meidroth: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.299, below average - a good day, not a model error

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

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
KC	Seth Lugo (Starter)	10 / 14	10.8 / 5.79 (+5.01)	4 / 3	7 / 4	0 / 0	0 / 6	1.63	3 / 6	3.8	venue home: 4.09 vs 6.1 ERA (ok) · opp run factor 1.03 (average) · 31 prior appearances this season
KC	Connor Thomas (Middle Reliever (High Leverage))	6 / 3	0 / 27 (-27.00)	0 / 3	1 / 4	0 / 0	0 / 0	0.46	1 / 1	0.83
KC	Steven Cruz (Closer)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.31	2 / 0	1.01

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Maikel Garcia (3B)	KC	3 / 4	3 / 1	1 / 1	0 / 0	0 / 0	3 / 0	1 / 0	1 / 0	0 / 0	1.27	0.313 average	no signal
Braden Montgomery (RF)	CHW	4 / 3	3 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 1	1.08	0.325 above average	projection too low
Munetaka Murakami (1B)	CHW	4 / 4	1 / 1	1 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	1 / 1	-0.95	0.344 elite	bad day
Tristan Peters (CF)	CHW	3 / 3	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	-0.78	0.315 average	no signal
Bobby Witt Jr. (SS)	KC	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	2 / 1	0 / 1	1 / 1	-0.64	0.379 elite	bad day
Andrew Benintendi (DH)	CHW	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	-0.47	0.307 average	—
Chase Meidroth (2B)	CHW	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	1 / 1	0.47	0.299 below average	—
Colson Montgomery (SS)	CHW	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	3 / 1	-0.47	0.301 below average	—
John Rave (CF)	KC	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.311 average	—
Michael Massey (2B)	KC	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.47	0.305 average	—
Isaac Collins (LF)	KC	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	2 / 1	-0.47	0.294 below average	—
Miguel Vargas (3B)	CHW	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	0 / 1	-0.33	0.38 elite	—
Carter Jensen (C)	KC	4 / 4	3 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	1 / 1	-0.32	0.31 average	—
Sam Antonacci (LF)	CHW	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	-0.31	0.341 elite	—
Vinnie Pasquantino (1B)	KC	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 1	1 / 1	0 / 0	0 / 0	0.31	0.336 above average	—
Kyle Teel (C)	CHW	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	2 / 1	1 / 0	1 / 1	2 / 1	0	0.303 below average	—
Jac Caglianone (RF)	KC	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	0	0.357 elite	—
Salvador Perez (DH)	KC	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	0	0.31 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.991	positive favors home; negative favors away
Recent form	-1.018	positive favors home; negative favors away
Starting-pitcher matchup	-0.039	positive favors home; negative favors away
Composite	-1.245	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.33-3.24	n/a	Base ML historical model prediction
variance_applied	5-3	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-5	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	7-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	7-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	7-5	n/a	Capped at max 12 realistic runs
simulation_result	7-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	5.33-3.24	5-3	-0.092	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	6-5	-1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	6-5	7-5	1	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
CHW	Andrew Benintendi	batter / DH	better	predicted_RBI 3.65
CHW	Tristan Peters	batter / CF	better	predicted_RBI 2.78
KC	John Rave	batter / CF	worse	predicted_BB 2.674
CHW	Sam Antonacci	batter / LF	better	predicted_AB 1.567
KC	Michael Massey	batter / 2B	worse	predicted_RBI 1.49
CHW	Colson Montgomery	batter / SS	worse	predicted_SO -1.48
KC	Jac Caglianone	batter / RF	worse	predicted_RBI 1.45
CHW	Kyle Teel	batter / C	better	predicted_H 1.41

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	CHW pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CHW
info	CURRENT_PITCHER_DUPLICATION	KC pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	KC
info	CURRENT_FINALIZED_MATCH_BOX	CHW model-prediction batting totals differ from final simulation box score (AB: current=31, sim=36; H: current=8, sim=9; 2B: current=2, sim=3; HR: current=1, sim=0; RBI: current=4, sim=7; R: current=5, sim=7; BB: current=4, sim=6).	CHW
info	CURRENT_PLAYER_MATCH_BOX	CHW current finalized hitter lines are model predictions and differ from the simulation box score: Sam Antonacci: AB current=4 box=5, H current=1 box=2, 2B current=1 box=0; Munetaka Murakami: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Kyle Teel: AB current=4 box=5, H current=1 box=2, RBI current=1 box=0, R current=0 box=1; Miguel Vargas: 2B current=0 box=1, RBI current=1 box=0, SO current=1 box=0; Andrew Benintendi: AB current=3 box=4, H current=1 box=3, 2B current=0 box=1, RBI current=1 box=4; Braden Montgomery: AB current=3 box=4, BB current=1 box=0, SO current=1 box=2; Tristan Peters: AB current=3 box=4, RBI current=0 box=3, R current=0 box=1, BB current=0 box=1; Chase Meidroth: H current=1 box=0, BB current=0 box=1, SO current=1 box=2	CHW
info	CURRENT_FINALIZED_MATCH_BOX	KC model pitcher projections differ from final simulation pitching box score (ER: current=6, sim=7).	KC
info	CURRENT_FINALIZED_MATCH_BOX	KC model-prediction batting totals differ from final simulation box score (H: current=9, sim=7; 2B: current=2, sim=1; HR: current=1, sim=0; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=7; SO: current=7, sim=8).	KC
info	CURRENT_PLAYER_MATCH_BOX	KC current finalized hitter lines are model predictions and differ from the simulation box score: Bobby Witt Jr.: AB current=4 box=5, H current=1 box=0, 2B current=1 box=0, R current=1 box=0; Maikel Garcia: 2B current=1 box=0, R current=0 box=1, BB current=0 box=1; Carter Jensen: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Jac Caglianone: RBI current=1 box=2, BB current=0 box=1, SO current=1 box=0; John Rave: AB current=3 box=1, H current=1 box=0, BB current=0 box=3; Michael Massey: AB current=4 box=3, 2B current=0 box=1, RBI current=0 box=2, R current=0 box=1; Vinnie Pasquantino: RBI current=1 box=0, BB current=0 box=1; Salvador Perez: H current=1 box=2, R current=0 box=1, SO current=1 box=0	KC
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

CIN@ATLHIT

Predicted: ATL 5-6 (conf +0.12, moderate)Actual: ATL 2-3Brier 0.194total bias +6

Root cause: Right winner, wrong score

Evidence-backed diagnosis

published system edge 0.12 (moderate); this is not the half-scale distance from 50%
team-run absolute error 6; total bias +6
pregame home-minus-away evidence: season +1.446, recent +0.540, pitcher +0.072, composite +1.269
confidence pipeline reached 0.649 before later publication/recency controls; published home probability was 0.560
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 6 (11 vs 5)
CIN offense over-projected by 3
ATL offense over-projected by 3

Team	Starter	Projected	Actual	ER err
CIN	Andrew Abbott	1 ER / 13 outs	2 ER / 21 outs	1
ATL	Chris Sale	2 ER / 19 outs	1 ER / 17 outs	-1

Decision internals: run factors home×1.081 / away×1.009; 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:

Sal Stewart (CIN 1B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Austin Riley (ATL 3B): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Juan Brito (CIN RF): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]
Matt McLain (CIN 2B): 2-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]
Elly De La Cruz (CIN SS): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Pitcher versus batter: what actually broke this game

AWAY pitching

Andrew Abbott (CIN) threw 21 outs, 2 ER = 2.57 ERA against a projected 2.08 ERA (0.49 ERA error)

Bullpen behind him: 0 ER over 1 outs.

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

Who did the damage to him:

Austin Riley: 2-for-4, 1 HR, 1 RBI [+1.4 runs vs projection] season xwOBA 0.298, below average - a good day, not a model error
Mauricio Dubón: 2-for-4, 0 HR, 1 RBI [+0.8 runs vs projection] season xwOBA 0.304, average - a good day, not a model error
Matt Olson: 2-for-4, 1 HR, 1 RBI [+0.45 runs vs projection] season xwOBA 0.351, elite - a good day, not a model error

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

HOME pitching

Chris Sale (ATL) threw 17 outs, 1 ER = 1.59 ERA against a projected 2.84 ERA (-1.25 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The AWAY lineup he faced produced 2 runs on 5 hits and 1 HR, against a projected 3 runs / 7 hits (-1 runs vs projection).

Who did the damage to him:

Juan Brito: 2-for-4, 1 HR, 1 RBI [+1.09 runs vs projection] season xwOBA 0.284, below average - a good day, not a model error
Matt McLain: 2-for-4, 0 HR, 0 RBI [+1.09 runs vs projection] season xwOBA 0.297, below average - a good day, not a model error
Oswaldo Cabrera: 1-for-4, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.12, weak - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
CIN	Andrew Abbott (Starter)	21 / 13	2.57 / 2.08	2 / 1	6 / 5	2 / 1	1 / 1	2.42	6 / 8	4.19	venue away: 4.91 vs 4.48 ERA (ok) · opp run factor 1.08 (strong) · 31 prior appearances this season
CIN	Luis Mey (Setup Man)	1 / 3	0 / 9 (-9.00)	0 / 1	1 / 2	0 / 0	0 / 0	0.68	0 / 0	1.59
ATL	Chris Sale (Starter)	17 / 19	1.59 / 2.84	1 / 2	4 / 3	1 / 0	0 / 2	0.71	10 / 5	7.77	venue home: 1.66 vs 2.74 ERA (ok) · opp run factor 1.01 (weak) · 26 prior appearances this season
ATL	Dylan Lee (Setup Man)	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 1	0 / 0	0 / 0	0.25	0 / 0	1.08
ATL	Raisel Iglesias (Closer)	3 / 3	0 / 18 (-18.00)	0 / 2	0 / 2	0 / 0	0 / 0	0.24	3 / 0	1.14

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Sal Stewart (1B)	CIN	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	3 / 1	-1.4	0.348 elite	bad day
Austin Riley (3B)	ATL	4 / 4	2 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	2 / 1	1.4	0.298 below average	good day
Juan Brito (RF)	CIN	4 / 4	2 / 1	0 / 1	0 / 0	1 / 0	1 / 1	1 / 0	0 / 0	0 / 1	1.09	0.284 below average	good day
Matt McLain (2B)	CIN	4 / 3	2 / 1	0 / 0	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	2 / 1	1.09	0.297 below average	good day
Elly De La Cruz (SS)	CIN	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.8	0.378 elite	bad day
Drake Baldwin (C)	ATL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.8	0.367 elite	bad day
Mauricio Dubón (LF)	ATL	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	1 / 0	0 / 0	0.8	0.304 average	no signal
Eugenio Suárez (DH)	CIN	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	3 / 2	-0.47	0.3 below average	—
Dane Myers (CF)	CIN	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	2 / 1	-0.47	0.311 average	—
Sean Murphy (DH)	ATL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.259 weak	—
Michael Harris II (CF)	ATL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.47	0.341 elite	—
Brewer Hicklen (RF)	ATL	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	-0.47	0.312 average	—
Ha-Seong Kim (SS)	ATL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.212 weak	—
Matt Olson (1B)	ATL	4 / 4	2 / 1	1 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	0 / 1	0.45	0.351 elite	—
Oswaldo Cabrera (LF)	CIN	4 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0.31	0.12 weak	—
Ozzie Albies (2B)	ATL	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	-0.31	0.278 weak	—
Will Banfield (C)	CIN	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0	0.198 weak	—
Ke'Bryan Hayes (3B)	CIN	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.293 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	1.446	positive favors home; negative favors away
Recent form	0.54	positive favors home; negative favors away
Starting-pitcher matchup	0.072	positive favors home; negative favors away
Composite	1.269	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.45-6.19	n/a	Base ML historical model prediction
variance_applied	3-6	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-6	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-6	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	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

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.45-6.19	3-6	0.258	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-6	5-6	-2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	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

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
ATL	Drake Baldwin	batter / C	worse	predicted_RBI 3.46
ATL	Chris Sale	pitcher / Starter	better	predicted_H -3
CIN	Dane Myers	batter / CF	boosted	predicted_RBI 2.66
ATL	Sean Murphy	batter / DH	deboosted	predicted_H 2.414
ATL	Raisel Iglesias	pitcher / Closer	worse	predicted_H 2
ATL	Ha-Seong Kim	batter / SS	worse	predicted_H 1.72
CIN	Will Banfield	batter / C	worse	predicted_AB 1.5
CIN	Sal Stewart	batter / 1B	worse	predicted_AB -1.42

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	ATL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	ATL
info	CURRENT_PITCHER_DUPLICATION	CIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CIN
info	CURRENT_FINALIZED_MATCH_BOX	ATL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=34; H: current=9, sim=11; 2B: current=1, sim=2; HR: current=1, sim=2; RBI: current=4, sim=6; R: current=4, sim=6; BB: current=3, sim=2; SO: current=7, sim=8).	ATL
info	CURRENT_PLAYER_MATCH_BOX	ATL current finalized hitter lines are model predictions and differ from the simulation box score: Drake Baldwin: H current=1 box=2, HR current=0 box=2, RBI current=1 box=4, BB current=1 box=0; Ozzie Albies: AB current=4 box=3, 2B current=1 box=0, BB current=0 box=1; Mauricio Dubón: RBI current=1 box=0, R current=0 box=1; Sean Murphy: H current=1 box=3, RBI current=0 box=1, R current=0 box=1, BB current=0 box=1; Matt Olson: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Michael Harris II: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=2; Brewer Hicklen: AB current=3 box=4, 2B current=0 box=1, RBI current=0 box=1, R current=0 box=1; Ha-Seong Kim: AB current=3 box=4, H current=1 box=2, R current=0 box=1	ATL
info	CURRENT_FINALIZED_MATCH_BOX	CIN model-prediction batting totals differ from final simulation box score (AB: current=32, sim=33; H: current=7, sim=6; 2B: current=1, sim=2; HR: current=1, sim=0; RBI: current=4, sim=5; R: current=3, sim=5; SO: current=10, sim=5).	CIN
info	CURRENT_PLAYER_MATCH_BOX	CIN current finalized hitter lines are model predictions and differ from the simulation box score: Juan Brito: 2B current=1 box=0, R current=0 box=1; Elly De La Cruz: 2B current=0 box=1, BB current=1 box=0; Eugenio Suárez: RBI current=1 box=0, SO current=2 box=0; Oswaldo Cabrera: H current=1 box=0, BB current=0 box=1; Sal Stewart: AB current=4 box=3, HR current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Dane Myers: 2B current=0 box=1, RBI current=0 box=3, R current=0 box=1, BB current=1 box=0; Will Banfield: AB current=3 box=4, SO current=1 box=0; Matt McLain: AB current=3 box=4, H current=1 box=0	CIN
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

CLE@BOSMISS

Predicted: CLE 5-4 (conf -0.01, toss_up)Actual: BOS 0-1Brier 0.255total bias +8

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

Evidence-backed diagnosis

published system edge 0.01 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 8; total bias +8
pregame home-minus-away evidence: season +0.058, recent -0.768, pitcher +0.060, composite -0.207
confidence pipeline reached 0.507 before later publication/recency controls; published home probability was 0.495
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 8 (9 vs 1)
CLE offense over-projected by 5
BOS offense over-projected by 3
Foster Griffin (CLE) allowed 3 fewer ER than projected (1 vs 4)
Sonny Gray (BOS) allowed 3 fewer ER than projected (0 vs 3)

Team	Starter	Projected	Actual	ER err
CLE	Foster Griffin	4 ER / 16 outs	1 ER / 18 outs	-3
BOS	Sonny Gray	3 ER / 18 outs	0 ER / 18 outs	-3

Decision internals: run factors home×1.005 / away×0.945; 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:

Chase DeLauter (CLE DH): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Jo Adell (CLE RF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.07]
Jahmai Jones (BOS DH): 2-for-4, 0 HR, 0 RBI (projected 0.0 H, 0.0 HR) [run impact 0.94]
José Ramírez (CLE 3B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]
Adley Rutschman (BOS C): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.64]

Starter miss: Foster Griffin (CLE) posted 1.50 ERA vs projected 6.75 (1 ER in 18 outs); he is materially worse pitching away (3.92 ERA there vs 3.35 on the other side), a venue-role weakness the model under-weighted.

