Weekly · 2026-10-19 to 2026-10-25
Week 2026-W43: model vs market
130 finished matches with a forecast published before kickoff. Lower Brier score and log loss are better. oddsxi-goal-model p6-v5.
Matches130
Most likely outcome right65
1X2 Brier: model / market0.615 / 0.587
1X2 log loss: model / market1.026 / 0.987
Where the model did best
| Match | League | Score | Model gave the result | Market gave the result |
|---|---|---|---|---|
| Levante - Athletic Club | La Liga | 3-1 | 43.6% | 25.9% |
| OFI Crete - PAOK | Super League Greece | 2-1 | 33.0% | 20.0% |
| FSV Mainz 05 - Werder Bremen | Bundesliga | 2-3 | 36.0% | 22.3% |
Where the model did worst
| Match | League | Score | Model gave the result | Market gave the result |
|---|---|---|---|---|
| Lincoln - Burnley | Championship | 4-1 | 17.9% | 33.4% |
| Eyupspor - Kasimpasa | Super Lig | 0-1 | 37.7% | 61.0% |
| Nurnberg - Kaiserslautern | 2. Bundesliga | 0-1 | 18.4% | 26.6% |
Research record
145 cases where the model disagreed with the market by 3 points or more: 54 came true; the market expected 55.1. One week says little; see the Performance Lab for the full record.
Football analysis, not betting advice.