Weekly · 2026-10-26 to 2026-11-01
Week 2026-W44: model vs market
135 finished matches with a forecast published before kickoff. Lower Brier score and log loss are better. oddsxi-goal-model p6-v5.
Matches135
Most likely outcome right84
1X2 Brier: model / market0.545 / 0.541
1X2 log loss: model / market0.924 / 0.924
Where the model did best
| Match | League | Score | Model gave the result | Market gave the result |
|---|---|---|---|---|
| Levadeiakos - Aris | Super League Greece | 3-0 | 45.5% | 24.0% |
| AC Milan - Bologna | Serie A | 2-4 | 32.5% | 17.7% |
| Fenerbahce - Goztep | Super Lig | 1-2 | 21.2% | 11.9% |
Where the model did worst
| Match | League | Score | Model gave the result | Market gave the result |
|---|---|---|---|---|
| Tottenham - Crystal Palace | Premier League | 2-0 | 39.7% | 61.9% |
| FC Augsburg - SC Freiburg | Bundesliga | 1-2 | 28.9% | 44.0% |
| Racing Santander - Real Madrid | La Liga | 0-3 | 52.3% | 77.4% |
Research record
153 cases where the model disagreed with the market by 3 points or more: 63 came true; the market expected 55.9. One week says little; see the Performance Lab for the full record.
Football analysis, not betting advice.