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

Pitcher versus batter: what actually broke this game

AWAY pitching

Foster Griffin (CLE) threw 18 outs, 1 ER = 1.5 ERA against a projected 6.75 ERA (-5.25 ERA error)

Bullpen behind him: 0 ER over 3 outs.

The HOME lineup he faced produced 0 runs on 10 hits and 0 HR, against a projected 3 runs / 8 hits (-3 runs vs projection).

Who did the damage to him:

Jahmai Jones: 2-for-4, 0 HR, 0 RBI [+0.94 runs vs projection] season xwOBA 0.326, above average - real projection error

Verdict: offense: the lineup scored well below its projection

HOME pitching

Sonny Gray (BOS) threw 18 outs, 0 ER = 0 ERA against a projected 4.5 ERA (-4.5 ERA error)

Bullpen behind him: 0 ER over 6 outs.

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

Who did the damage to him:

Petey Halpin: 1-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.289, below average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
CLE	Foster Griffin (Starter)	18 / 16	1.5 / 6.75 (-5.25)	1 / 4	10 / 4	0 / 1	0 / 3	2	7 / 6	4.02	venue away: 3.92 vs 3.35 ERA (worse here) · opp run factor 1 (average) · 30 prior appearances this season
CLE	Tim Herrin (Middle Reliever (High Leverage))	3 / 0	0 / n/a	0 / 0	1 / 1	0 / 0	0 / 0	0.38	3 / 0	0.52
BOS	Sonny Gray (Starter)	18 / 18	0 / 4.5 (-4.50)	0 / 3	4 / 6	0 / 0	1 / 3	1.88	3 / 6	5.61	venue home: 2.91 vs 2.74 ERA (ok) · opp run factor 0.94 (average) · 28 prior appearances this season
BOS	Aroldis Chapman (Closer)	3 / 4	0 / 6.75 (-6.75)	0 / 1	0 / 1	0 / 0	0 / 0	0.46	3 / 0	1.4
BOS	Garrett Whitlock (Middle Reliever (High Leverage))	3 / 2	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0.16	0 / 0	1.24

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Chase DeLauter (DH)	CLE	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-1.11	0.348 elite	bad day
Jo Adell (RF)	CLE	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	1 / 1	-1.07	0.324 above average	bad day
Jahmai Jones (DH)	BOS	4 / 3	2 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0.94	0.326 above average	projection too low
José Ramírez (3B)	CLE	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	-0.78	0.327 above average	bad day
Adley Rutschman (C)	BOS	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.64	0.328 above average	bad day
Nathaniel Lowe (1B)	CLE	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.336 above average	—
Petey Halpin (CF)	CLE	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.47	0.289 below average	—
Trevor Story (SS)	BOS	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.47	0.254 weak	—
Caleb Durbin (3B)	BOS	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	-0.47	0.295 below average	—
Wilyer Abreu (RF)	BOS	4 / 4	2 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-0.46	0.349 elite	—
Roman Anthony (LF)	BOS	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	-0.33	0.341 elite	—
Ceddanne Rafaela (CF)	BOS	1 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	-0.31	0.282 weak	—
Steven Kwan (LF)	CLE	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0.14	0.327 above average	—
Travis Bazzana (2B)	CLE	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	2 / 1	-0.14	0.302 below average	—
Nick Sogard (1B)	BOS	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0.14	0.309 average	—
Patrick Bailey (C)	CLE	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.302 below average	—
Brayan Rocchio (SS)	CLE	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0	0.295 below average	—
Andruw Monasterio (2B)	BOS	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0	0.291 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.058	positive favors home; negative favors away
Recent form	-0.768	positive favors home; negative favors away
Starting-pitcher matchup	0.06	positive favors home; negative favors away
Composite	-0.207	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.91-5	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.505	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	n/a	Capped at max 12 realistic runs
simulation_result	5-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.91-5	4-5	0.089	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	5-4	2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	5-4	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	5-4	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	5-4	-0	Capped at max 12 realistic runs
simulation_result	5-4	5-4	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
BOS	Jahmai Jones	batter / DH	worse	predicted_AB 2.5
BOS	Trevor Story	batter / SS	better	predicted_RBI 2.42
BOS	Sonny Gray	pitcher / Starter	better	predicted_SO 2
CLE	Patrick Bailey	batter / C	worse	predicted_RBI 1.69
CLE	Nathaniel Lowe	batter / 1B	worse	predicted_H 1.62
CLE	José Ramírez	batter / 3B	better	predicted_RBI 1.43
BOS	Adley Rutschman	batter / C	worse	predicted_SO 1.319
CLE	Petey Halpin	batter / CF	deboosted	predicted_AB 1.306

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	BOS pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	BOS
info	CURRENT_PITCHER_DUPLICATION	CLE pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CLE
info	CURRENT_FINALIZED_MATCH_BOX	BOS model-prediction batting totals differ from final simulation box score (AB: current=33, sim=36; H: current=8, sim=6; 2B: current=2, sim=0; R: current=3, sim=4; BB: current=3, sim=4).	BOS
info	CURRENT_PLAYER_MATCH_BOX	BOS current finalized hitter lines are model predictions and differ from the simulation box score: Nick Sogard: H current=1 box=0; Adley Rutschman: 2B current=1 box=0, RBI current=1 box=0, BB current=1 box=0, SO current=1 box=2; Roman Anthony: H current=1 box=0, SO current=1 box=0; Jahmai Jones: AB current=3 box=5, H current=0 box=1, R current=0 box=1, SO current=1 box=2; Wilyer Abreu: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Andruw Monasterio: AB current=3 box=4, H current=1 box=0, SO current=1 box=0; Trevor Story: AB current=4 box=3, H current=1 box=3, HR current=0 box=1, RBI current=1 box=3; Ceddanne Rafaela: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0	BOS
info	CURRENT_FINALIZED_MATCH_BOX	CLE model pitcher projections differ from final simulation pitching box score (H: current=5, sim=6).	CLE
info	CURRENT_FINALIZED_MATCH_BOX	CLE model-prediction batting totals differ from final simulation box score (AB: current=32, sim=38; 2B: current=2, sim=3; HR: current=1, sim=0; RBI: current=4, sim=5).	CLE
info	CURRENT_PLAYER_MATCH_BOX	CLE current finalized hitter lines are model predictions and differ from the simulation box score: Chase DeLauter: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; José Ramírez: AB current=4 box=5, H current=1 box=2, RBI current=1 box=2, BB current=1 box=0; Steven Kwan: RBI current=0 box=1, SO current=0 box=1; Jo Adell: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Travis Bazzana: AB current=3 box=4, H current=1 box=0, BB current=0 box=1, SO current=1 box=2; Nathaniel Lowe: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, R current=0 box=1; Petey Halpin: AB current=3 box=4, H current=0 box=1, 2B current=0 box=1, R current=0 box=1; Patrick Bailey: AB current=3 box=4, RBI current=0 box=2, R current=0 box=1, SO current=1 box=2	CLE
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

HOU@SEAHIT

Predicted: SEA 4-5 (conf +0.03, toss_up)Actual: SEA 5-6Brier 0.235total bias -2

Root cause: Toss-up that went our way

Evidence-backed diagnosis

published system edge 0.03 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 2; total bias -2
pregame home-minus-away evidence: season +0.083, recent +0.526, pitcher -0.000, composite +0.357
confidence pipeline reached 0.531 before later publication/recency controls; published home probability was 0.515
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

Ethan Pecko (HOU) allowed 3 more ER than projected (3 vs 0)

Team	Starter	Projected	Actual	ER err
HOU	Ethan Pecko	0 ER / 15 outs	3 ER / 8 outs	3
SEA	George Kirby	2 ER / 18 outs	4 ER / 15 outs	2

Decision internals: run factors home×0.996 / away×0.996; 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:

Yainer Diaz (HOU C): 2-for-5, 1 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 2.02]
Julio Rodríguez (SEA CF): 3-for-4, 0 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.58]
Nick Allen (HOU SS): 2-for-4, 0 HR, 0 RBI (projected 0.0 H, 0.0 HR) [run impact 1.25]
Taylor Trammell (HOU LF): 0-for-5, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Starter miss: Ethan Pecko (HOU) posted 10.12 ERA vs projected 0.00 (3 ER in 8 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

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

Pitcher versus batter: what actually broke this game

AWAY pitching

Ethan Pecko (HOU) threw 8 outs, 3 ER = 10.12 ERA against a projected 0 ERA (10.12 ERA error) only 6 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 1 ER over 14 outs.

The HOME lineup he faced produced 6 runs on 8 hits and 0 HR, against a projected 4 runs / 7 hits (2 runs vs projection).

Who did the damage to him:

Julio Rodríguez: 3-for-4, 0 HR, 2 RBI [+1.58 runs vs projection] season xwOBA 0.334, above average - real projection error
Josh Naylor: 1-for-4, 0 HR, 1 RBI [+0.33 runs vs projection] season xwOBA 0.311, average - a good day, not a model error

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

HOME pitching

George Kirby (SEA) threw 15 outs, 4 ER = 7.2 ERA against a projected 3 ERA (4.2 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The AWAY lineup he faced produced 5 runs on 10 hits and 1 HR, against a projected 4 runs / 8 hits (1 runs vs projection).

Who did the damage to him:

Yainer Diaz: 2-for-5, 1 HR, 3 RBI [+2.02 runs vs projection] season xwOBA 0.309, average - a good day, not a model error
Nick Allen: 2-for-4, 0 HR, 0 RBI [+1.25 runs vs projection] season xwOBA 0.246, weak - a good day, not a model error

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

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
HOU	Ethan Pecko (Starter)	8 / 15	10.12 / 0 (+10.12)	3 / 0	4 / 4	0 / 0	2 / 2	2.63	2 / 8	2.92	venue away: 6.75 vs 3.38 ERA (ok) · opp run factor 1 (weak) · 6 prior appearances this season
HOU	Logan VanWey (Setup Man)	1 / 4	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0	0 / 0	1.05
HOU	Josh Hader (Closer)	4 / 4	6.75 / 0 (+6.75)	1 / 0	2 / 0	0 / 0	4 / 0	0.19	0 / 1	1.11
HOU	AJ Blubaugh (Middle Reliever (High Leverage))	6 / 1	0 / 27 (-27.00)	0 / 1	0 / 1	0 / 1	1 / 0	0.57	1 / 0	0.85
HOU	Bryan Abreu (Middle Reliever (High Leverage))	3 / 0	0 / n/a	0 / 2	1 / 4	0 / 0	0 / 0	0.36	1 / 2	0.94
SEA	George Kirby (Starter)	15 / 18	7.2 / 3 (+4.20)	4 / 2	6 / 6	1 / 1	3 / 1	2.06	5 / 8	5.34	venue home: 3.27 vs 5.45 ERA (ok) · opp run factor 1 (average) · 29 prior appearances this season
SEA	José A. Ferrer (Setup Man)	3 / 0	0 / n/a	0 / 0	0 / 0	0 / 0	0 / 1	0.16	2 / 0	0.6
SEA	Gabe Speier (Middle Reliever (High Leverage))	3 / 0	0 / n/a	0 / 0	1 / 0	0 / 0	0 / 0	0.23	2 / 0	0.74

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Yainer Diaz (C)	HOU	5 / 4	2 / 1	0 / 0	1 / 0	1 / 0	3 / 1	1 / 0	0 / 0	0 / 1	2.02	0.309 average	no signal
Julio Rodríguez (CF)	SEA	4 / 4	3 / 1	1 / 0	0 / 0	0 / 0	2 / 0	0 / 0	1 / 0	0 / 1	1.58	0.334 above average	projection too low
Nick Allen (SS)	HOU	4 / 3	2 / 0	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	1.25	0.246 weak	good day
Taylor Trammell (LF)	HOU	5 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	3 / 1	-0.8	0.292 below average	projection too high
Cole Young (2B)	SEA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.314 average	—
Dominic Canzone (DH)	SEA	5 / 4	2 / 1	0 / 0	0 / 0	0 / 1	3 / 1	0 / 1	0 / 0	0 / 1	-0.46	0.355 elite	—
J.P. Crawford (SS)	SEA	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0 / 1	1 / 1	-0.33	0.322 above average	—
Josh Naylor (1B)	SEA	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0 / 1	0.33	0.311 average	—
Jose Altuve (2B)	HOU	5 / 4	2 / 1	0 / 1	0 / 0	0 / 0	1 / 0	1 / 1	0 / 0	2 / 1	0.16	0.272 weak	—
Yordan Alvarez (DH)	HOU	4 / 4	2 / 1	1 / 0	0 / 0	0 / 1	0 / 1	2 / 1	1 / 1	0 / 1	-0.15	0.445 elite	—
Lucas Spence (CF)	HOU	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	2 / 1	-0.14	0.236 weak	—
Cam Smith (RF)	HOU	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	-0.14	0.318 average	—
Lazaro Montes (RF)	SEA	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	1 / 1	-0.14	0.32 average	—
Isaac Paredes (3B)	HOU	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	1 / 0	-0.02	0.323 above average	—
Randy Arozarena (LF)	SEA	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	3 / 1	2 / 1	0 / 1	0.02	0.349 elite	—
Christian Walker (1B)	HOU	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	2 / 1	0	0.3 below average	—
Cal Raleigh (C)	SEA	4 / 4	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	2 / 1	0	0.311 average	—
Weston Wilson (3B)	SEA	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.26 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.083	positive favors home; negative favors away
Recent form	0.526	positive favors home; negative favors away
Starting-pitcher matchup	-0	positive favors home; negative favors away
Composite	0.357	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.72-4.44	n/a	Base ML historical model prediction
variance_applied	4-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.515	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	n/a	Capped at max 12 realistic runs
simulation_result	4-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.72-4.44	4-4	-0.724	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-4	4-5	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	4-5	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	4-5	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	4-5	0	Capped at max 12 realistic runs
simulation_result	4-5	4-5	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
SEA	George Kirby	pitcher / Starter	better	predicted_SO 3
HOU	Lucas Spence	batter / CF	better	predicted_RBI 2.75
HOU	Nick Allen	batter / SS	worse	predicted_AB 2.5
SEA	Weston Wilson	batter / 3B	better	predicted_AB 2.417
SEA	Randy Arozarena	batter / LF	worse	predicted_AB 2.295
SEA	Dominic Canzone	batter / DH	worse	predicted_SO 2.223
SEA	Lazaro Montes	batter / RF	deboosted	predicted_AB 2.2
HOU	Taylor Trammell	batter / LF	worse	predicted_AB 2.184

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	HOU pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	HOU
info	CURRENT_PITCHER_DUPLICATION	SEA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	SEA
info	CURRENT_FINALIZED_MATCH_BOX	HOU model pitcher projections differ from final simulation pitching box score (H: current=12, sim=11).	HOU
info	CURRENT_FINALIZED_MATCH_BOX	HOU model-prediction batting totals differ from final simulation box score (AB: current=32, sim=45; H: current=8, sim=10; 3B: current=0, sim=1; HR: current=1, sim=2; BB: current=3, sim=4; SO: current=8, sim=9; SF: current=0, sim=1).	HOU
info	CURRENT_PLAYER_MATCH_BOX	HOU current finalized hitter lines are model predictions and differ from the simulation box score: Yordan Alvarez: AB current=4 box=6, H current=1 box=2, BB current=1 box=0, SO current=1 box=2; Jose Altuve: AB current=4 box=6, 2B current=1 box=0, R current=1 box=0; Yainer Diaz: AB current=4 box=5, RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Isaac Paredes: AB current=4 box=5, 2B current=0 box=1, RBI current=1 box=0, SO current=0 box=2; Lucas Spence: AB current=3 box=5, H current=1 box=3, HR current=0 box=1, RBI current=0 box=3; Taylor Trammell: AB current=3 box=5, 3B current=0 box=1, BB current=1 box=0, SO current=1 box=0; Christian Walker: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Cam Smith: BB current=0 box=1, SO current=1 box=0, SF current=0 box=1	HOU
info	CURRENT_FINALIZED_MATCH_BOX	SEA model-prediction batting totals differ from final simulation box score (AB: current=34, sim=47; H: current=7, sim=11; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=2; SO: current=9, sim=12).	SEA
info	CURRENT_PLAYER_MATCH_BOX	SEA current finalized hitter lines are model predictions and differ from the simulation box score: Randy Arozarena: AB current=4 box=6, H current=1 box=2, 2B current=1 box=0, RBI current=1 box=0; J.P. Crawford: AB current=4 box=5, H current=1 box=2, SO current=1 box=2; Dominic Canzone: AB current=4 box=6, 2B current=0 box=1, HR current=1 box=0, SO current=1 box=3; Cal Raleigh: AB current=4 box=5, RBI current=1 box=0, R current=1 box=0; Josh Naylor: AB current=4 box=5, H current=1 box=2, RBI current=0 box=2, R current=0 box=1; Lazaro Montes: AB current=3 box=5, R current=0 box=1, SO current=1 box=0; Cole Young: AB current=4 box=5; Julio Rodríguez: AB current=4 box=5, SO current=1 box=2	SEA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

LAA@OAKMISS

Predicted: LAA 5-4 (conf -0.08, lean)Actual: OAK 3-7Brier 0.292total bias -1

Root cause: Small edge on the losing side

Evidence-backed diagnosis

published system edge 0.08 (lean); this is not the half-scale distance from 50%
team-run absolute error 5; total bias -1
pregame home-minus-away evidence: season -0.649, recent -0.398, pitcher +0.059, composite -0.507
confidence pipeline reached 0.582 before later publication/recency controls; published home probability was 0.460
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

OAK offense under-projected by 3

Team	Starter	Projected	Actual	ER err
LAA	Walbert Ureña	3 ER / 18 outs	5 ER / 12 outs	2
OAK	Jeffrey Springs	4 ER / 14 outs	3 ER / 13 outs	-1

Decision internals: run factors home×1.048 / away×0.99; 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:

Henry Bolte (OAK CF): 4-for-5, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.7]
Vaughn Grissom (LAA 1B): 2-for-4, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]
Donovan Walton (OAK 2B): 1-for-2, 1 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]
Mike Trout (LAA DH): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.76]
Zach Neto (LAA SS): 1-for-2, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.6]

Starter miss: Walbert Ureña (LAA) posted 11.25 ERA vs projected 4.50 (5 ER in 12 outs); he is materially worse pitching away (3.59 ERA there vs 2.22 on the other side), a venue-role weakness the model under-weighted.

