Luna deep analysis — slate 20260921

Coverage receipt: COMPLETE · 9/8 sections; 3/3 games; 7/7 causes; 6/6 sources

1. Evidence receipt and gaps

Sources received

All manifest entries were checked against the supplied evidence:

SourceStatusPurpose
prediction review — /app/output/reports/prediction_review_20260921.jsonLoaded, 79,110 bytesOfficial outcomes and scoring errors
forensic deep audit — /app/output/reports/PREDICTION_SYSTEM_DEEP_AUDIT_20260921.jsonLoaded, 60,154 bytesProbability, provenance, starter, and decision contracts
continuous learning — /app/output/learning/continuous_learning_daily_20260921.jsonLoaded, 473,280 bytesStage quality, calibration, lineage, score distribution, and starter comparisons
detailed report — /app/output/reports/detailed_20260921.htmlLoaded, 76,436 bytesFull game and simulation rendering
ultra detailed player audit — /app/output/reports/20260921_ultra_detailed_player_audit.htmlLoaded, 6,405,727 bytesPlayer-level prediction trace
structural audit — /app/output/reports/audit_report_20260921.htmlLoaded, 362,566 bytesInvariant findings
integrity revision audit — /app/output/reports/audit_report_20260921_integrity_revision.htmlMissingPost-Luna remediation proof for immutable revision artifacts

The missing integrity-revision audit is a material evidence gap for remediation verification. It does not prevent review of the game outcomes or the supplied structural findings, but it prevents confirming that a corrected immutable artifact was generated and validated.

Games reviewed

Coverage is complete for the three-game slate:

The coverage matrix reports:

The slate was therefore game-complete, although not artifact-complete because the integrity-revision audit is missing.

Internal contradictions and limitations

  1. Published probability lineage is not uniformly reconciled.
  2. WAS@DET has a calibrated-to-published difference of -0.041591, with unexplained_material_delta: true.
  3. The causal register correctly labels this a critical probability-lineage failure.
  4. MIN@SF also says "transformation":"no recorded transformation explains the publication delta", although its delta is only -0.017297 and is marked non-material.
  5. TOR@BAL has a small published-minus-calibrated difference of -0.002043, while the record says the published value matches the calibrated/governed value. The numerical values are not identical, but the difference is small and not material under the stated threshold.
  1. Starter-line data are internally unusual in WAS@DET.
  2. River Ryan is recorded as a starter with 0.2 IP, nine outs, and a projected 0.2 IP; the structural audit flags this as an implausible starter workload.
  3. This is an artifact/data-contract defect, not proof that the winner prediction was wrong. The game winner was correctly predicted.
  1. The confidence terminology is potentially misleading.
  2. The MIN@SF record calls the home-minus-away edge -0.135, while the diagnostic describes a 0.14-confidence away commitment. This is directionally consistent: the away probability was approximately 0.568, but the edge is not close to certainty.
  3. A 0.12 home-away confidence edge must be treated as moderate, not near-even and not high conviction.
  1. No market cross-check was available.
  2. Moneyline source coverage: zero games.
  3. Therefore, no conclusion about model-versus-market disagreement can be made.
  1. The provided data do not include the exact text of all 22 informational audit findings.
  2. Their count is available, and the per-game audit summaries report 8, 7, and 7 informational findings, but their individual messages are not supplied.

2. Executive diagnosis

Outcome summary

The slate finished:

This is a poor slate, but the causes differ by game.

Winner accuracy

Winner accuracy was poor: 33.3%, with only WAS@DET correct.

The quality of the winner path was not stable:

The variance stage caused a measurable winner-direction regression:

The enhanced stage later recovered the WAS@DET direction, but that recovery does not erase the intermediate regression.

Calibration quality

Calibration was weak and overconfident on this slate:

The second bucket is especially unstable, but it contains only one game. The correct conclusion is that the slate exhibited poor calibration, not that a stable calibration curve has been proven.

MIN@SF was the clearest confidence problem:

TOR@BAL was essentially a toss-up:

WAS@DET was a correct lean:

A correct winner does not make the probability well calibrated; it only records that the selected side won.

Score accuracy

Score accuracy was materially poor:

The run models have almost no validated holdout signal:

The model therefore should not be treated as an exact-score system. The weak_run_model_signal finding is appropriately high severity, not critical: it is a serious modeling limitation, but not itself an evidence-contract failure.

