History & model performance
We publish the model's track record — judge it yourself before trusting it.
6/6
leagues where the model beats league averages
7.8%
lower log loss than benchmark (1X2)
2,011
matches in holdout test
Settled forecasts
No settled forecasts yet — they appear after matches the model forecasted are finished.
Backtest — validation · 2023/2024
poisson-v1-hl180-k3 · Nov 19, 2026 · N=1979
| Model | Brier | Log loss | ECE |
|---|---|---|---|
| Poisson 1X2 | 0.5952 | 0.9959 | 0.0120 |
| Elo 1X2 | 0.5986 | 1.0013 | 0.0247 |
| Benchmark (league frequencies) 1X2 | 0.6535 | 1.0794 | 0.0197 |
| Poisson O/U 2.5 | 0.4827 | 0.6760 | 0.0184 |
| Benchmark (league frequencies) O/U 2.5 | 0.4916 | 0.6847 | 0.0034 |
| Poisson BTTS | 0.4938 | 0.6870 | 0.0253 |
| Benchmark (league frequencies) BTTS | 0.4940 | 0.6871 | 0.0108 |
Backtest — test (holdout) · 2024/2025
poisson-v1-hl180-k3 · Nov 19, 2026 · N=2011
| Model | Brier | Log loss | ECE |
|---|---|---|---|
| Poisson 1X2 | 0.5924 | 0.9926 | 0.0113 |
| Elo 1X2 | 0.5965 | 0.9982 | 0.0242 |
| Benchmark (league frequencies) 1X2 | 0.6520 | 1.0764 | 0.0086 |
| Poisson O/U 2.5 | 0.4898 | 0.6831 | 0.0368 |
| Benchmark (league frequencies) O/U 2.5 | 0.4954 | 0.6885 | 0.0131 |
| Poisson BTTS | 0.4984 | 0.6917 | 0.0367 |
| Benchmark (league frequencies) BTTS | 0.4945 | 0.6877 | 0.0135 |
Walk-forward backtest: each forecast uses only matches played earlier. Lower Brier/log loss is better. Past results do not guarantee future ones.