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

Pitcher versus batter: what actually broke this game

AWAY pitching

Walbert Ureña (LAA) threw 12 outs, 5 ER = 11.25 ERA against a projected 4.5 ERA (6.75 ERA error)

Bullpen behind him: 2 ER over 3 outs.

The HOME lineup he faced produced 7 runs on 13 hits and 1 HR, against a projected 4 runs / 8 hits (3 runs vs projection).

Who did the damage to him:

Henry Bolte: 4-for-5, 0 HR, 1 RBI [+1.7 runs vs projection] season xwOBA 0.319, average - a good day, not a model error
Donovan Walton: 1-for-2, 1 HR, 3 RBI [+0.93 runs vs projection] season xwOBA 0.327, above average - real projection error

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

HOME pitching

Jeffrey Springs (OAK) threw 13 outs, 3 ER = 6.23 ERA against a projected 7.71 ERA (-1.48 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The AWAY lineup he faced produced 3 runs on 8 hits and 2 HR, against a projected 4 runs / 9 hits (-1 runs vs projection).

Who did the damage to him:

Vaughn Grissom: 2-for-4, 1 HR, 2 RBI [+1.09 runs vs projection] season xwOBA 0.309, average - a good day, not a model error
Mike Trout: 2-for-4, 1 HR, 1 RBI [+0.76 runs vs projection] season xwOBA 0.38, elite - real projection error

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

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
LAA	Walbert Ureña (Starter)	12 / 18	11.25 / 4.5 (+6.75)	5 / 3	7 / 3	1 / 0	1 / 4	1.93	2 / 6	6.58	venue away: 3.59 vs 2.22 ERA (worse here) · opp run factor 1.05 (weak) · 28 prior appearances this season
LAA	Tayler Saucedo (Middle Reliever (High Leverage))	1 / 2	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0.22	1 / 0	0.85
LAA	José Fermin (Middle Reliever (High Leverage))	2 / 1	27 / 27	2 / 1	3 / 1	0 / 1	0 / 0	0.5	0 / 1	1.27
OAK	Jeffrey Springs (Starter)	13 / 14	6.23 / 7.71	3 / 4	5 / 8	2 / 1	1 / 1	1.75	3 / 6	3.95	venue home: 6.94 vs 5.34 ERA (worse here) · opp run factor 0.99 (weak) · 28 prior appearances this season
OAK	Hogan Harris (Reliever)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	n/a	1 / 1	n/a
OAK	Luis Medina (Reliever)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	n/a	1 / 1	n/a

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Henry Bolte (CF)	OAK	5 / 4	4 / 1	2 / 0	0 / 0	0 / 0	1 / 0	2 / 1	0 / 1	0 / 1	1.7	0.319 average	no signal
Vaughn Grissom (1B)	LAA	4 / 4	2 / 1	0 / 1	0 / 0	1 / 0	2 / 1	1 / 0	0 / 0	0 / 1	1.09	0.309 average	no signal
Donovan Walton (2B)	OAK	2 / 3	1 / 1	0 / 0	0 / 0	1 / 0	3 / 0	1 / 0	1 / 1	0 / 1	0.93	0.327 above average	projection too low
Mike Trout (DH)	LAA	4 / 4	2 / 1	0 / 1	0 / 0	1 / 0	1 / 0	1 / 1	0 / 1	1 / 1	0.76	0.38 elite	projection too low
Zach Neto (SS)	LAA	2 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	0 / 1	-0.6	0.312 average	no signal
Denzer Guzman (3B)	LAA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	2 / 1	-0.47	0.271 weak	—
Christian Moore (2B)	LAA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	3 / 1	-0.47	0.228 weak	—
Tyler Heineman (C)	LAA	1 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	-0.47	0.237 weak	—
Carlos Cortes (LF)	OAK	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.305 average	—
Alika Williams (SS)	OAK	4 / 3	2 / 1	0 / 1	0 / 0	0 / 0	0 / 0	2 / 0	0 / 0	0 / 1	0.16	0.244 weak	—
Zack Gelof (3B)	OAK	3 / 4	2 / 1	1 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 0	0 / 1	-0.15	0.322 above average	—
Wade Meckler (RF)	LAA	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0.14	0.311 average	—
Jose Siri (LF)	LAA	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0 / 1	-0.14	0.298 below average	—
Lawrence Butler (RF)	OAK	5 / 4	2 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	1 / 1	0.14	0.327 above average	—
Shea Langeliers (DH)	OAK	5 / 4	1 / 1	1 / 0	0 / 0	0 / 0	2 / 1	0 / 1	0 / 1	2 / 1	-0.02	0.359 elite	—
Jeff McNeil (1B)	OAK	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	0 / 0	0.02	0.299 below average	—
Bryce Teodosio (CF)	LAA	2 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.194 weak	—
Brian Serven (C)	OAK	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	0	0.281 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.649	positive favors home; negative favors away
Recent form	-0.398	positive favors home; negative favors away
Starting-pitcher matchup	0.059	positive favors home; negative favors away
Composite	-0.507	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.35-4.19	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.54	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	n/a	Capped at max 12 realistic runs
simulation_result	5-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	5.35-4.19	5-4	-0.161	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	5-4	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	5-4	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	5-4	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	5-4	-0	Capped at max 12 realistic runs
simulation_result	5-4	5-4	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
OAK	Jeffrey Springs	pitcher / Starter	worse	predicted_IP -4.8
OAK	Seth Johnson	pitcher / Reliever	better	predicted_H 3
LAA	Bryce Teodosio	batter / CF	better	predicted_H 2.524
LAA	Jose Siri	batter / LF	better	predicted_AB 2.424
LAA	Walbert Ureña	pitcher / Starter	better	predicted_SO 2
LAA	Ben Joyce	pitcher / Closer	worse	predicted_SO -2
OAK	Chris Roycroft	pitcher / Reliever	worse	predicted_H 2
LAA	Wade Meckler	batter / RF	worse	predicted_AB 1.77

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	LAA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	LAA
info	CURRENT_PITCHER_DUPLICATION	OAK pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	OAK
info	CURRENT_FINALIZED_MATCH_BOX	LAA model-prediction batting totals differ from final simulation box score (AB: current=33, sim=40; H: current=9, sim=13; HR: current=1, sim=2; R: current=4, sim=5; BB: current=3, sim=1; SO: current=8, sim=11; HBP: current=0, sim=1).	LAA
info	CURRENT_PLAYER_MATCH_BOX	LAA current finalized hitter lines are model predictions and differ from the simulation box score: Mike Trout: AB current=4 box=5, H current=1 box=2, R current=1 box=0, BB current=1 box=0; Vaughn Grissom: AB current=4 box=5, 2B current=1 box=0, R current=0 box=1, SO current=1 box=2; Wade Meckler: AB current=4 box=5, R current=1 box=0, BB current=1 box=0; Denzer Guzman: HBP current=0 box=1; Zach Neto: AB current=4 box=5, H current=1 box=2, HR current=1 box=0, RBI current=1 box=0; Bryce Teodosio: AB current=3 box=4, H current=1 box=3, RBI current=0 box=1, R current=0 box=1; Christian Moore: H current=1 box=0, RBI current=1 box=0; Jose Siri: AB current=3 box=5, H current=1 box=3, 2B current=0 box=1, HR current=0 box=2	LAA
info	CURRENT_FINALIZED_MATCH_BOX	OAK model-prediction batting totals differ from final simulation box score (AB: current=32, sim=33; H: current=8, sim=6; RBI: current=5, sim=4; HBP: current=0, sim=1).	OAK
info	CURRENT_PLAYER_MATCH_BOX	OAK current finalized hitter lines are model predictions and differ from the simulation box score: Shea Langeliers: RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Henry Bolte: AB current=4 box=5, 2B current=0 box=1, RBI current=0 box=2, BB current=1 box=0; Jeff McNeil: AB current=4 box=3, 2B current=1 box=0, RBI current=1 box=0, HBP current=0 box=1; Carlos Cortes: R current=0 box=1, BB current=0 box=1, SO current=1 box=0; Alika Williams: AB current=3 box=4, H current=1 box=0, 2B current=1 box=0, SO current=1 box=2; Brian Serven: AB current=3 box=4, H current=0 box=1, HR current=0 box=1, R current=0 box=1; Donovan Walton: AB current=3 box=4, H current=1 box=0, BB current=1 box=0; Zack Gelof: AB current=4 box=3, 2B current=0 box=1, HR current=1 box=0, BB current=0 box=1	OAK
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

MIA@CHCMISS

Predicted: CHC 5-6 (conf +0.06, lean)Actual: MIA 3-2Brier 0.281total bias +6

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

Evidence-backed diagnosis

published system edge 0.06 (lean); this is not the half-scale distance from 50%
team-run absolute error 6; total bias +6
pregame home-minus-away evidence: season +0.797, recent +0.408, pitcher -0.013, composite +0.692
confidence pipeline reached 0.569 before later publication/recency controls; published home probability was 0.530
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 6 (11 vs 5)
CHC offense over-projected by 4

Team	Starter	Projected	Actual	ER err
MIA	Ryan Gusto	4 ER / 14 outs	2 ER / 16 outs	-2
CHC	Kevin Gausman	2 ER / 17 outs	2 ER / 17 outs	0

Decision internals: run factors home×0.967 / away×0.979; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

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

Top deviations that drove the game:

Pete Crow-Armstrong (CHC CF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Jakob Marsee (MIA CF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Griffin Conine (MIA RF): 2-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.79]
Seiya Suzuki (CHC RF): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.64]
Kyle Stowers (MIA 1B): 1-for-4, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 0.6]

Starter miss: Ryan Gusto (MIA) posted 3.38 ERA vs projected 7.71 (2 ER in 16 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

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

Pitcher versus batter: what actually broke this game

AWAY pitching

Ryan Gusto (MIA) threw 16 outs, 2 ER = 3.38 ERA against a projected 7.71 ERA (-4.33 ERA error)

Bullpen behind him: 0 ER over 5 outs.

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

Who did the damage to him:

Dansby Swanson: 1-for-3, 0 HR, 2 RBI [+0.33 runs vs projection] season xwOBA 0.305, average - a good day, not a model error
Michael Conforto: 1-for-4, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.336, above average - a good day, not a model error

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

HOME pitching

Kevin Gausman (CHC) threw 17 outs, 2 ER = 3.18 ERA against a projected 3.18 ERA (0 ERA error)

Bullpen behind him: 0 ER over 4 outs.

The AWAY lineup he faced produced 3 runs on 8 hits and 1 HR, against a projected 4 runs / 9 hits (-1 runs vs projection).

Who did the damage to him:

Kyle Stowers: 1-for-4, 1 HR, 2 RBI [+0.6 runs vs projection] season xwOBA 0.326, above average - real projection error
Javier Sanoja: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.294, below average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIA	Ryan Gusto (Starter)	16 / 14	3.38 / 7.71 (-4.33)	2 / 4	6 / 3	0 / 1	2 / 3	1.91	5 / 3	3.53	venue away: 4.15 vs 4.19 ERA (ok) · opp run factor 0.97 (strong) · 22 prior appearances this season
MIA	Calvin Faucher (Closer)	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 1	0 / 0	0 / 0	0.52	0 / 1	1
MIA	Michael Petersen (Middle Reliever (High Leverage))	2 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.43	2 / 1	1.42
CHC	Kevin Gausman (Starter)	17 / 17	3.18 / 3.18	2 / 2	6 / 4	1 / 1	1 / 0	1.5	7 / 5	6.55	venue home: 3.87 vs 5.4 ERA (ok) · opp run factor 0.98 (average) · 31 prior appearances this season
CHC	Jacob Webb (Closer)	4 / 3	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0.49	2 / 3	1.11

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Pete Crow-Armstrong (CF)	CHC	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	1 / 1	-1.4	0.376 elite	bad day
Jakob Marsee (CF)	MIA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	3 / 1	-0.8	0.313 average	no signal
Griffin Conine (RF)	MIA	4 / 4	2 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	2 / 1	-0.79	0.363 elite	bad day
Seiya Suzuki (RF)	CHC	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.64	0.338 above average	bad day
Kyle Stowers (1B)	MIA	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	2 / 1	1 / 1	0 / 1	1 / 1	0.6	0.326 above average	projection too low
Javier Sanoja (2B)	MIA	4 / 4	2 / 1	1 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 0	0.47	0.294 below average	—
Joe Mack (C)	MIA	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.284 below average	—
Dansby Swanson (SS)	CHC	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	2 / 0	0 / 1	1 / 0	0 / 1	0.33	0.305 average	—
Michael Conforto (DH)	CHC	4 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	0.31	0.336 above average	—
Heriberto Hernández (LF)	MIA	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 0	0 / 1	0.19	0.347 elite	—
Michael Busch (1B)	CHC	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	2 / 1	0.14	0.332 above average	—
Ian Happ (LF)	CHC	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	-0.02	0.315 average	—
Otto Lopez (SS)	MIA	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	1 / 1	0	0.327 above average	—
Graham Pauley (3B)	MIA	2 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.246 weak	—
Agustín Ramírez (DH)	MIA	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0	0.287 below average	—
Pedro Ramirez (3B)	CHC	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 0	0	0.284 below average	—
Nico Hoerner (2B)	CHC	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0	0.322 above average	—
Miguel Amaya (C)	CHC	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.291 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.797	positive favors home; negative favors away
Recent form	0.408	positive favors home; negative favors away
Starting-pitcher matchup	-0.013	positive favors home; negative favors away
Composite	0.692	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.56-5.23	n/a	Base ML historical model prediction
variance_applied	5-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-6	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	5-6	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

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.56-5.23	5-5	-0.673	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-5	5-6	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	5-6	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