Starter accuracy

Starter projection was the primary cause of MIN@SF:

Starter projection contributed materially to WAS@DET as well:

However, Detroit still won, and the extreme River Ryan line was favorable to the pick:

TOR@BAL did not show a decisive starter failure:

Artifact integrity

Artifact integrity is not clean:

This is a high-severity process/artifact issue, not proof that the model’s Detroit winner was wrong. The causal register correctly says not to attribute accuracy changes to the malformed file.

Causal issue register reconciliation

Critical: unexplained published probability delta

This is the only critical finding and is correctly classified as an evidence/contract failure.

High: artifact integrity failure

High: pipeline winner instability

High: calibration instability

Because this finding has since escalated to a three-slate streak, it now supports a bounded shadow experiment, discussed below.

High: weak run-model signal

Informational: complete audit coverage

Informational: simulation inertness

3. Every game

MIN@SF — wrong winner and wrong score

Prediction versus outcome

Errors

Starter projection versus actual

StarterProjectedActualError
Zebby Matthews, MIN1 ER, 17 outs, ERA 1.595 ER, 15 outs, ERA 9.00+4 ER, -2 outs, +7.41 ERA
Blade Tidwell, SF3 ER, 16 outs, ERA 5.061 ER, 10 outs, ERA 2.70-2 ER, -6 outs, -2.36 ERA

The principal starter failure was Matthews. The audit specifically identifies his venue weakness as credible and says the model underweighted it.

Pregame edges

The directional evidence favored Minnesota, but the edge was not large enough to justify treating the 0.14 confidence difference as near-certainty.

Controls

The no-change volatility decision did not protect against the eventual wide outcome. This is consistent with the broader finding that the score model’s tails are too compressed.

Score path

  1. Historical ML: MIN 4.697, SF 4.343
  2. Variance-applied deterministic target: MIN 5, SF 4
  3. Enhanced prediction: MIN 5, SF 4
  4. Embedded Savant environment: MIN 5, SF 4
  5. Away delta: +0.1928
  6. Home delta: -0.2595
  7. Fundamental edge review: unchanged
  8. Capped target: unchanged
  9. Simulation result: unchanged

The simulation did not provide an independent correction; it merely preserved the narrow target.

Player-level adjustments and outcome

There were 25 material player changes. Important listed adjustments included:

Important realized deviations:

The matchup verdict was:

Audit findings

Causes

TOR@BAL — toss-up miss and wrong score

Prediction versus outcome

Errors

This was a close, low-conviction game. It should not be grouped with the moderate-confidence MIN@SF miss.

Starter projection versus actual

StarterProjectedActualError
Trey Yesavage, TOR3 ER, 15 outs, ERA 5.401 ER, 6 outs, ERA 4.50-2 ER, -9 outs, -0.90 ERA
Shane Baz, BAL4 ER, 17 outs, ERA 6.353 ER, 20 outs, ERA 4.05-1 ER, +3 outs, -2.30 ERA

Both starters allowed fewer earned runs than projected. No decisive starter-direction failure is identified.

Pregame edges

The inputs conflicted: season quality favored Baltimore, while recent and pitcher edges leaned Toronto. The composite favored Toronto, but only weakly.

Recent form:

Controls

Score path

  1. Historical ML: TOR 4.297, BAL 4.059
  2. Variance-applied: TOR 4, BAL 4
  3. Enhanced prediction: TOR 5, BAL 4
  4. Embedded Savant environment: TOR 5, BAL 4
  5. Away delta: -0.2485
  6. Home delta: +0.0277
  7. Fundamental review: unchanged
  8. Capped target: unchanged
  9. Simulation result: unchanged

The model promoted Toronto from a near-even historical projection to a five-run target while maintaining a one-run margin. The actual game remained close but favored Baltimore.

Probability reversion

This game contains the only recorded blend-stage side reversal:

This is not automatically a defect, but it is a decision-path instability worth tracing alongside the broader winner-instability finding.

Player-level adjustments and outcome

There were 23 material player changes. Notable adjustments included:

Important realized contributions:

The forensic conclusion was a genuine toss-up, with no starter or one-sided team breakdown.

Audit findings

Causes

WAS@DET — correct winner, wrong score

Prediction versus outcome

This is the required example showing that winner correctness does not imply a good prediction. The winner was right, but the six-run team-score error and seven-run actual margin were badly missed.

