Back to home
Public learning proof

Learning Proof

A first-class proof chain showing whether the V2 engine is actually improving over time, whether the aggregate still matches ledger truth, and whether calibration and challenger workflows are healthy enough to support public claims.

Generated 08/08/2026, 05:17:44

HEALTHY
Predictions Analysed
84K+
83,775 exact surfaced forecasts
Settled Forecasts
79,696
Exact settled surfaced V2 ledger
Model Brier
0.2065
Lower is better
Alpha
-0.0045
Market Brier minus model Brier
ROI
-2.41%
Flat-stake surfaced ROI
Calibration Freshness
19.6h
SLA: 24h
Aggregate Reconciliation
Healthy
Last reconciled 08/08/2026, 05:18:38
Max drift 0.000000
Calibration SLA
Within SLA
Last updated 07/08/2026, 09:40:59
Challenger Activity
Active
Last activity 08/08/2026, 04:05:01
Pending promotions: 0 · recent decisions: 25

Rolling Walk-Forward Evidence

Shareable audit of persisted pre-match V2 forecasts. Rows are grouped by forecast-created week, and a cohort is only evaluated once that market has at least 100 prior surfaced settled forecasts. Forecast rows are not the same as executed bets: many markets can come from one fixture. This is deliberately separate from the tiny scheduled weekly comparison jobs.

Generated 08/08/2026, 05:18:42

Scope: 365d, surfaced-only, engine opportunities-engine-v2, cohort field createdAt, settled-outcome field settledAt. This validates stored forecasts against the market Brier baseline; it is not a claim that every row was executable at unlimited stake size.

Evaluated Holdout
79,198
5,842 fixtures, 79,696 eligible rows
Cohorts
18
1 warm-up cohorts skipped
Alpha vs Market
-0.0046
Market Brier minus model Brier
Flat ROI
-2.44%
-1932.77u over holdout rows
CLV Check
-1.56%
24.9% hit, 29,750 samples
MarketEligibleHoldoutFixturesCohortsAlphaROIModel BrierMarket BrierAvg CLVRange
All Markets79,69679,1985,84218-0.0046-2.44%0.20650.2019-1.56%27/03/2026 to 07/08/2026
Asian Handicap20,42517,7694,17917-0.0138-9.84%0.21940.2055n/a27/03/2026 to 07/08/2026
Goals Over/Under14,59314,4704,22818-0.0048-2.25%0.19340.1886-1.58%27/03/2026 to 07/08/2026
Match Winner9,5758,3783,62217-0.0032-0.04%0.20200.1988-1.31%27/03/2026 to 07/08/2026
Half Goals9,5817,5153,57817+0.0420+7.48%0.20410.2461+1.08%27/03/2026 to 07/08/2026
Double Chance5,7174,4102,10517-0.0097-5.20%0.23020.2205-0.78%27/03/2026 to 07/08/2026
Corners5,7563,7641,16011-0.0130-6.43%0.21680.2038n/a28/03/2026 to 07/08/2026
BTTS3,6012,8311,76217-0.0049-7.31%0.24420.2393-1.01%28/03/2026 to 07/08/2026
Team Goals2,7292,5037369+0.0189+60.70%0.21900.2379-1.54%06/05/2026 to 07/08/2026
Correct Score3,2412,4321,1874-0.0019-25.15%0.05250.0506n/a29/03/2026 to 08/05/2026
Clean Sheet1,6211,4541,40617-0.0081-6.55%0.22090.2127n/a28/03/2026 to 07/08/2026
Even/Odd Goals1,91070064314-0.0036-5.46%0.25350.2498+0.34%27/03/2026 to 07/08/2026
HT/FT6093092613-0.0031-16.89%0.11720.1141n/a03/04/2026 to 05/05/2026
Win to Nil evidence quarantined

Old Win to Nil rows were recomputed with strict settlement rules, then excluded from headline proof because the original odds feed mapped stale market IDs into Win to Nil. These rows remain visible here as an audit trail, but they do not count toward public ROI, Brier, alpha, CLV, or market-family proof.