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
CHC	Pedro Ramirez	batter / 3B	better	predicted_RBI 2.556
CHC	Seiya Suzuki	batter / RF	better	predicted_AB -2.45
CHC	Nico Hoerner	batter / 2B	worse	predicted_AB -2.43
CHC	Dansby Swanson	batter / SS	worse	predicted_SO 2.085
MIA	Ryan Gusto	pitcher / Starter	neutral	predicted_SO 2
MIA	Kyle Stowers	batter / 1B	worse	predicted_H 1.96
CHC	Pete Crow-Armstrong	batter / CF	worse	predicted_AB -1.77
MIA	Heriberto Hernández	batter / LF	worse	predicted_AB -1.73

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	CHC pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CHC
info	CURRENT_PITCHER_DUPLICATION	MIA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	MIA
info	CURRENT_FINALIZED_MATCH_BOX	CHC model-prediction batting totals differ from final simulation box score (AB: current=33, sim=30; H: current=8, sim=7; 2B: current=1, sim=0; 3B: current=0, sim=1; RBI: current=5, sim=6; R: current=5, sim=6; BB: current=4, sim=6; SF: current=0, sim=1).	CHC
info	CURRENT_PLAYER_MATCH_BOX	CHC current finalized hitter lines are model predictions and differ from the simulation box score: Seiya Suzuki: AB current=4 box=3, 2B current=1 box=0, HR current=0 box=1, SO current=1 box=0; Michael Busch: R current=1 box=0, BB current=1 box=0; Pete Crow-Armstrong: HR current=1 box=0, RBI current=1 box=0; Pedro Ramirez: H current=1 box=3, 3B current=0 box=1, RBI current=1 box=3, R current=0 box=1; Michael Conforto: RBI current=1 box=0, R current=0 box=1, BB current=0 box=1, SO current=1 box=0; Nico Hoerner: AB current=4 box=2, H current=1 box=0, RBI current=0 box=1, R current=0 box=1; Ian Happ: AB current=4 box=3, H current=1 box=0, SO current=1 box=0; Miguel Amaya: BB current=0 box=1	CHC
info	CURRENT_FINALIZED_MATCH_BOX	MIA model pitcher projections differ from final simulation pitching box score (H: current=5, sim=7; BB: current=3, sim=6).	MIA
info	CURRENT_FINALIZED_MATCH_BOX	MIA model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=9, sim=8; HR: current=1, sim=2; R: current=4, sim=5; BB: current=3, sim=2; SO: current=8, sim=9).	MIA
info	CURRENT_PLAYER_MATCH_BOX	MIA current finalized hitter lines are model predictions and differ from the simulation box score: Griffin Conine: AB current=4 box=3, HR current=1 box=0, SO current=1 box=0; Javier Sanoja: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, SO current=0 box=2; Heriberto Hernández: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Otto Lopez: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0; Kyle Stowers: H current=1 box=3, HR current=0 box=1, RBI current=1 box=2, BB current=1 box=0; Graham Pauley: AB current=3 box=4, R current=0 box=1, SO current=1 box=2; Joe Mack: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Jakob Marsee: H current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=1 box=2	MIA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

MIL@PHIHIT

Predicted: MIL 5-3 (conf -0.12, moderate)Actual: MIL 4-1Brier 0.194total bias +3

Root cause: Strong correct call

Evidence-backed diagnosis

published system edge 0.12 (moderate); this is not the half-scale distance from 50%
team-run absolute error 3; total bias +3
pregame home-minus-away evidence: season -1.009, recent -0.688, pitcher -0.051, composite -1.199
confidence pipeline reached 0.642 before later publication/recency controls; published home probability was 0.440
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 3 (8 vs 5)

Team	Starter	Projected	Actual	ER err
MIL	Logan Henderson	3 ER / 18 outs	1 ER / 15 outs	-2
PHI	Aaron Nola	3 ER / 18 outs	4 ER / 18 outs	1

Decision internals: run factors home×1.014 / away×1.065; 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:

Bryson Stott (PHI 3B): 2-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 1.42]
Kyle Schwarber (PHI DH): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Brice Turang (MIL 2B): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.07]
Jackson Chourio (MIL LF): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Christian Yelich (MIL DH): 1-for-3, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]

Pitcher versus batter: what actually broke this game

AWAY pitching

Logan Henderson (MIL) threw 15 outs, 1 ER = 1.8 ERA against a projected 4.5 ERA (-2.7 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The HOME lineup he faced produced 1 runs on 5 hits and 0 HR, against a projected 4 runs / 7 hits (-3 runs vs projection).

Who did the damage to him:

Bryson Stott: 2-for-3, 0 HR, 0 RBI [+1.42 runs vs projection] season xwOBA 0.328, above average - real projection error
Justin Crawford: 1-for-2, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.283, below average - a good day, not a model error

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

HOME pitching

Aaron Nola (PHI) threw 18 outs, 4 ER = 6 ERA against a projected 4.5 ERA (1.5 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 4 runs on 6 hits and 2 HR, against a projected 5 runs / 9 hits (-1 runs vs projection).

Who did the damage to him:

Brice Turang: 2-for-4, 1 HR, 1 RBI [+1.07 runs vs projection] season xwOBA 0.342, elite - real projection error
Christian Yelich: 1-for-3, 1 HR, 1 RBI [+0.93 runs vs projection] season xwOBA 0.307, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIL	Logan Henderson (Starter)	15 / 18	1.8 / 4.5	1 / 3	4 / 2	0 / 2	3 / 2	0.93	3 / 7	6.34	venue away: 3.1 vs 1.84 ERA (worse here) · opp run factor 1.01 (average) · 17 prior appearances this season
MIL	Abner Uribe (Setup Man)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.34	1 / 1	1.15
MIL	Trevor Megill (Closer)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.28	1 / 0	1.15
PHI	Aaron Nola (Starter)	18 / 18	6 / 4.5	4 / 3	4 / 6	2 / 1	1 / 2	2.38	8 / 7	4.31	venue home: 5.3 vs 3.84 ERA (worse here) · opp run factor 1.07 (strong) · 31 prior appearances this season

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Bryson Stott (3B)	PHI	3 / 3	2 / 1	2 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 0	0 / 1	1.42	0.328 above average	projection too low
Kyle Schwarber (DH)	PHI	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	1 / 2	-1.4	0.364 elite	bad day
Brice Turang (2B)	MIL	4 / 4	2 / 1	1 / 1	0 / 0	1 / 0	1 / 1	2 / 1	0 / 1	1 / 1	1.07	0.342 elite	projection too low
Jackson Chourio (LF)	MIL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	2 / 1	-0.93	0.339 above average	bad day
Christian Yelich (DH)	MIL	3 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 1	0 / 0	0 / 1	0.93	0.307 average	no signal
Garrett Mitchell (CF)	MIL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 1	1 / 1	-0.8	0.353 elite	bad day
Jake Bauers (1B)	MIL	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	3 / 1	-0.8	0.378 elite	bad day
Luis Arraez (2B)	PHI	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	-0.78	0.308 average	no signal
Sal Frelick (RF)	MIL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.279 weak	—
Bryce Harper (RF)	PHI	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	-0.47	0.381 elite	—
Justin Crawford (CF)	PHI	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.47	0.283 below average	—
David Hamilton (3B)	MIL	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	-0.14	0.27 weak	—
Brandon Marsh (LF)	PHI	1 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	-0.14	0.302 below average	—
William Contreras (C)	MIL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	2 / 1	0 / 1	0 / 0	1 / 1	0	0.335 above average	—
Joey Ortiz (SS)	MIL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.296 below average	—
Alec Bohm (1B)	PHI	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	2 / 1	0	0.331 above average	—
Trea Turner (SS)	PHI	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0	0.295 below average	—
J.T. Realmuto (C)	PHI	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0	0.319 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-1.009	positive favors home; negative favors away
Recent form	-0.688	positive favors home; negative favors away
Starting-pitcher matchup	-0.051	positive favors home; negative favors away
Composite	-1.199	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.95-3.61	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	5-3	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-3	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-3	n/a	Capped at max 12 realistic runs
simulation_result	5-3	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.95-3.61	5-4	-0.337	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	5-3	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	5-3	5-3	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-3	5-3	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-3	5-3	-0	Capped at max 12 realistic runs
simulation_result	5-3	5-3	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
MIL	William Contreras	batter / C	better	predicted_H 1.85
MIL	Brice Turang	batter / 2B	worse	predicted_H 1.82
MIL	Jake Bauers	batter / 1B	worse	predicted_AB 1.632
PHI	Bryson Stott	batter / 3B	deboosted	predicted_RBI 1.564
MIL	Garrett Mitchell	batter / CF	boosted	predicted_RBI 1.52
PHI	Alec Bohm	batter / 1B	worse	predicted_SO 1.422
MIL	Joey Ortiz	batter / SS	worse	predicted_SO 1.31
PHI	Luis Arraez	batter / 2B	worse	predicted_H -1.29

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	MIL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	MIL
info	CURRENT_PITCHER_DUPLICATION	PHI pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	PHI
info	CURRENT_FINALIZED_MATCH_BOX	MIL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=38; H: current=9, sim=11; 2B: current=1, sim=5; SO: current=9, sim=11).	MIL
info	CURRENT_PLAYER_MATCH_BOX	MIL current finalized hitter lines are model predictions and differ from the simulation box score: Garrett Mitchell: HR current=0 box=1, RBI current=1 box=2, R current=0 box=1, SO current=1 box=2; Jake Bauers: AB current=3 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Jackson Chourio: AB current=4 box=5, 2B current=0 box=1, HR current=1 box=0, SO current=1 box=0; William Contreras: H current=1 box=3, 2B current=0 box=1, BB current=0 box=1; Brice Turang: AB current=4 box=5, H current=1 box=3, 2B current=1 box=2, RBI current=1 box=0; Joey Ortiz: H current=1 box=0, SO current=1 box=2; Sal Frelick: AB current=3 box=4, H current=1 box=0; Christian Yelich: H current=1 box=2, RBI current=0 box=1, SO current=1 box=0	MIL
info	CURRENT_FINALIZED_MATCH_BOX	PHI model-prediction batting totals differ from final simulation box score (AB: current=32, sim=29; H: current=7, sim=2; 2B: current=1, sim=0; HR: current=1, sim=2; RBI: current=4, sim=3; R: current=4, sim=3; SO: current=9, sim=8; HBP: current=0, sim=1).	PHI
info	CURRENT_PLAYER_MATCH_BOX	PHI current finalized hitter lines are model predictions and differ from the simulation box score: Bryce Harper: AB current=4 box=3, HR current=0 box=1; Alec Bohm: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, R current=0 box=1; Bryson Stott: HR current=0 box=1, RBI current=1 box=2, SO current=1 box=0; Trea Turner: H current=1 box=0, R current=1 box=0; Luis Arraez: H current=1 box=0, 2B current=1 box=0; Kyle Schwarber: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Brandon Marsh: H current=1 box=0, SO current=1 box=0; J.T. Realmuto: SO current=1 box=2	PHI
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

MIN@SFHIT

Predicted: MIN 4-3 (conf -0.02, toss_up)Actual: MIN 4-2Brier 0.239total bias +1

Root cause: Toss-up that went our way

Evidence-backed diagnosis

published system edge 0.02 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 1; total bias +1
pregame home-minus-away evidence: season -0.137, recent -0.044, pitcher -0.073, composite -0.426
confidence pipeline reached 0.527 before later publication/recency controls; published home probability was 0.489
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

No large single-factor errors.

Team	Starter	Projected	Actual	ER err
MIN	Connor Prielipp	3 ER / 17 outs	2 ER / 16 outs	-1
SF	Cesar Perdomo	2 ER / 14 outs	2 ER / 16 outs	0

Decision internals: run factors home×0.928 / away×1.001; 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:

Victor Bericoto (SF LF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Turner Hill (SF CF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Jonah Cox (SF 2B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]
Ryan Jeffers (MIN C): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.64]
Ben Ross (MIN SS): 1-for-2, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.62]

Pitcher versus batter: what actually broke this game

AWAY pitching

Connor Prielipp (MIN) threw 16 outs, 2 ER = 3.38 ERA against a projected 4.76 ERA (-1.38 ERA error)

Bullpen behind him: 0 ER over 8 outs.

The HOME lineup he faced produced 2 runs on 4 hits and 0 HR, against a projected 4 runs / 8 hits (-2 runs vs projection).

Who did the damage to him:

Shay Whitcomb: 2-for-4, 0 HR, 2 RBI [+0.47 runs vs projection] season xwOBA 0.263, weak - a good day, not a model error
Andrew Knizner: 1-for-3, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.295, below average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

HOME pitching

Cesar Perdomo (SF) threw 16 outs, 2 ER = 3.38 ERA against a projected 3.86 ERA (-0.48 ERA error) only 4 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 2 ER over 3 outs.

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

Who did the damage to him:

Ben Ross: 1-for-2, 0 HR, 0 RBI [+0.62 runs vs projection] season xwOBA 0.147, weak - a good day, not a model error
Luke Keaschall: 1-for-3, 1 HR, 2 RBI [+0.6 runs vs projection] season xwOBA 0.315, average - a good day, not a model error
Alan Roden: 1-for-3, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.349, elite - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIN	Connor Prielipp (Starter)	16 / 17	3.38 / 4.76	2 / 3	4 / 2	0 / 0	2 / 1	1.87	3 / 4	5.31	venue away: 5.54 vs 4.85 ERA (worse here) · opp run factor 0.93 (weak) · 23 prior appearances this season
MIN	A.J. Minter (Middle Reliever (High Leverage))	3 / 4	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0.19	0 / 1	1.03
MIN	Yoendrys Gómez (Closer)	5 / 3	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0.54	2 / 0	0.71
SF	Cesar Perdomo (Starter)	16 / 14	3.38 / 3.86	2 / 2	5 / 4	1 / 0	0 / 2	1.69	2 / 5	2.86	venue home: 9.64 vs 1.69 ERA (ok) · opp run factor 1 (average) · thin sample: 4 prior appearances this season (high uncertainty)
SF	Dylan Smith (Closer)	2 / 3	27 / 0 (+27.00)	2 / 0	1 / 0	0 / 0	1 / 0	0.54	1 / 1	1.01
SF	Trent Harris (Setup Man)	1 / 4	0 / 13.5 (-13.50)	0 / 2	0 / 2	0 / 0	0 / 1	0.23	0 / 2	1.39

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Victor Bericoto (LF)	SF	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	0 / 1	-1.4	0.299 below average	projection too high
Turner Hill (CF)	SF	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0 / 1	-0.8	0.325 above average	bad day
Jonah Cox (2B)	SF	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	-0.78	0.265 weak	projection too high
Ryan Jeffers (C)	MIN	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.64	0.345 elite	bad day
Ben Ross (SS)	MIN	2 / 3	1 / 1	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 1	0.62	0.147 weak	good day
Luke Keaschall (RF)	MIN	3 / 4	1 / 1	0 / 0	0 / 0	1 / 0	2 / 0	1 / 1	0 / 1	1 / 0	0.6	0.315 average	no signal
Brooks Lee (3B)	MIN	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	1 / 1	-0.6	0.278 weak	projection too high
Christian Koss (SS)	SF	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	-0.47	0.248 weak	—
Shay Whitcomb (3B)	SF	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	2 / 0	0 / 0	0 / 0	1 / 1	0.47	0.263 weak	—
Alan Roden (LF)	MIN	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0 / 0	0.33	0.349 elite	—
Emmanuel Rodriguez (CF)	MIN	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0.33	0.349 elite	—
Josh Bell (DH)	MIN	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 1	0 / 1	-0.33	0.333 above average	—
Bo Davidson (RF)	SF	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	-0.33	0.389 elite	—
Andrew Knizner (C)	SF	3 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.31	0.295 below average	—
Marcelo Mayer (DH)	SF	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	1 / 1	0.19	0.254 weak	—
Royce Lewis (2B)	MIN	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.305 average	—
Aaron Sabato (1B)	MIN	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.156 weak	—
Brett Harris (1B)	SF	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.263 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.137	positive favors home; negative favors away
Recent form	-0.044	positive favors home; negative favors away
Starting-pitcher matchup	-0.073	positive favors home; negative favors away
Composite	-0.426	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.91-4.06	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-3	0.511	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-3	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-3	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-3	n/a	Capped at max 12 realistic runs
simulation_result	4-3	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.91-4.06	5-4	0.151	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	4-3	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-3	4-3	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-3	4-3	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-3	4-3	-0	Capped at max 12 realistic runs
simulation_result	4-3	4-3	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
SF	Cesar Perdomo	pitcher / Starter	worse	predicted_H 3
MIN	Ryan Jeffers	batter / C	worse	predicted_AB -2.45
SF	Jason Foley	pitcher / Middle Reliever (High Leverage)	better	predicted_H -2
SF	Brett Harris	batter / 1B	better	predicted_2B 1.844
MIN	Brooks Lee	batter / 3B	worse	predicted_AB -1.78
MIN	Ben Ross	batter / SS	better	predicted_RBI 1.73
MIN	Alan Roden	batter / LF	better	predicted_SO 1.586
MIN	Luke Keaschall	batter / RF	worse	predicted_AB 1.497

