Weekly · 2026-11-02 to 2026-11-08
Week 2026-W45: model vs market
151 finished matches with a forecast published before kickoff. Lower Brier score and log loss are better. oddsxi-goal-model p6-v5.
Matches151
Most likely outcome right83
1X2 Brier: model / market0.563 / 0.547
1X2 log loss: model / market0.955 / 0.930
Where the model did best
| Match | League | Score | Model gave the result | Market gave the result |
|---|---|---|---|---|
| Blackburn - West Ham | Championship | 2-1 | 30.8% | 20.3% |
| Oviedo - Las Palmas | Segunda Division | 2-3 | 29.0% | 20.8% |
| Heerenveen - Feyenoord | Eredivisie | 5-3 | 29.5% | 22.5% |
Where the model did worst
| Match | League | Score | Model gave the result | Market gave the result |
|---|---|---|---|---|
| Auxerre - Paris FC | Ligue 1 | 0-1 | 31.7% | 54.1% |
| Venezia - Udinese | Serie A | 2-1 | 26.4% | 44.7% |
| Iraklis - Panathinaikos | Super League Greece | 1-2 | 46.7% | 68.7% |
Research record
174 cases where the model disagreed with the market by 3 points or more: 69 came true; the market expected 67.9. One week says little; see the Performance Lab for the full record.
Football analysis, not betting advice.