Errors

Starter projection versus actual

StarterProjectedActualError
DJ Herz, WAS1 ER, 7 outs, ERA 3.864 ER, 12 outs, ERA 9.00+3 ER, +5 outs, +5.14 ERA
River Ryan, DET1 ER, 2 outs, ERA 13.500 ER, 9 outs, ERA 0.00-1 ER, +7 outs, -13.50 ERA

The Herz projection understated Detroit’s scoring opportunity. River Ryan’s actual performance was unusually good relative to the very low projected workload, but the record explicitly classifies this as an outlier and says no model change is warranted from this one start.

Pregame edges

The directional evidence consistently favored Detroit, albeit not with a large enough probability to support a blowout score.

Recent form details were not separately included in this game’s trace, but the model had a coherent positive home-side composite.

Controls

The unavailable starter experience and venue fields are important data gaps for this matchup, particularly because the actual starter outcomes were extreme.

Score path

  1. Historical ML: WAS 4.697, DET 4.343
  2. Variance-applied: WAS 4, DET 5
  3. Enhanced prediction: unchanged
  4. Embedded Savant environment: unchanged
  5. Fundamental review: unchanged
  6. Capped target: unchanged
  7. Simulation result: unchanged
  8. Published final: unchanged

The final score path never represented the possibility of Detroit winning by a wide margin.

Player-level adjustments and outcome

Important realized contributors:

The slate-level player audit also identifies Keibert Ruiz as a below-average hitter who was over-projected, with run impact -0.78.

Audit findings

Two structural errors were recorded:

  1. Current artifact:
  2. DET River Ryan has an implausible 0.2 IP starter outing
  3. Rule: PITCHING_REALISM
  4. Severity: error
  1. Simulation artifact:
  2. Same implausible 0.2 IP starter outing
  3. Rule: PITCHING_REALISM
  4. Severity: error

Summary:

The errors are presentation/data-contract defects. They should not be treated as proof that the Detroit winner was incorrectly modeled.

Causes

4. Miss reconstruction

MIN@SF

Earliest wrong-direction stage

The earliest evidence that moved the model toward the wrong result was already present in the pregame directional inputs:

Those inputs led to the away pick. The critical model error was not a later score-stage reversal; it was insufficiently discounting Minnesota’s vulnerable starter profile, especially Matthews’ credible away split.

The score pipeline then reinforced the wrong side:

Cause classification

The 0.14 edge is moderate. It does not mean Minnesota was a near-certain selection; it means the published probability was about 56.8%, which still implies a substantial loss probability.

TOR@BAL

Earliest wrong-direction stage

The historical stage favored Toronto slightly:

At variance_applied, the winner direction regressed to a tie:

The enhanced stage then restored Toronto:

The model therefore never found a strong evidence-based Baltimore direction. It ended as a one-run Toronto pick with a probability of only 0.505257.

Cause classification

The blend-stage reversal is a pipeline-quality concern:

This should be traced, but it is not evidence that Baltimore should simply have been selected.

Pipeline-stage quality across all stages and transitions

StageWinner accuracyWinner hitsTeam-run MAETotal-run MAEMargin MAE
Historical ML33.33%1/31.91531.87573.7913
Variance applied0%0/32.00002.00004.0000
Enhanced prediction33.33%1/32.00002.00004.0000
Embedded Savant environment33.33%1/32.00002.00004.0000
Fundamental edge review33.33%1/32.00002.00004.0000
Capped target33.33%1/32.00002.00004.0000
Simulation result33.33%1/32.00002.00004.0000
Published final33.33%1/32.00002.00004.0000

Transitions:

  1. Historical ML → variance applied
  2. Changed TOR → TIE
  3. Changed DET → TIE
  4. WAS@DET is recorded as the regression
  5. Winner accuracy: -0.3333
  6. Winner hits: -1
  7. Team-run MAE: +0.0847
  8. Margin MAE: +0.2087
  1. Variance applied → enhanced prediction
  2. TOR@BAL: TIE → TOR
  3. WAS@DET: TIE → DET
  4. Recovered one winner hit, including the Detroit pick
  5. No remaining regression relative to variance stage
  1. All later transitions
  2. No winner changes
  3. No run-error changes
  4. No score improvement

The simulation did not materially intervene. Its zero-flip behavior is not automatically a single-slate defect, but it is an established cross-slate diagnostic failure.