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	MIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	MIN
info	CURRENT_PITCHER_DUPLICATION	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_FINALIZED_MATCH_BOX	MIN model-prediction batting totals differ from final simulation box score (AB: current=32, sim=33; H: current=8, sim=6; 2B: current=1, sim=2; 3B: current=0, sim=1; HR: current=1, sim=0; BB: current=3, sim=4; SO: current=7, sim=10).	MIN
info	CURRENT_PLAYER_MATCH_BOX	MIN current finalized hitter lines are model predictions and differ from the simulation box score: Luke Keaschall: AB current=4 box=5, H current=1 box=0, R current=1 box=0, BB current=1 box=0; Alan Roden: AB current=3 box=4, 2B current=0 box=1, RBI current=1 box=0, SO current=0 box=2; Royce Lewis: 2B current=0 box=1, RBI current=0 box=1, R current=0 box=1; Ryan Jeffers: AB current=4 box=2, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Brooks Lee: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Emmanuel Rodriguez: AB current=3 box=4, H current=1 box=2, RBI current=0 box=1, R current=0 box=1; Josh Bell: RBI current=1 box=0, R current=0 box=1, BB current=1 box=0; Aaron Sabato: AB current=3 box=4, SO current=1 box=2	MIN
info	CURRENT_FINALIZED_MATCH_BOX	SF model-prediction batting totals differ from final simulation box score (H: current=8, sim=4; 2B: current=1, sim=3; HR: current=1, sim=0; RBI: current=4, sim=3; R: current=4, sim=3; BB: current=3, sim=1; SO: current=9, sim=5; HBP: current=0, sim=1).	SF
info	CURRENT_PLAYER_MATCH_BOX	SF current finalized hitter lines are model predictions and differ from the simulation box score: Jonah Cox: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Victor Bericoto: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Bo Davidson: RBI current=1 box=0, BB current=1 box=0, SO current=1 box=2; Brett Harris: AB current=3 box=4, H current=0 box=2, 2B current=0 box=2, RBI current=0 box=1; Marcelo Mayer: H current=1 box=0, SO current=1 box=0; Christian Koss: H current=1 box=0, R current=1 box=0, SO current=1 box=2; Shay Whitcomb: H current=1 box=0, SO current=1 box=0; Turner Hill: 2B current=0 box=1, RBI current=1 box=2, R current=0 box=1, BB current=1 box=0	SF
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

NYM@TEXHIT

Predicted: NYM 6-3 (conf -0.19, moderate)Actual: NYM 7-2Brier 0.164total bias +0

Root cause: Strong correct call

Evidence-backed diagnosis

published system edge 0.19 (moderate); this is not the half-scale distance from 50%
team-run absolute error 2; total bias +0
pregame home-minus-away evidence: season +0.116, recent -0.138, pitcher -0.072, composite -0.669
confidence pipeline reached 0.622 before later publication/recency controls; published home probability was 0.405
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

No large single-factor errors.

Team	Starter	Projected	Actual	ER err
NYM	Nolan McLean	0 ER / 18 outs	2 ER / 21 outs	2
TEX	Cody Bradford	2 ER / 14 outs	2 ER / 13 outs	0

Decision internals: run factors home×0.927 / away×0.999; 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:

Corey Seager (TEX SS): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Marcus Semien (NYM 2B): 1-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]
Wyatt Langford (TEX LF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Juan Soto (NYM DH): 1-for-2, 1 HR, 2 RBI (projected 1.0 H, 1.0 HR) [run impact 0.66]

Pitcher versus batter: what actually broke this game

AWAY pitching

Nolan McLean (NYM) threw 21 outs, 2 ER = 2.57 ERA against a projected 0 ERA (2.57 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 2 runs on 4 hits and 0 HR, against a projected 4 runs / 7 hits (-2 runs vs projection).

Who did the damage to him:

Danny Jansen: 0-for-2, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.266, weak - a good day, not a model error

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

HOME pitching

Cody Bradford (TEX) threw 13 outs, 2 ER = 4.15 ERA against a projected 3.86 ERA (0.29 ERA error) only 8 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 1 ER over 6 outs.

The AWAY lineup he faced produced 6 runs on 7 hits and 2 HR, against a projected 4 runs / 9 hits (2 runs vs projection).

Who did the damage to him:

Marcus Semien: 1-for-4, 1 HR, 1 RBI [+0.93 runs vs projection] season xwOBA 0.302, below average - a good day, not a model error
Juan Soto: 1-for-2, 1 HR, 2 RBI [+0.66 runs vs projection] season xwOBA 0.411, elite - real projection error
A.J. Ewing: 1-for-3, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.319, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
NYM	Nolan McLean (Starter)	21 / 18	2.57 / 0	2 / 0	3 / 0	0 / 0	1 / 0	2.81	5 / 6	5.75	venue away: 2.89 vs 3.44 ERA (ok) · opp run factor 0.93 (average) · 30 prior appearances this season
TEX	Cody Bradford (Starter)	13 / 14	4.15 / 3.86	2 / 2	4 / 4	1 / 2	0 / 0	1.21	1 / 11	2.35	venue home: 4.61 vs 4.79 ERA (ok) · opp run factor 1 (average) · 8 prior appearances this season
TEX	Jacob Latz (Closer)	5 / 5	5.4 / 0 (+5.40)	1 / 0	1 / 0	1 / 0	1 / 0	0.41	2 / 1	1.65
TEX	Chase Silseth (Middle Reliever (High Leverage))	1 / 3	0 / 27 (-27.00)	0 / 3	1 / 3	0 / 1	1 / 2	0.32	1 / 0	0.9

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Corey Seager (SS)	TEX	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	2 / 1	-1.73	0.349 elite	bad day
Marcus Semien (2B)	NYM	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 1	0 / 0	0 / 0	0.93	0.302 below average	good day
Wyatt Langford (LF)	TEX	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	-0.8	0.324 above average	bad day
Juan Soto (DH)	NYM	2 / 4	1 / 1	0 / 0	0 / 0	1 / 1	2 / 1	2 / 1	3 / 1	0 / 0	0.66	0.411 elite	projection too low
Carson Benge (RF)	NYM	5 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	-0.47	0.34 elite	—
Mark Vientos (1B)	NYM	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.311 average	—
Ezequiel Duran (3B)	TEX	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.47	0.303 below average	—
Bo Bichette (3B)	NYM	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	0 / 1	-0.45	0.328 above average	—
A.J. Ewing (CF)	NYM	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	0.33	0.319 average	—
Francisco Lindor (SS)	NYM	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	-0.33	0.332 above average	—
Danny Jansen (C)	TEX	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0.33	0.266 weak	—
Evan Carter (CF)	TEX	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0 / 1	-0.33	0.278 weak	—
Brandon Nimmo (RF)	TEX	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	1 / 1	-0.31	0.343 elite	—
Francisco Alvarez (C)	NYM	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0.14	0.324 above average	—
Nick Morabito (LF)	NYM	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	2 / 1	0	0.315 average	—
Josh Jung (DH)	TEX	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0	0.345 elite	—
Jake Burger (1B)	TEX	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	1 / 0	0 / 0	0 / 1	0	0.318 average	—
Nicky Lopez (2B)	TEX	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.266 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.116	positive favors home; negative favors away
Recent form	-0.138	positive favors home; negative favors away
Starting-pitcher matchup	-0.072	positive favors home; negative favors away
Composite	-0.669	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.84-3.25	n/a	Base ML historical model prediction
variance_applied	5-3	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	0.595	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	6-3	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-3	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-3	n/a	Capped at max 12 realistic runs
simulation_result	6-3	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.84-3.25	5-3	0.411	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-3	5-3	-0	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	5-3	6-3	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-3	6-3	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-3	6-3	-0	Capped at max 12 realistic runs
simulation_result	6-3	6-3	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
NYM	Marcus Semien	batter / 2B	better	predicted_RBI 3.65
TEX	Nicky Lopez	batter / 2B	worse	predicted_SO 2.517
NYM	Carson Benge	batter / RF	worse	predicted_SO 2.082
TEX	Brandon Nimmo	batter / RF	worse	predicted_AB -1.54
TEX	Wyatt Langford	batter / LF	better	predicted_RBI 1.48
TEX	Corey Seager	batter / SS	worse	predicted_AB -1.45
NYM	Juan Soto	batter / DH	worse	predicted_SO 1.394
NYM	Nick Morabito	batter / LF	better	predicted_AB 1.385

Structural audit findings

Severity	Rule	Finding	Subject
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_PITCHER_DUPLICATION	TEX pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	TEX
info	CURRENT_FINALIZED_MATCH_BOX	NYM model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=9, sim=8; HR: current=1, sim=3; RBI: current=4, sim=6; R: current=4, sim=6; BB: current=3, sim=2; SO: current=7, sim=14).	NYM
info	CURRENT_PLAYER_MATCH_BOX	NYM current finalized hitter lines are model predictions and differ from the simulation box score: Juan Soto: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Carson Benge: SO current=1 box=3; Bo Bichette: H current=1 box=2, 2B current=1 box=0, HR current=0 box=1, R current=0 box=1; Mark Vientos: H current=1 box=0, RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; A.J. Ewing: H current=1 box=0, BB current=0 box=1; Nick Morabito: AB current=3 box=4, 2B current=0 box=1, RBI current=0 box=1, R current=0 box=1; Marcus Semien: H current=1 box=3, HR current=0 box=2, RBI current=0 box=4; Francisco Lindor: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0	NYM
info	CURRENT_FINALIZED_MATCH_BOX	TEX model-prediction batting totals differ from final simulation box score (AB: current=32, sim=29; H: current=7, sim=2; RBI: current=4, sim=3; R: current=4, sim=3; BB: current=3, sim=5).	TEX
info	CURRENT_PLAYER_MATCH_BOX	TEX current finalized hitter lines are model predictions and differ from the simulation box score: Josh Jung: H current=1 box=0, R current=1 box=0, SO current=1 box=2; Ezequiel Duran: 2B current=0 box=1, SO current=1 box=0; Wyatt Langford: AB current=4 box=3, HR current=0 box=1, RBI current=0 box=2, SO current=1 box=0; Corey Seager: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Jake Burger: H current=1 box=0, RBI current=1 box=0, BB current=0 box=1, SO current=1 box=0; Brandon Nimmo: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Nicky Lopez: AB current=3 box=4, H current=1 box=0, SO current=1 box=3; Danny Jansen: AB current=3 box=2, BB current=0 box=1, SO current=1 box=2	TEX
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

SD@LADMISS

Predicted: LAD 5-7 (conf +0.07, lean)Actual: SD 5-1Brier 0.286total bias +6

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

Evidence-backed diagnosis

published system edge 0.07 (lean); this is not the half-scale distance from 50%
team-run absolute error 6; total bias +6
pregame home-minus-away evidence: season +0.749, recent +0.210, pitcher +0.042, composite +0.917
confidence pipeline reached 0.541 before later publication/recency controls; published home probability was 0.535
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 9 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 6 (12 vs 6)
LAD offense over-projected by 6

Team	Starter	Projected	Actual	ER err
SD	Robbie Ray	3 ER / 15 outs	1 ER / 10 outs	-2
LAD	Yoshinobu Yamamoto	3 ER / 20 outs	2 ER / 18 outs	-1

Decision internals: run factors home×1.029 / away×0.987; 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:

Mookie Betts (LAD SS): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Manny Machado (SD 3B): 1-for-4, 1 HR, 3 RBI (projected 0.0 H, 0.0 HR) [run impact 1.07]
Teoscar Hernández (LAD LF): 3-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.94]
Andy Pages (LAD CF): 3-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.94]
Ty France (SD 1B): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]

Pitcher versus batter: what actually broke this game

AWAY pitching

Robbie Ray (SD) threw 10 outs, 1 ER = 2.7 ERA against a projected 5.4 ERA (-2.7 ERA error)

Bullpen behind him: 0 ER over 12 outs.

The HOME lineup he faced produced 1 runs on 10 hits and 0 HR, against a projected 5 runs / 8 hits (-4 runs vs projection).

Who did the damage to him:

Teoscar Hernández: 3-for-4, 0 HR, 0 RBI [+0.94 runs vs projection] season xwOBA 0.331, above average - real projection error
Andy Pages: 3-for-4, 0 HR, 0 RBI [+0.94 runs vs projection] season xwOBA 0.342, elite - real projection error
Will Smith: 1-for-2, 0 HR, 0 RBI [+0.66 runs vs projection] season xwOBA 0.385, elite - real projection error

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

HOME pitching

Yoshinobu Yamamoto (LAD) threw 18 outs, 2 ER = 3 ERA against a projected 4.05 ERA (-1.05 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 5 runs on 6 hits and 1 HR, against a projected 6 runs / 7 hits (-1 runs vs projection).

Who did the damage to him:

Manny Machado: 1-for-4, 1 HR, 3 RBI [+1.07 runs vs projection] season xwOBA 0.329, above average - real projection error
Fernando Tatis Jr.: 3-for-5, 0 HR, 0 RBI [+0.61 runs vs projection] season xwOBA 0.38, elite - real projection error
Ethan Salas: 1-for-2, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.199, weak - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
SD	Robbie Ray (Starter)	10 / 15	2.7 / 5.4	1 / 3	7 / 5	0 / 2	1 / 2	2.92	2 / 5	3.99	venue away: 4.36 vs 3.2 ERA (worse here) · opp run factor 1.03 (strong) · 30 prior appearances this season
SD	Bradgley Rodriguez (Middle Reliever (High Leverage))	6 / 3	0 / 0	0 / 0	1 / 1	0 / 0	1 / 0	0.2	3 / 0	1.02
SD	Mason Miller (Closer)	3 / 3	0 / 27 (-27.00)	0 / 3	0 / 0	0 / 0	1 / 1	0.35	2 / 0	1.72
SD	Yuki Matsui (Middle Reliever (High Leverage))	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0.62	0 / 1	0.99
LAD	Yoshinobu Yamamoto (Starter)	18 / 20	3 / 4.05	2 / 3	5 / 6	1 / 1	1 / 2	1.56	4 / 8	6.91	venue home: 3.13 vs 1.77 ERA (worse here) · opp run factor 0.99 (average) · 27 prior appearances this season

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Mookie Betts (SS)	LAD	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	1 / 0	-1.4	0.339 above average	bad day
Manny Machado (3B)	SD	4 / 4	1 / 0	0 / 0	0 / 0	1 / 0	3 / 1	1 / 0	0 / 1	0 / 1	1.07	0.329 above average	projection too low
Teoscar Hernández (LF)	LAD	4 / 4	3 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	1 / 1	0.94	0.331 above average	projection too low
Andy Pages (CF)	LAD	4 / 4	3 / 1	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	0.94	0.342 elite	projection too low
Ty France (1B)	SD	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	0 / 1	-0.93	0.333 above average	bad day
Freddie Freeman (1B)	LAD	5 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.78	0.365 elite	bad day
Will Smith (C)	LAD	2 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	3 / 1	0 / 1	0.66	0.385 elite	projection too low
Fernando Tatis Jr. (RF)	SD	5 / 4	3 / 1	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	1 / 1	0.61	0.38 elite	projection too low
Dustin Harris (LF)	SD	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	0 / 1	-0.47	0.309 average	—
Jackson Merrill (CF)	SD	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.47	0.347 elite	—
Jake Cronenworth (2B)	SD	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	-0.47	0.273 weak	—
Ethan Salas (C)	SD	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.47	0.199 weak	—
Xander Bogaerts (SS)	SD	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	2 / 1	-0.47	0.323 above average	—
Max Muncy (3B)	LAD	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.349 elite	—
Samad Taylor (DH)	SD	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 0	1 / 1	-0.14	0.318 average	—
Kyle Tucker (RF)	LAD	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0.14	0.334 above average	—
Shohei Ohtani (DH)	LAD	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	2 / 1	0	0.399 elite	—
Tommy Edman (2B)	LAD	3 / 4	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0	0.305 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.749	positive favors home; negative favors away
Recent form	0.21	positive favors home; negative favors away
Starting-pitcher matchup	0.042	positive favors home; negative favors away
Composite	0.917	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.18-5.33	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-6	0.535	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

