BACK/TEST
Read this part.
Not the chart.
The model placed 521 bets at a yield of +1.13%. Across 521 settled picks the 95% Wilson interval is 33.94% – 42.25% — it brackets enough outcomes both clearly profitable and clearly losing that the headline is noise, not signal. The interesting result is below, inside the edge bands: the model's most confident picks lost the most money.
A bumpy flat line.cumulative P/L · 1u per pick
HIGHER EDGE, lower yield.
If the picks contained real signal, profitability would scale with conviction. It doesn't. The model's most confident bets (15%+ edge) lose the most.
This is the over-confidence pattern Pinnacle's closing line is designed to enforce against amateur models. The model has not yet beaten it.
CI —
CI 35%–50%
CI 31%–44%
CI 24%–45%
CI 17%–48%
1X2 is where it fails; totals is where it accidentally lives.
| Market | Picks | Win rate | Yield | P / L | Brier | Log loss | N scored |
|---|---|---|---|---|---|---|---|
| 1X2match result | 298 | 31.2% | -0.57% | -1.70u | 0.1970 | 0.5784 | 1137 |
| O/U 2.5total goals | 223 | 47.1% | +3.40% | +7.58u | 0.2436 | 0.6803 | 758 |
| Informational — no picks made on these markets in v1. | |||||||
| BTTSboth teams to score | — | — | — | — | 0.2500 | 0.6985 | 758 |
| Date▼ | Fixture· | Mkt· | Pick | Odds· | Edge· | Result | P / L· |
|---|---|---|---|---|---|---|---|
| May 25 | WolvesvBrentford | O/U 2.5 | Underunder | 2.63 | 4.9% | won | +1.63u |
| May 25 | TottenhamvBrighton | O/U 2.5 | Overover | 1.49 | 3.7% | won | +0.49u |
| May 25 | TottenhamvBrighton | 1X2 | Hhome | 3.83 | 20.8% | lost | -1.00u |
| May 25 | SouthamptonvArsenal | O/U 2.5 | Underunder | 2.99 | 7.5% | lost | -1.00u |
| May 25 | SouthamptonvArsenal | 1X2 | Aaway | 1.25 | 4.1% | won | +0.25u |
| May 25 | Nott'm ForestvChelsea | 1X2 | Hhome | 3.50 | 4.9% | lost | -1.00u |
| May 25 | NewcastlevEverton | O/U 2.5 | Underunder | 2.75 | 16.6% | won | +1.75u |
| May 25 | NewcastlevEverton | 1X2 | Aaway | 10.36 | 10.5% | won | +9.36u |
| May 25 | NewcastlevEverton | 1X2 | Ddraw | 6.19 | 8.5% | lost | -1.00u |
| May 25 | Man UnitedvAston Villa | O/U 2.5 | Underunder | 2.44 | 8.8% | won | +1.44u |
| May 25 | Man UnitedvAston Villa | 1X2 | Ddraw | 4.37 | 3.5% | lost | -1.00u |
| May 25 | Man UnitedvAston Villa | 1X2 | Hhome | 4.82 | 14.4% | won | +3.82u |
| May 25 | LiverpoolvCrystal Palace | O/U 2.5 | Underunder | 3.48 | 11.1% | won | +2.48u |
| May 25 | LiverpoolvCrystal Palace | 1X2 | Ddraw | 6.04 | 3.3% | won | +5.04u |
| May 25 | IpswichvWest Ham | 1X2 | Aaway | 2.23 | 7.0% | won | +1.23u |
| May 25 | FulhamvMan City | 1X2 | Hhome | 6.32 | 4.3% | lost | -1.00u |
| May 25 | BournemouthvLeicester | O/U 2.5 | Underunder | 2.67 | 8.6% | won | +1.67u |
| May 25 | BournemouthvLeicester | 1X2 | Aaway | 9.01 | 8.0% | lost | -1.00u |
| May 25 | BournemouthvLeicester | 1X2 | Ddraw | 5.67 | 4.9% | lost | -1.00u |
| May 20 | Man CityvBournemouth | O/U 2.5 | Underunder | 2.66 | 3.6% | lost | -1.00u |
| May 20 | Crystal PalacevWolves | 1X2 | Hhome | 2.41 | 7.3% | won | +1.41u |
| May 19 | BrightonvLiverpool | 1X2 | Aaway | 3.09 | 26.9% | lost | -1.00u |
| May 18 | ArsenalvNewcastle | O/U 2.5 | Overover | 1.74 | 3.8% | lost | -1.00u |
| May 18 | ArsenalvNewcastle | 1X2 | Hhome | 1.94 | 6.2% | won | +0.94u |
| May 18 | LeicestervIpswich | O/U 2.5 | Underunder | 2.30 | 3.6% | won | +1.30u |
For the curious.
Skip if you trust us.
For every calendar date in the test season, a Dixon-Coles model is refit on every completed match strictly prior to that date — including completed fixtures within the test season itself. No future data leaks into training. The fit uses exponential time-decay weighting (xi = 0.0019, calibrated to give recent fixtures roughly 3.5× the weight of matches eighteen months ago).
Each match emits an 11 × 11 score distribution which is then collapsed into 1X2 and Over/Under 2.5 market probabilities. Edge is computed against the de-vigged closing line — Pinnacle preferred, Bet365 as fallback. A pick is taken when edge ≥ 3% and the stake is 1u. BTTS is reported as informational only because football-data.co.uk does not carry a consistent BTTS odds feed.
Closing-line value is zero by construction here because edge already uses the closing price. The column is preserved against the day this harness is wired to a live odds source. Sample sizes inside the edge buckets are small; treat any band whose 95% Wilson CI crosses the break-even win rate as statistically unproven, regardless of point estimate.