Quarantined Rows
6,459
6,801 historical Win to Nil rows
Active Proof Rows
0
0 settled rows currently counting
Removed ROI
+846.47%
Audit-only; +57568.60u is excluded
Strict Audit ROI
+1007.87%
5,957 recomputable rows, still excluded
Dominant Bad Pattern
31.27
Win to Nil - Home / No / key=no / side=home
Reason: win_to_nil_market_id_contamination_2026_04_14. Last quarantine write: 04/07/2026, 13:58:27. Latest captured row: 11/04/2026, 17:43:06. Latest active row: n/a. Generated 08/08/2026, 10:38:11.
MatchStored BetBookCapturedScoreStoredStrictUnit
Rayo Vallecano vs Elche
La Liga
Win to Nil - Home / No
@ 41.00 - Stored outcome differed from strict Win to Nil settlement
bet36503/04/2026, 14:21:341-0wonlost-1.00u
Grenoble Foot 38 vs Clermont
Ligue 2
Win to Nil - Home / No
@ 29.00 - Stored outcome differed from strict Win to Nil settlement
bet36503/04/2026, 14:21:242-2lostwon+28.00u
Grenoble Foot 38 vs Clermont
Ligue 2
Win to Nil - Home / No
@ 29.00 - Stored outcome differed from strict Win to Nil settlement
bet36503/04/2026, 14:18:322-2lostwon+28.00u
Nottingham Forest vs Burnley
Premier League
Win to Nil - Home / No
@ 41.00 - Highest contaminated stored odds
bet36511/04/2026, 14:55:364-1wonwon+40.00u
Rayo Vallecano vs Espanyol
La Liga
Win to Nil - Home / No
@ 41.00 - Highest contaminated stored odds
bet36511/04/2026, 14:05:011-0lostlost-1.00u
Cambridge United vs Notts County
League Two
Win to Nil - Home / No
@ 41.00 - Highest contaminated stored odds
bet36511/04/2026, 13:46:514-0lostlost-1.00u
Proof Chain Windows
Exact surfaced settled windows comparing older periods against recent behavior.
WindowSampleBrierMarket BrierAlphaROIAvg CLVCLV Hit
Older 7d2,4970.21790.2216+0.003712.90%-0.7812.8%
Recent 7d1,8150.21600.2188+0.00291.98%-0.0320.6%
Older 30d6,9220.21000.2098-0.00024.57%-0.419.8%
Recent 30d7,7830.21270.2159+0.00328.49%-0.4916.1%
7-Day Improvement
Older 7d vs recent 7d, with 95% confidence intervals on the change.
Brier Improvement
+0.0020
95% CI -0.0078 to 0.0117
Directional only · improved
Alpha Improvement
0.0000
95% CI 0.0000 to 0.0000
Directional only · worsened
ROI Improvement
-10.91pp
95% CI -20.09 to -1.74pp
Statistically clear · worsened
Avg CLV Improvement
+0.76%
95% CI 0.12 to 1.39%
Statistically clear · improved
CLV Hit Improvement
+7.74pp
95% CI 4.85 to 10.62pp
Statistically clear · improved
30-Day Improvement
Older 30d vs recent 30d for a broader view of whether the learning story survives over longer horizons.
Brier Improvement
-0.0026
95% CI -0.0084 to 0.0032
Directional only · worsened
Alpha Improvement
0.0000
95% CI 0.0000 to 0.0100
Statistically clear · improved
ROI Improvement
+3.91pp
95% CI -1.38 to 9.20pp
Directional only · improved
Avg CLV Improvement
-0.08%
95% CI -0.41 to 0.24%
Directional only · worsened
CLV Hit Improvement
+6.29pp
95% CI 4.90 to 7.68pp
Statistically clear · improved