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.18-5.33	4-5	-0.146	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

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
SD	Robbie Ray	pitcher / Starter	better	predicted_SO 3
LAD	Yoshinobu Yamamoto	pitcher / Starter	better	predicted_SO 3
LAD	Kyle Tucker	batter / RF	better	predicted_RBI 2.51
SD	Jackson Merrill	batter / CF	worse	predicted_AB -2.38
SD	Manny Machado	batter / 3B	worse	predicted_SO 2.092
LAD	Alex Vesia	pitcher / Middle Reliever (High Leverage)	worse	predicted_SO -2
LAD	Freddie Freeman	batter / 1B	better	predicted_H 1.76
SD	Dustin Harris	batter / LF	better	predicted_AB 1.67

Structural audit findings

Severity	Rule	Finding	Subject
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_PITCHER_DUPLICATION	SD pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	SD
info	CURRENT_FINALIZED_MATCH_BOX	LAD model pitcher projections differ from final simulation pitching box score (BB: current=2, sim=3).	LAD
info	CURRENT_FINALIZED_MATCH_BOX	LAD model-prediction batting totals differ from final simulation box score (AB: current=34, sim=31; H: current=8, sim=7; HR: current=1, sim=2; RBI: current=5, sim=7; R: current=5, sim=7; BB: current=4, sim=3; SO: current=8, sim=6).	LAD
info	CURRENT_PLAYER_MATCH_BOX	LAD current finalized hitter lines are model predictions and differ from the simulation box score: Shohei Ohtani: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Will Smith: AB current=4 box=3, H current=1 box=0, R current=0 box=1, SO current=1 box=0; Mookie Betts: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Teoscar Hernández: AB current=4 box=3, HR current=0 box=1, RBI current=1 box=2, BB current=0 box=1; Freddie Freeman: H current=1 box=3, RBI current=0 box=2, R current=0 box=1, SO current=1 box=0; Kyle Tucker: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, HR current=0 box=1; Andy Pages: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Max Muncy: AB current=3 box=2, H current=1 box=0, RBI current=1 box=0, R current=0 box=1	LAD
info	CURRENT_FINALIZED_MATCH_BOX	SD model pitcher projections differ from final simulation pitching box score (H: current=6, sim=7; ER: current=6, sim=7).	SD
info	CURRENT_FINALIZED_MATCH_BOX	SD model-prediction batting totals differ from final simulation box score (AB: current=34, sim=35; H: current=7, sim=8; 2B: current=1, sim=2; 3B: current=0, sim=1; RBI: current=4, sim=5; R: current=6, sim=5; SO: current=9, sim=11).	SD
info	CURRENT_PLAYER_MATCH_BOX	SD current finalized hitter lines are model predictions and differ from the simulation box score: Manny Machado: H current=0 box=1, 2B current=0 box=1, R current=0 box=1, BB current=1 box=0; Dustin Harris: AB current=4 box=5, 3B current=0 box=1, R current=0 box=1, SO current=1 box=2; Jackson Merrill: AB current=4 box=3, RBI current=1 box=0, R current=1 box=0, BB current=0 box=1; Samad Taylor: AB current=3 box=4, RBI current=0 box=1, SO current=1 box=0; Fernando Tatis Jr.: H current=1 box=0, 2B current=1 box=0, R current=1 box=0; Jake Cronenworth: H current=1 box=2, R current=1 box=0; Ethan Salas: AB current=3 box=4; Xander Bogaerts: AB current=4 box=3, 2B current=0 box=1, RBI current=0 box=1, SO current=1 box=2	SD
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

STL@PITMISS

Predicted: PIT 5-6 (conf +0.11, moderate)Actual: STL 5-1Brier 0.309total bias +5

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

Evidence-backed diagnosis

published system edge 0.11 (moderate); this is not the half-scale distance from 50%
team-run absolute error 5; total bias +5
pregame home-minus-away evidence: season +0.367, recent +0.526, pitcher +0.044, composite +0.609
confidence pipeline reached 0.581 before later publication/recency controls; published home probability was 0.556
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 5 (11 vs 6)
PIT offense over-projected by 5
committed to PIT at 0.11 confidence and it lost

Team	Starter	Projected	Actual	ER err
STL	Matthew Liberatore	3 ER / 15 outs	1 ER / 17 outs	-2
PIT	Lake Bachar	2 ER / 15 outs	0 ER / 5 outs	-2

Decision internals: run factors home×1.023 / away×0.98; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

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

Top deviations that drove the game:

Rafael Flores Jr. (PIT C): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Ryan O'Hearn (PIT 1B): 2-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]
Brandon Lowe (PIT 2B): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.07]
JJ Wetherholt (STL 2B): 3-for-4, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.94]
Konnor Griffin (PIT SS): 0-for-4, 0 HR, 0 RBI (projected 2.0 H, 0.0 HR) [run impact -0.94]

Starter miss: Matthew Liberatore (STL) posted 1.59 ERA vs projected 5.40 (1 ER in 17 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

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

Pitcher versus batter: what actually broke this game

AWAY pitching

Matthew Liberatore (STL) threw 17 outs, 1 ER = 1.59 ERA against a projected 5.4 ERA (-3.81 ERA error)

Bullpen behind him: 0 ER over 3 outs.

The HOME lineup he faced produced 1 runs on 6 hits and 1 HR, against a projected 5 runs / 10 hits (-4 runs vs projection).

Who did the damage to him:

Ryan O'Hearn: 2-for-3, 0 HR, 0 RBI [+1.09 runs vs projection] season xwOBA 0.316, average - a good day, not a model error
Brandon Lowe: 2-for-4, 1 HR, 1 RBI [+1.07 runs vs projection] season xwOBA 0.33, above average - real projection error

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

HOME pitching

Lake Bachar (PIT) threw 5 outs, 0 ER = 0 ERA against a projected 3.6 ERA (-3.6 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 5 runs on 10 hits and 0 HR, against a projected 4 runs / 8 hits (1 runs vs projection).

Who did the damage to him:

JJ Wetherholt: 3-for-4, 0 HR, 1 RBI [+0.94 runs vs projection] season xwOBA 0.347, elite - real projection error
Masyn Winn: 1-for-3, 0 HR, 1 RBI [+0.64 runs vs projection] season xwOBA 0.298, below average - a good day, not a model error
Nolan Gorman: 1-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.28, weak - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
STL	Matthew Liberatore (Starter)	17 / 15	1.59 / 5.4 (-3.81)	1 / 3	6 / 6	1 / 1	0 / 1	2.09	6 / 4	4.41	venue away: 5.27 vs 5.45 ERA (ok) · opp run factor 1.02 (average) · 30 prior appearances this season
STL	George Soriano (Setup Man)	3 / 4	0 / 6.75 (-6.75)	0 / 1	0 / 1	0 / 0	1 / 0	0.41	2 / 0	0.56
PIT	Lake Bachar (Starter)	5 / 15	0 / 3.6 (-3.60)	0 / 2	1 / 2	0 / 1	1 / 3	2.13	2 / 5	4.32	venue home: 2.13 vs 2.53 ERA (ok) · opp run factor 0.98 (average) · 51 prior appearances this season

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Rafael Flores Jr. (C)	PIT	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	1 / 1	-1.73	0.358 elite	bad day
Ryan O'Hearn (1B)	PIT	3 / 4	2 / 1	2 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	1.09	0.316 average	no signal
Brandon Lowe (2B)	PIT	4 / 4	2 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 1	0 / 1	1 / 1	1.07	0.33 above average	projection too low
JJ Wetherholt (2B)	STL	4 / 4	3 / 1	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	1 / 1	1 / 1	0.94	0.347 elite	projection too low
Konnor Griffin (SS)	PIT	4 / 4	0 / 2	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	-0.94	0.307 average	no signal
Jordan Walker (RF)	STL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-0.93	0.342 elite	bad day
Bryan Reynolds (LF)	PIT	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	-0.78	0.354 elite	bad day
Masyn Winn (SS)	STL	3 / 3	1 / 1	1 / 0	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	2 / 1	0.64	0.298 below average	good day
Iván Herrera (DH)	STL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	0 / 1	-0.47	0.333 above average	—
Nolan Gorman (3B)	STL	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	0.47	0.28 weak	—
José Fermín (LF)	STL	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	1 / 0	0.47	0.295 below average	—
Nathan Church (CF)	STL	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.287 below average	—
Ronny Simon (RF)	PIT	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	-0.47	0.333 above average	—
Jake Mangum (CF)	PIT	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.289 below average	—
Leonardo Bernal (C)	STL	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	1 / 1	0 / 1	0 / 0	1 / 1	-0.31	0.366 elite	—
Alec Burleson (1B)	STL	5 / 4	1 / 1	0 / 1	0 / 0	0 / 0	1 / 1	1 / 0	0 / 0	3 / 1	-0.31	0.371 elite	—
Nick Gonzales (DH)	PIT	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	-0.31	0.324 above average	—
Jared Triolo (3B)	PIT	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.286 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.367	positive favors home; negative favors away
Recent form	0.526	positive favors home; negative favors away
Starting-pitcher matchup	0.044	positive favors home; negative favors away
Composite	0.609	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.14-5.51	n/a	Base ML historical model prediction
variance_applied	4-6	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-6	0.556	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-6	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

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.14-5.51	4-6	0.63	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-6	5-6	-1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-6	5-6	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-6	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

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
PIT	Jared Triolo	batter / 3B	better	predicted_RBI 2.736
STL	José Fermín	batter / LF	worse	predicted_SO 2.593
PIT	Lake Bachar	pitcher / Starter	better	predicted_ER -2
STL	Masyn Winn	batter / SS	better	predicted_2B 1.841
STL	Nathan Church	batter / CF	better	predicted_BB 1.798
STL	Leonardo Bernal	batter / C	worse	predicted_AB -1.66
PIT	Bryan Reynolds	batter / LF	neutral	predicted_RBI 1.59
STL	JJ Wetherholt	batter / 2B	worse	predicted_BB 1.527

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	PIT pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	PIT
info	CURRENT_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	PIT model-prediction batting totals differ from final simulation box score (AB: current=34, sim=32; H: current=10, sim=9; 2B: current=2, sim=3; RBI: current=5, sim=6; R: current=5, sim=6; BB: current=3, sim=2; SO: current=8, sim=4; SF: current=0, sim=1).	PIT
info	CURRENT_PLAYER_MATCH_BOX	PIT current finalized hitter lines are model predictions and differ from the simulation box score: Konnor Griffin: H current=2 box=1, 2B current=0 box=1; Bryan Reynolds: H current=1 box=2, RBI current=1 box=2, BB current=1 box=0; Nick Gonzales: H current=1 box=0, 2B current=1 box=0; Ryan O'Hearn: AB current=4 box=3, RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Jared Triolo: AB current=3 box=2, H current=1 box=2, HR current=0 box=1, RBI current=0 box=3; Rafael Flores Jr.: H current=1 box=2, 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0; Brandon Lowe: RBI current=1 box=0, BB current=1 box=0, SO current=1 box=0; Ronny Simon: AB current=4 box=3, H current=1 box=0, R current=1 box=0, SO current=1 box=0	PIT
info	CURRENT_FINALIZED_MATCH_BOX	STL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=32; H: current=8, sim=5; 3B: current=0, sim=1; HR: current=1, sim=2; R: current=4, sim=5; BB: current=3, 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: JJ Wetherholt: AB current=4 box=3, H current=1 box=0, BB current=1 box=2; Jordan Walker: BB current=0 box=1, SO current=1 box=0; Iván Herrera: HR current=0 box=1, BB current=1 box=0, HBP current=0 box=1; Leonardo Bernal: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Alec Burleson: AB current=4 box=5, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; José Fermín: H current=1 box=0, RBI current=1 box=0, SO current=0 box=3; Masyn Winn: AB current=3 box=4, H current=1 box=2, 2B current=0 box=2, RBI current=0 box=2; Nathan Church: AB current=3 box=2, 3B current=0 box=1, RBI current=0 box=1, R current=0 box=1	STL
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

TB@NYYMISS

Predicted: TB 5-4 (conf -0.01, toss_up)Actual: NYY 2-9Brier 0.255total bias -2

Root cause: Starting-pitcher projection was wrong

Evidence-backed diagnosis

published system edge 0.01 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 8; total bias -2
pregame home-minus-away evidence: season -0.064, recent -0.714, pitcher +0.004, composite -0.385
confidence pipeline reached 0.507 before later publication/recency controls; published home probability was 0.495
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

TB offense over-projected by 3
NYY offense under-projected by 5
Mason Englert (TB) allowed 5 more ER than projected (8 vs 3)

Team	Starter	Projected	Actual	ER err
TB	Mason Englert	3 ER / 15 outs	8 ER / 11 outs	5
NYY	Gerrit Cole	1 ER / 17 outs	2 ER / 17 outs	1

Decision internals: run factors home×1.021 / away×1.017; 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:

Ben Rice (NYY DH): 3-for-4, 2 HR, 2 RBI (projected 1.0 H, 1.0 HR) [run impact 1.87]
Junior Caminero (TB 3B): 0-for-2, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Cedric Mullins (TB CF): 1-for-4, 1 HR, 1 RBI (projected 0.0 H, 0.0 HR) [run impact 1.4]
Ryan McMahon (NYY 3B): 2-for-4, 1 HR, 5 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Jonny DeLuca (TB RF): 3-for-4, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.25]

Starter miss: Mason Englert (TB) posted 19.64 ERA vs projected 5.40 (8 ER in 11 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

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

Pitcher versus batter: what actually broke this game

AWAY pitching

Mason Englert (TB) threw 11 outs, 8 ER = 19.64 ERA against a projected 5.4 ERA (14.24 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 9 runs on 12 hits and 4 HR, against a projected 4 runs / 8 hits (5 runs vs projection).

Who did the damage to him:

Ben Rice: 3-for-4, 2 HR, 2 RBI [+1.87 runs vs projection] season xwOBA 0.375, elite - real projection error
Ryan McMahon: 2-for-4, 1 HR, 5 RBI [+1.4 runs vs projection] season xwOBA 0.293, below average - a good day, not a model error
George Lombard Jr.: 1-for-5, 1 HR, 1 RBI [+0.93 runs vs projection] season xwOBA 0.291, below average - a good day, not a model error

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

HOME pitching

Gerrit Cole (NYY) threw 17 outs, 2 ER = 3.18 ERA against a projected 1.59 ERA (1.59 ERA error)

Bullpen behind him: 0 ER over 9 outs.