5. Cross-game counterfactuals

Lever tension entries

score.variance_multiplier / score.variance_floor

player.pitcher.v4 form weights / player.pitcher_prior_games

recent.* / season.* run-mean weighting

Tempting one-game fixes that must not be made

  1. Do not force Baltimore in future games merely because it won TOR@BAL.
  2. The game was a toss-up with a 0.505 Toronto probability and no decisive starter breakdown.
  1. Do not increase Detroit confidence simply because Detroit won by seven.
  2. The score was wrong by six runs, the artifact had two errors, and the large margin was not forecast.
  1. Do not tune directly to River Ryan’s result.
  2. The audit calls it an outlier with no credible explanatory signal.
  1. Do not globally increase starter pessimism or decrease starter projections.
  2. The slate-level starter mean suggests the opposite direction, but the 12-slate persistence test does not support a change.
  1. Do not use player-level deviations as broad training targets.
  2. Only 7 of 22 material deviations, or 31.8%, were supported by the hitter’s own season contact quality.
  1. Do not interpret zero simulation flips as proof that simulation is unnecessary.
  2. The cross-slate diagnosis says the simulation may be silently no-oping.
  1. Do not promote the slate or learn parameters from it while the critical probability-lineage and error-level artifact findings remain unresolved.

6. Data and contract defects

Critical findings

Unexplained probability publication delta

The data do not establish a governor bypass. They establish an unexplained material transformation.

Error findings

Two error-level findings, both in WAS@DET:

  1. Current artifact flags River Ryan’s 0.2 IP starter outing as implausible.
  2. Simulation artifact flags the same 0.2 IP starter outing.

These are artifact/presentation or data-contract defects. They are not direct evidence of an incorrect Detroit winner.

Warning findings

Informational findings

Coverage reconciliation

The slate is not “clean” merely because coverage is complete.

Positive coverage facts:

Negative integrity facts:

Probability-lineage audit

MIN@SF

This is not a material bypass under the registered threshold, but the transformation record is incomplete.

TOR@BAL

The numerical difference is small, but the record should still state whether the score-coherence value or another final-stage transformation produced 0.505257.

WAS@DET

This is the only clear material lineage bypass/gap in the supplied evidence.

Presentation/artifact defects versus model defects

Model/process defects

Artifact/presentation defects

These categories must remain separate.

7. Cross-slate persistence

A single slate is insufficient for many inferences, but the escalated findings are explicitly established failures and must not be dismissed as single-slate noise.

Raw structural blocks

Margin distribution

Current slate:

This is consistent with margin compression.

Pipeline-stage quality and transitions

The current slate shows a material intermediate regression:

This confirms that the pipeline instability finding is present now.

Calibration monitoring

Current slate:

High-raw cluster

Simulation control

Finding streaks

FindingConsecutive-slate countPresent now?Interpretation
high_raw_probability_cluster0NoNot persistent now; do not tune from this slate
pipeline_winner_instability4YesEstablished recurring winner-direction regression
score_margin_compression0NoCurrent slate shows compression, but the registered streak is not active
simulation_inert_on_slate11YesEstablished recurring simulation failure
single_slate_calibration_instability3YesPersistent enough for a bounded shadow experiment
weak_run_model_signal11YesEstablished structural modeling failure

Escalated findings

Pipeline winner instability

Simulation inertness

Calibration instability

Weak run-model signal

Metric trends

Each supplied trend direction:

The top-confidence-pick loss streak is:

8. Adjustments for today

Recommendations are ranked by persistence, scope, expected metric impact, affected games, and regression risk.

1. Fix winner-stage instability — highest priority

2. Repair or isolate the simulation module

3. Run a bounded calibration shadow experiment

A larger simultaneous change to confidence.empirical_max_weight, confidence.base_cap, and confidence.moderate_cap is not recommended; use one bounded shadow lever first.

4. Address weak run-model signal through isolated training work

5. Shadow-test starter venue/form weighting as the strongest single-slate driver

This is a bounded experiment, not a conclusion that all starter projections are systematically wrong.

Parameter abstentions

9. Bottom line

Known

Uncertain

Single safest next action

Do not promote or learn from this slate. First repair and trace the artifact and probability contracts: preserve pre/post winner picks, reconcile the WAS@DET published probability transformation, and correct the River Ryan workload artifact. Then run the calibration-cap and simulation fixes in shadow mode with rollback monitors before changing production parameters.