Loading the page
Every trained model, ranked by how far it beats the benchmark its own market has to clear. Accuracy across different markets is not comparable, because a line that falls one way most of the time hands a model a high score for learning very little. The gap to the benchmark is what ranks here.
| # | Model | Edge | Accuracy | Baseline | Log loss | Sample | Since last train |
|---|---|---|---|---|---|---|---|
| 1 | Match result football · 1X2 · Always pick the majority class | +5.52 | 49.22% | 43.70% | 1.0189 | 23,602 | ▼ 0.12 |
| 2 | Corners over 9.5 football · Corners · Always pick the majority class | +3.68 | 53.98% | 50.30% | 0.6880 | 13,739 | ▼ 0.52 |
| 3 | Goals over 2.5 football · Goals · Always pick the majority class | +3.52 | 55.55% | 52.03% | 0.6821 | 23,602 | ▼ 0.67 |
| 4 | Cards over 4.5 football · Cards · Always pick the majority class | +3.23 | 61.12% | 57.89% | 0.6657 | 14,798 | ▲ 0.69 |
| 5 | Both teams to score football · BTTS · Always pick the majority class | +0.66 | 55.14% | 54.48% | 0.6863 | 23,602 | ▼ 0.03 |
| 6 | Cards over 2.5 football · Cards · Always pick the majority class | 0.00 | 77.25% | 77.25% | 0.5280 | 14,798 | ▼ 0.86 |
| 7 | Goals over 1.5 football · Goals · Always pick the majority class | 0.00 | 75.66% | 75.66% | 0.5473 | 23,602 | ▼ 0.14 |
| 8 | Corners over 7.5 football · Corners · Always pick the majority class | 0.00 | 73.27% | 73.27% | 0.5757 | 13,739 | ▲ 0.39 |
| 9 | Corners over 8.5 football · Corners · Always pick the majority class | 0.00 | 61.75% | 61.75% | 0.6596 | 13,739 | ▲ 0.54 |
| 10 | Half-time goals over 0.5 football · Goals · Always pick the majority class | 0.00 | 70.67% | 70.67% | 0.6014 | 16,829 | ▼ 0.49 |
| 11 | Cards over 3.5 football · Cards · Always pick the majority class | -0.76 | 59.45% | 60.21% | 0.6593 | 14,798 | ▼ 3.60 |
Edge is accuracy minus the baseline, in percentage points, measured on the held-out test period each model was evaluated against, never on data it trained on. A model with no recorded baseline shows its accuracy but no edge, because there is nothing to say how much of that accuracy is skill.