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

Who did the damage to him:

Cedric Mullins: 1-for-4, 1 HR, 1 RBI [+1.4 runs vs projection] season xwOBA 0.264, weak - a good day, not a model error
Jonny DeLuca: 3-for-4, 0 HR, 1 RBI [+1.25 runs vs projection] season xwOBA 0.257, weak - a good day, not a model error

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

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
TB	Mason Englert (Starter)	11 / 15	19.64 / 5.4 (+14.24)	8 / 3	9 / 9	4 / 0	1 / 2	1.05	6 / 8	4.9	venue away: 5.91 vs 7.36 ERA (ok) · opp run factor 1.02 (strong) · 14 prior appearances this season
NYY	Gerrit Cole (Starter)	17 / 17	3.18 / 1.59	2 / 1	3 / 2	1 / 1	3 / 5	0.58	8 / 5	5.84	venue home: 3.93 vs 3.41 ERA (worse here) · opp run factor 1.02 (average) · 21 prior appearances this season
NYY	Brent Headrick (Setup Man)	1 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.36	0 / 1	1.43
NYY	David Bednar (Closer)	3 / 5	0 / 0	0 / 0	0 / 2	0 / 0	0 / 0	0.18	1 / 2	1.2
NYY	Paul Blackburn (Middle Reliever (High Leverage))	3 / 2	0 / 27 (-27.00)	0 / 2	1 / 5	0 / 0	0 / 0	0.43	0 / 0	0.91
NYY	Tim Hill (Middle Reliever (High Leverage))	2 / 0	0 / n/a	0 / 2	0 / 2	0 / 0	0 / 0	0.23	0 / 1	0.51

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Ben Rice (DH)	NYY	4 / 4	3 / 1	0 / 0	0 / 0	2 / 1	2 / 1	2 / 1	1 / 1	0 / 1	1.87	0.375 elite	projection too low
Junior Caminero (3B)	TB	2 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	0 / 1	-1.73	0.361 elite	bad day
Cedric Mullins (CF)	TB	4 / 3	1 / 0	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	0 / 1	1.4	0.264 weak	good day
Ryan McMahon (3B)	NYY	4 / 3	2 / 1	0 / 0	0 / 0	1 / 0	5 / 0	1 / 0	0 / 0	1 / 1	1.4	0.293 below average	good day
Jonny DeLuca (RF)	TB	4 / 4	3 / 1	1 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	1 / 1	1.25	0.257 weak	good day
Jonathan Aranda (DH)	TB	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-1.11	0.345 elite	bad day
George Lombard Jr. (SS)	NYY	5 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 1	0 / 0	4 / 1	0.93	0.291 below average	good day
Ryan Vilade (1B)	TB	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.333 above average	—
Heliot Ramos (RF)	NYY	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	1 / 1	0.47	0.327 above average	—
Ali Sánchez (C)	NYY	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.47	0.244 weak	—
Richie Palacios (2B)	TB	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	-0.33	0.309 average	—
Liam Hicks (C)	TB	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	2 / 0	0 / 0	0.19	0.319 average	—
Victor Mesa Jr. (LF)	TB	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 1	-0.14	0.334 above average	—
José Caballero (LF)	NYY	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 1	-0.14	0.28 weak	—
Spencer Jones (CF)	NYY	5 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 1	2 / 2	-0.02	0.325 above average	—
Luis García Jr. (1B)	NYY	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	2 / 1	0.02	0.341 elite	—
Jorge Mateo (SS)	TB	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	3 / 1	0	0.278 weak	—
Jazz Chisholm Jr. (2B)	NYY	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0	0.295 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.064	positive favors home; negative favors away
Recent form	-0.714	positive favors home; negative favors away
Starting-pitcher matchup	0.004	positive favors home; negative favors away
Composite	-0.385	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.88-5.27	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.505	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	n/a	Capped at max 12 realistic runs
simulation_result	5-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.88-5.27	4-5	0.389	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	5-4	2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	5-4	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	5-4	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	5-4	-0	Capped at max 12 realistic runs
simulation_result	5-4	5-4	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
TB	Jorge Mateo	batter / SS	better	predicted_RBI 2.85
TB	Victor Mesa Jr.	batter / LF	worse	predicted_AB 2.746
NYY	Ali Sánchez	batter / C	better	predicted_AB 2.5
TB	Ryan Vilade	batter / 1B	better	predicted_H 2.321
TB	Jonathan Aranda	batter / DH	worse	predicted_AB 2.24
NYY	Luis García Jr.	batter / 1B	worse	predicted_AB 2.07
NYY	Ryan McMahon	batter / 3B	worse	predicted_SO 2.064
TB	Alex Cook	pitcher / Reliever	worse	predicted_H 2

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	NYY pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	NYY
info	CURRENT_PITCHER_DUPLICATION	TB pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	TB
info	CURRENT_FINALIZED_MATCH_BOX	NYY model-prediction batting totals differ from final simulation box score (AB: current=31, sim=45; H: current=8, sim=12; 3B: current=0, sim=1; HR: current=1, sim=0; BB: current=3, sim=5; SO: current=10, sim=14).	NYY
info	CURRENT_PLAYER_MATCH_BOX	NYY current finalized hitter lines are model predictions and differ from the simulation box score: Heliot Ramos: AB current=3 box=5, R current=1 box=0, BB current=0 box=1, SO current=1 box=2; George Lombard Jr.: AB current=4 box=6, H current=1 box=2, RBI current=1 box=0, SO current=1 box=2; Spencer Jones: AB current=4 box=5, 3B current=0 box=1, R current=0 box=1; José Caballero: AB current=3 box=5, H current=1 box=2; Luis García Jr.: AB current=4 box=6, H current=1 box=2, 2B current=1 box=0, RBI current=1 box=0; Ben Rice: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, HR current=1 box=0; Jazz Chisholm Jr.: AB current=3 box=4, H current=1 box=0, R current=1 box=0; Ryan McMahon: AB current=3 box=4, H current=1 box=0, BB current=0 box=1, SO current=1 box=3	NYY
info	CURRENT_FINALIZED_MATCH_BOX	TB model-prediction batting totals differ from final simulation box score (AB: current=31, sim=44; H: current=8, sim=11; 2B: current=1, sim=3; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=5; SO: current=8, sim=9; HBP: current=0, sim=1).	TB
info	CURRENT_PLAYER_MATCH_BOX	TB current finalized hitter lines are model predictions and differ from the simulation box score: Jonathan Aranda: AB current=4 box=6, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Victor Mesa Jr.: AB current=4 box=6, H current=1 box=0, R current=1 box=0, SO current=1 box=2; Richie Palacios: AB current=3 box=4, H current=1 box=0, SO current=1 box=2; Ryan Vilade: AB current=3 box=4, H current=1 box=3, HR current=0 box=1, RBI current=0 box=2; Jonny DeLuca: AB current=4 box=5, 2B current=0 box=1, RBI current=1 box=0, R current=0 box=1; Liam Hicks: AB current=3 box=5, RBI current=1 box=0, R current=0 box=1, BB current=0 box=1; Junior Caminero: AB current=4 box=5, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Cedric Mullins: AB current=3 box=5, H current=0 box=2, 2B current=0 box=1, R current=0 box=1	TB
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

TOR@BALHIT

Predicted: BAL 4-5 (conf +0.01, toss_up)Actual: BAL 2-4Brier 0.245total bias +3

Root cause: Toss-up that went our way

Evidence-backed diagnosis

published system edge 0.01 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 3; total bias +3
pregame home-minus-away evidence: season +0.077, recent +0.014, pitcher +0.079, composite +0.400
confidence pipeline reached 0.507 before later publication/recency controls; published home probability was 0.505
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 3 (9 vs 6)
Chris Bassitt (BAL) allowed 3 fewer ER than projected (0 vs 3)

Team	Starter	Projected	Actual	ER err
TOR	Max Scherzer	3 ER / 15 outs	4 ER / 15 outs	1
BAL	Chris Bassitt	3 ER / 15 outs	0 ER / 20 outs	-3

Decision internals: run factors home×1.089 / away×1.01; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

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

Top deviations that drove the game:

Andrés Giménez (TOR SS): 2-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 1.09]
Kazuma Okamoto (TOR 3B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.07]

Pitcher versus batter: what actually broke this game

AWAY pitching

Max Scherzer (TOR) threw 15 outs, 4 ER = 7.2 ERA against a projected 5.4 ERA (1.8 ERA error)

Bullpen behind him: 0 ER over 0 outs.

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

Who did the damage to him:

Leody Taveras: 1-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.27, weak - a good day, not a model error

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

HOME pitching

Chris Bassitt (BAL) threw 15 outs, 2 ER = 3.6 ERA against a projected 5.4 ERA (-1.8 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The AWAY lineup he faced produced 2 runs on 9 hits and 0 HR, against a projected 3 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

Andrés Giménez: 2-for-4, 0 HR, 0 RBI [+1.09 runs vs projection] season xwOBA 0.273, weak - a good day, not a model error
Ernie Clement: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.26, weak - a good day, not a model error

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

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
TOR	Max Scherzer (Starter)	15 / 15	7.2 / 5.4	4 / 3	4 / 7	1 / 1	3 / 1	1.77	7 / 3	3.57	venue away: 5.7 vs 6.25 ERA (ok) · opp run factor 1.09 (average) · 16 prior appearances this season
BAL	Chris Bassitt (Starter)	15 / 15	3.6 / 5.4	2 / 3	7 / 6	0 / 0	1 / 0	2.04	2 / 8	5.59	venue home: 3.54 vs 5.53 ERA (ok) · opp run factor 1.01 (weak) · 19 prior appearances this season
BAL	Andrew Kittredge (Closer)	3 / 5	0 / 5.4 (-5.40)	0 / 1	1 / 2	0 / 0	0 / 1	0.18	1 / 0	0.98
BAL	Rico Garcia (Setup Man)	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.22	0 / 1	0.8

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Andrés Giménez (SS)	TOR	4 / 3	2 / 1	2 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	1.09	0.273 weak	good day
Kazuma Okamoto (3B)	TOR	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 0	0 / 1	-1.07	0.328 above average	bad day
Sean Keys (1B)	TOR	1 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	-0.47	0.26 weak	—
Ernie Clement (2B)	TOR	4 / 3	2 / 1	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.47	0.26 weak	—
Coby Mayo (DH)	BAL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	2 / 1	-0.47	0.336 above average	—
Gunnar Henderson (SS)	BAL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	2 / 1	-0.47	0.323 above average	—
Leody Taveras (RF)	BAL	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.47	0.27 weak	—
Josh Smith (LF)	TOR	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	-0.33	0.281 weak	—
George Springer (DH)	TOR	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	1 / 1	-0.33	0.304 average	—
Dylan Beavers (LF)	BAL	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	1 / 1	1 / 1	-0.31	0.293 below average	—
Brandon Valenzuela (C)	TOR	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0 / 1	0.14	0.312 average	—
Colton Cowser (CF)	BAL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	2 / 1	-0.14	0.314 average	—
Nathan Lukes (RF)	TOR	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0	0.289 below average	—
Myles Straw (CF)	TOR	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0	0.296 below average	—
Christian Encarnacion-Strand (3B)	BAL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	0 / 1	0	0.334 above average	—
Pete Alonso (1B)	BAL	3 / 4	1 / 1	0 / 0	0 / 0	1 / 1	2 / 1	1 / 1	1 / 1	2 / 1	0	0.397 elite	—
Jeremiah Jackson (2B)	BAL	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0	0.283 below average	—
Carlos Narváez (C)	BAL	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.26 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.077	positive favors home; negative favors away
Recent form	0.014	positive favors home; negative favors away
Starting-pitcher matchup	0.079	positive favors home; negative favors away
Composite	0.4	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.54-4.5	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.505	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	n/a	Capped at max 12 realistic runs
simulation_result	4-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.54-4.5	5-4	-0.958	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	4-5	2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	4-5	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	4-5	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	4-5	0	Capped at max 12 realistic runs
simulation_result	4-5	4-5	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
BAL	Jeremiah Jackson	batter / 2B	better	predicted_RBI 1.89
TOR	Brandon Valenzuela	batter / C	boosted	predicted_2B 1.882
BAL	Carlos Narváez	batter / C	worse	predicted_H 1.82
BAL	Gunnar Henderson	batter / SS	worse	predicted_H 1.7
TOR	Myles Straw	batter / CF	worse	predicted_AB 1.5
BAL	Colton Cowser	batter / CF	worse	predicted_H 1.44
TOR	Josh Smith	batter / LF	worse	predicted_H 1.417
TOR	Sean Keys	batter / 1B	worse	predicted_AB 1.316

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	BAL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	BAL
info	CURRENT_PITCHER_DUPLICATION	TOR pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	TOR
info	CURRENT_FINALIZED_MATCH_BOX	BAL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=7, sim=11; 2B: current=1, sim=2; HR: current=1, sim=2; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=1; SO: current=9, sim=4).	BAL
info	CURRENT_PLAYER_MATCH_BOX	BAL current finalized hitter lines are model predictions and differ from the simulation box score: Coby Mayo: AB current=4 box=3, BB current=0 box=1, SO current=1 box=0; Christian Encarnacion-Strand: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=0; Pete Alonso: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Gunnar Henderson: H current=1 box=3, 2B current=0 box=1, BB current=1 box=0, SO current=1 box=0; Dylan Beavers: 2B current=1 box=0, R current=0 box=1, BB current=1 box=0, SO current=1 box=0; Leody Taveras: AB current=3 box=4; Jeremiah Jackson: H current=1 box=2, HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Carlos Narváez: AB current=3 box=4, H current=0 box=2, 2B current=0 box=1, SO current=1 box=0	BAL
info	CURRENT_FINALIZED_MATCH_BOX	TOR model-prediction batting totals differ from final simulation box score (AB: current=30, sim=34; 2B: current=1, sim=2; HR: current=1, sim=0; R: current=3, sim=4; BB: current=3, sim=1; SO: current=6, sim=10; SF: current=0, sim=1).	TOR
info	CURRENT_PLAYER_MATCH_BOX	TOR current finalized hitter lines are model predictions and differ from the simulation box score: Josh Smith: H current=0 box=2, RBI current=0 box=1, R current=0 box=1, BB current=1 box=0; Sean Keys: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0; Nathan Lukes: RBI current=0 box=1, SO current=0 box=1; Andrés Giménez: H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Brandon Valenzuela: AB current=3 box=4, H current=1 box=2, 2B current=0 box=2, RBI current=0 box=1; Ernie Clement: AB current=3 box=4, H current=1 box=2, 2B current=1 box=0, RBI current=0 box=1; George Springer: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Kazuma Okamoto: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0	TOR
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

TOR@BALMISS

Predicted: TOR 6-5 (conf -0.02, toss_up)Actual: BAL 2-4Brier 0.262total bias +5

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

Evidence-backed diagnosis

published system edge 0.02 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 5; total bias +5
pregame home-minus-away evidence: season +0.077, recent +0.014, pitcher +0.118, composite +0.126
confidence pipeline reached 0.526 before later publication/recency controls; published home probability was 0.488
score-volatility actions: no_change
forecast-file integrity: 2 structural defect(s) on the saved artifact; see the per-game file-integrity table (this is a file check, not why the pick missed)

Quantified contributing factors

total runs over-projected by 5 (11 vs 6)
TOR offense over-projected by 4

Decision internals: run factors home×1.089 / away×0.971; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

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

Top deviations that drove the game:

Pete Alonso (BAL 1B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Kazuma Okamoto (TOR 3B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Christian Encarnacion-Strand (BAL 3B): 2-for-3, 1 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Dylan Beavers (BAL LF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]

Starter miss: Trey Gibson (BAL) posted 0.00 ERA vs projected 13.50 (0 ER in 20 outs); he is materially worse pitching home (9.35 ERA there vs 5.27 on the other side), a venue-role weakness the model under-weighted.

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

Pitcher versus batter: what actually broke this game

AWAY pitching

No settled pitching line for this club. The forecast named Max Scherzer (TOR) at a projected 7.2 ERA, but the per-player feed carries no pitching row for him in this game, so the actual-versus-projected comparison cannot be made on this side. That is a source-coverage gap in the settled box, not a forecasting failure.

Bullpen behind him: 0 ER over 0 outs.

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

Who did the damage to him:

Christian Encarnacion-Strand: 2-for-3, 1 HR, 3 RBI [+1.4 runs vs projection] season xwOBA 0.334, above average - real projection error
Coby Mayo: 1-for-3, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.336, above average - a good day, not a model error
Colton Cowser: 1-for-3, 0 HR, 1 RBI [+0.31 runs vs projection] season xwOBA 0.314, average - a good day, not a model error

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

HOME pitching

Trey Gibson (BAL) threw 20 outs, 0 ER = 0 ERA against a projected 13.5 ERA (-13.5 ERA error) only 8 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 0 ER over 4 outs.

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

Who did the damage to him:

Nathan Lukes: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.289, below average - a good day, not a model error

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

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
BAL	Trey Gibson (Starter)	20 / 12	0 / 13.5 (-13.50)	0 / 6	3 / 6	0 / 0	2 / 2	3.33	7 / 2	5.59	venue home: 9.35 vs 5.27 ERA (worse here) · opp run factor 0.97 (weak) · 8 prior appearances this season
BAL	Josh Walker (Middle Reliever (High Leverage))	3 / 5	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	0.33	0 / 1	1.1
BAL	Yennier Cano (Middle Reliever (High Leverage))	1 / 4	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0.14	1 / 1	0.69

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Pete Alonso (1B)	BAL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	1 / 1	-1.73	0.397 elite	bad day
Kazuma Okamoto (3B)	TOR	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	2 / 1	-1.4	0.328 above average	bad day
Christian Encarnacion-Strand (3B)	BAL	3 / 4	2 / 1	0 / 0	0 / 0	1 / 0	3 / 1	2 / 1	0 / 0	0 / 1	1.4	0.334 above average	projection too low
Dylan Beavers (LF)	BAL	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	2 / 1	-1.11	0.293 below average	projection too high
Nathan Lukes (RF)	TOR	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	0.47	0.289 below average	—
Ernie Clement (2B)	TOR	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	-0.47	0.26 weak	—
Gunnar Henderson (SS)	BAL	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	-0.47	0.323 above average	—
Jeremiah Jackson (2B)	BAL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.283 below average	—
Coby Mayo (DH)	BAL	3 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 0	0 / 1	0.31	0.336 above average	—
Colton Cowser (CF)	BAL	3 / 3	1 / 1	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0 / 1	0.31	0.314 average	—
Josh Smith (LF)	TOR	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.281 weak	—
Myles Straw (CF)	TOR	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	0	0.296 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.077	positive favors home; negative favors away
Recent form	0.014	positive favors home; negative favors away
Starting-pitcher matchup	0.118	positive favors home; negative favors away
Composite	0.126	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.79-3.85	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-5	0.512	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	6-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-5	n/a	Capped at max 12 realistic runs
simulation_result	6-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.79-3.85	5-4	0.059	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	6-5	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	6-5	6-5	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-5	6-5	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-5	6-5	-0	Capped at max 12 realistic runs
simulation_result	6-5	6-5	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
TOR	Myles Straw	batter / CF	better	predicted_RBI 3.86
TOR	Brandon Valenzuela	batter / C	worse	predicted_H 2.73
BAL	Dylan Beavers	batter / LF	better	predicted_RBI 2.67
BAL	Christian Encarnacion-Strand	batter / 3B	better	predicted_H 2.19
BAL	Coby Mayo	batter / DH	worse	predicted_SO 1.917
TOR	Vladimir Guerrero Jr.	batter / 1B	boosted	predicted_RBI 1.66
TOR	Kazuma Okamoto	batter / 3B	worse	predicted_AB -1.65
BAL	Gunnar Henderson	batter / SS	worse	predicted_BB 1.576

Structural audit findings

Severity	Rule	Finding	Subject
warning	INNING_OUTS	Inning 9 home has outs=0; expected 3 for a played half inning.
warning	PITCH_COUNT_PITCHER_MATCH	TOR pitch_count row for Max Scherzer differs from pitching box score: batters_faced/BF: summary=24, box=21.	TOR Max Scherzer
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	TOR pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	TOR
info	CURRENT_FINALIZED_MATCH_BOX	BAL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=34; H: current=7, sim=10; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=7; SO: current=9, sim=7).	BAL
info	CURRENT_PLAYER_MATCH_BOX	BAL current finalized hitter lines are model predictions and differ from the simulation box score: Coby Mayo: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=0 box=1; Christian Encarnacion-Strand: H current=1 box=3, BB current=0 box=1, SO current=1 box=0; Gunnar Henderson: AB current=4 box=3, H current=1 box=2, BB current=1 box=2, SO current=1 box=0; Dylan Beavers: AB current=4 box=5, 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=3; Leody Taveras: H current=0 box=1, RBI current=0 box=1, R current=0 box=1, BB current=0 box=1; Pete Alonso: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Colton Cowser: R current=0 box=1, BB current=0 box=1, SO current=1 box=2; Carlos Narváez: AB current=3 box=4, H current=0 box=1, 2B current=0 box=1, SO current=1 box=0	BAL
info	CURRENT_FINALIZED_MATCH_BOX	TOR model-prediction batting totals differ from final simulation box score (AB: current=31, sim=36; H: current=8, sim=9; 2B: current=1, sim=4; HR: current=1, sim=0; RBI: current=4, sim=6; R: current=3, sim=6; BB: current=3, sim=2; HBP: current=0, sim=1).	TOR
info	CURRENT_PLAYER_MATCH_BOX	TOR current finalized hitter lines are model predictions and differ from the simulation box score: Kazuma Okamoto: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Josh Smith: H current=0 box=1, R current=0 box=1, BB current=0 box=1, SO current=1 box=0; George Springer: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Vladimir Guerrero Jr.: AB current=4 box=5, 2B current=0 box=1, RBI current=0 box=2, BB current=1 box=0; Nathan Lukes: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0; Andrés Giménez: AB current=3 box=4, H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0; Brandon Valenzuela: AB current=3 box=4, H current=1 box=3, R current=0 box=1, BB current=1 box=0; Myles Straw: AB current=3 box=4, H current=1 box=2, 2B current=0 box=2, RBI current=0 box=4	TOR
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

WAS@DETHIT

Predicted: WAS 6-4 (conf -0.01, toss_up)Actual: WAS 4-2Brier 0.245total bias +4

Root cause: Toss-up that went our way

Evidence-backed diagnosis

published system edge 0.01 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 4; total bias +4
pregame home-minus-away evidence: season +0.217, recent -0.156, pitcher -0.098, composite -0.494
confidence pipeline reached 0.507 before later publication/recency controls; published home probability was 0.495
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

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

Team	Starter	Projected	Actual	ER err
WAS	Richard Lovelady	2 ER / 11 outs	0 ER / 4 outs	-2
DET	Framber Valdez	2 ER / 18 outs	2 ER / 16 outs	0

Decision internals: run factors home×0.994 / away×1.092; 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:

Ben Malgeri (DET RF): 2-for-5, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.07]
Dillon Dingler (DET C): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.07]
Yohandy Morales (WAS 1B): 1-for-4, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]
Kevin McGonigle (DET 3B): 3-for-5, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.92]
James Wood (WAS RF): 0-for-4, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Starter miss: Richard Lovelady (WAS) posted 0.00 ERA vs projected 4.91 (0 ER in 4 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

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

Pitcher versus batter: what actually broke this game

AWAY pitching

Richard Lovelady (WAS) threw 4 outs, 0 ER = 0 ERA against a projected 4.91 ERA (-4.91 ERA error)

Bullpen behind him: 1 ER over 17 outs.

The HOME lineup he faced produced 1 runs on 7 hits and 1 HR, against a projected 4 runs / 8 hits (-3 runs vs projection).

Who did the damage to him:

Ben Malgeri: 2-for-5, 1 HR, 2 RBI [+1.07 runs vs projection] season xwOBA 0.315, average - a good day, not a model error
Kevin McGonigle: 3-for-5, 0 HR, 0 RBI [+0.92 runs vs projection] season xwOBA 0.357, elite - real projection error
Max Clark: 1-for-2, 0 HR, 0 RBI [+0.64 runs vs projection] season xwOBA 0.326, above average - real projection error

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

HOME pitching

Framber Valdez (DET) threw 16 outs, 2 ER = 3.38 ERA against a projected 3 ERA (0.38 ERA error)

Bullpen behind him: 2 ER over 11 outs.

The AWAY lineup he faced produced 4 runs on 6 hits and 1 HR, against a projected 5 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

Yohandy Morales: 1-for-4, 1 HR, 2 RBI [+0.93 runs vs projection] season xwOBA 0.331, above average - real projection error
Nasim Nuñez: 1-for-2, 0 HR, 0 RBI [+0.8 runs vs projection] season xwOBA 0.266, weak - a good day, not a model error
Brady House: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.334, above average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
WAS	Richard Lovelady (Starter)	4 / 11	0 / 4.91 (-4.91)	0 / 2	1 / 3	0 / 1	1 / 0	2.31	0 / 8	4.05	venue away: 0 vs 5.4 ERA (ok) · opp run factor 0.99 (average) · 36 prior appearances this season
WAS	Brad Lord (Setup Man)	10 / 6	0 / 0	0 / 0	1 / 0	0 / 0	2 / 0	0.57	3 / 1	1.42
WAS	Erik Tolman (Closer)	2 / 5	0 / 5.4 (-5.40)	0 / 1	0 / 1	0 / 0	2 / 1	0.49	0 / 3	1.26
WAS	Yovanny Cruz (Middle Reliever (High Leverage))	4 / 2	6.75 / 0 (+6.75)	1 / 0	3 / 0	0 / 0	0 / 1	0.7	2 / 2	0.85
WAS	Jack Sinclair (Reliever)	1 / 0	0 / n/a	0 / 0	0 / 1	0 / 0	0 / 1	n/a	0 / 2	n/a
DET	Framber Valdez (Starter)	16 / 18	3.38 / 3	2 / 2	4 / 2	0 / 0	3 / 4	2.33	6 / 7	3.99	venue home: 4 vs 4.1 ERA (ok) · opp run factor 1.09 (strong) · 30 prior appearances this season
DET	Kenley Jansen (Closer)	3 / 4	0 / 27 (-27.00)	0 / 4	0 / 5	0 / 0	0 / 0	0.56	0 / 2	1.15
DET	Jacob Waguespack (Middle Reliever (High Leverage))	6 / 0	4.5 / n/a	1 / 0	1 / 0	0 / 0	1 / 1	0.5	2 / 5	1.12
DET	Tyler Kinley (Middle Reliever)	2 / 0	13.5 / n/a	1 / 0	1 / 0	1 / 0	0 / 0	0.47	0 / 0	0.68

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

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Ben Malgeri (RF)	DET	5 / 3	2 / 1	0 / 0	0 / 0	1 / 0	2 / 0	1 / 0	0 / 1	0 / 1	1.07	0.315 average	no signal
Dillon Dingler (C)	DET	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	0 / 1	-1.07	0.34 elite	bad day
Yohandy Morales (1B)	WAS	4 / 3	1 / 1	0 / 0	0 / 0	1 / 0	2 / 0	1 / 1	0 / 0	1 / 1	0.93	0.331 above average	projection too low
Kevin McGonigle (3B)	DET	5 / 4	3 / 1	1 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	0.92	0.357 elite	projection too low
James Wood (RF)	WAS	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 1	2 / 1	-0.8	0.406 elite	bad day
Nasim Nuñez (SS)	WAS	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 1	0.8	0.266 weak	good day
Spencer Torkelson (1B)	DET	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	4 / 1	-0.8	0.315 average	no signal
Andrés Chaparro (DH)	WAS	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	1 / 0	0 / 1	2 / 1	-0.64	0.343 elite	bad day
Max Clark (CF)	DET	2 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 1	0.64	0.326 above average	projection too low
Dylan Crews (LF)	WAS	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	-0.47	0.318 average	—
Harry Ford (C)	WAS	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	-0.47	0.288 below average	—
Brady House (3B)	WAS	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	1 / 1	0.47	0.334 above average	—
Jacob Young (CF)	WAS	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	-0.47	0.315 average	—
Zach McKinstry (2B)	DET	1 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0.33	0.283 below average	—
Hao-Yu Lee (DH)	DET	5 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.31	0.304 average	—
CJ Abrams (2B)	WAS	2 / 4	1 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	2 / 0	0 / 1	-0.27	0.331 above average	—
Riley Greene (LF)	DET	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	0 / 1	-0.14	0.373 elite	—
John Peck (SS)	DET	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	-0.14	0.326 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.217	positive favors home; negative favors away
Recent form	-0.156	positive favors home; negative favors away
Starting-pitcher matchup	-0.098	positive favors home; negative favors away
Composite	-0.494	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.04-4.64	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.505	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, starter_quality_favours_the_other_side, matchup_recency_guardrail)
embedded_savant_run_environment	6-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-4	n/a	Capped at max 12 realistic runs
simulation_result	6-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.04-4.64	4-5	-0.404	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	5-4	2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, starter_quality_favours_the_other_side, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	6-4	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-4	6-4	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-4	6-4	-0	Capped at max 12 realistic runs
simulation_result	6-4	6-4	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

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

Team	Player	Role	Net effect	Largest change
WAS	Yohandy Morales	batter / 1B	worse	predicted_AB 3.214
DET	Hao-Yu Lee	batter / DH	better	predicted_SO 3.194
WAS	Dylan Crews	batter / LF	worse	predicted_SO 3
DET	Riley Greene	batter / LF	better	predicted_H 2.95
DET	Spencer Torkelson	batter / 1B	worse	predicted_AB 2.651
DET	Ben Malgeri	batter / RF	better	predicted_AB 2.5
WAS	Andrés Chaparro	batter / DH	better	predicted_AB 2.438
DET	Kevin McGonigle	batter / 3B	worse	predicted_SO 2.38

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	DET pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	DET
info	CURRENT_PITCHER_DUPLICATION	WAS pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	WAS
info	CURRENT_FINALIZED_MATCH_BOX	DET model-prediction batting totals differ from final simulation box score (AB: current=31, sim=48; H: current=8, sim=9; 2B: current=1, sim=4; HR: current=1, sim=2; SO: current=8, sim=17).	DET
info	CURRENT_PLAYER_MATCH_BOX	DET current finalized hitter lines are model predictions and differ from the simulation box score: Kevin McGonigle: AB current=4 box=6, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Hao-Yu Lee: AB current=4 box=6, 2B current=1 box=0, HR current=0 box=1, R current=0 box=1; Ben Malgeri: AB current=3 box=5, H current=1 box=2, 2B current=0 box=1, R current=0 box=1; Max Clark: AB current=3 box=6, H current=1 box=0, R current=1 box=0; Dillon Dingler: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Zach McKinstry: AB current=3 box=4, BB current=0 box=1, SO current=0 box=2; Spencer Torkelson: AB current=3 box=6, HR current=0 box=1, RBI current=0 box=2, R current=0 box=1; Riley Greene: AB current=4 box=5, H current=1 box=4, 2B current=0 box=3, SO current=1 box=0	DET
info	CURRENT_FINALIZED_MATCH_BOX	WAS model-prediction batting totals differ from final simulation box score (AB: current=32, sim=46; H: current=8, sim=7; 3B: current=0, sim=1; HR: current=1, sim=0; RBI: current=4, sim=6; R: current=5, sim=6; BB: current=3, sim=5; SO: current=8, sim=16).	WAS
info	CURRENT_PLAYER_MATCH_BOX	WAS current finalized hitter lines are model predictions and differ from the simulation box score: Andrés Chaparro: AB current=4 box=5, 2B current=1 box=0, RBI current=1 box=0, R current=0 box=1; CJ Abrams: AB current=4 box=6, 3B current=0 box=1, HR current=1 box=0, RBI current=1 box=3; Dylan Crews: AB current=4 box=5, H current=1 box=0, R current=1 box=0, BB current=0 box=1; Yohandy Morales: AB current=3 box=6, 2B current=0 box=1, R current=1 box=0, SO current=1 box=2; James Wood: AB current=4 box=5, SO current=1 box=0; Nasim Nuñez: AB current=3 box=5, SO current=1 box=3; Harry Ford: AB current=3 box=4, SO current=1 box=2; Brady House: AB current=4 box=5, R current=0 box=1, BB current=0 box=1, SO current=1 box=2	WAS
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 20260922. Last recorded change: v4.30.0 (2026-09-08) Hitter shrinkage is derived from the between-hitter spread rather than fitted to slate error, so an established extreme profile is no longer flattened toward the league.

Metric	Prior	Current	Δ	Result
brier_score	0.2448	0.2442	-0.0006	improved
accuracy	0.563	0.5546	-0.0084	regressed
team_run_mae	2.4958	2.5168	+0.021	regressed
total_run_mae	3.3109	3.3697	+0.0588	regressed
calibration_error	0.1034	0.101	-0.0024	improved
starter_er_mae	1.712	1.756	+0.044	regressed
starter_outs_mae	3.516	3.572	+0.056	regressed

Deep slate report — 20260923

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 (16)

ARI@COLHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

CHW@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

CIN@ATLHIT

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@BOSMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

HOU@SEAHIT

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@OAKMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

MIA@CHCMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

MIL@PHIHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

MIN@SFHIT

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@TEXHIT

Player-level breakdown (projected vs actual)