Public results · auto-settled · no betting advice
Splits our home-side picks from our away-side picks. If one side wins markedly more than the other, the model is mis-pricing home-court advantage.
Recent trend
| League | Model | Win rate | Hits / attempts | Settled | Avg confidence |
|---|---|---|---|---|---|
| MLB | elo-v1 | Sample too small (n=14) | 4 / 7 | 14 | 54.5% |
| LOL | lol-elo-v1 | Sample too small (n=1) | 1 / 1 | 1 | 57.7% |
No total settlements yet
Monthly trend
| League | Model | Win rate | Hits / attempts | Settled | Avg confidence |
|---|---|---|---|---|---|
| MLB | elo-v1 | 58.1% | 118 / 203 | 211 | 54.0% |
| LOL | lol-elo-v1 | Sample too small (n=1) | 1 / 1 | 1 | 57.7% |
No total settlements yet
All-time
| League | Model | Win rate | Hits / attempts | Settled | Avg confidence |
|---|---|---|---|---|---|
| MLB | elo-v1 | 54.4% | 1015 / 1867 | 1886 | 53.8% |
| NBA | elo-v1 | 40.0% | 10 / 25 | 35 | 61.3% |
| LOL | lol-elo-v1 | Sample too small (n=1) | 1 / 1 | 1 | 57.7% |
No total settlements yet
Brier is a three-class normalized score (0–1, lower is better); not yet benchmarked against a favourite baseline or market odds.
| League | Model | Win rate | Hits / attempts | Brier | Settled |
|---|---|---|---|---|---|
| FIFA World Cup | wc-elo-v1 | 62.0% | 57 / 92 | 0.2567 | 92 |
| LaLiga | laliga-elo-v1 | 52.2% | 36 / 69 | 0.2777 | 69 |
| Ligue 1 | ligue1-elo-v1 | 55.6% | 30 / 54 | 0.3040 | 54 |
| Premier League | epl-elo-v1 | 40.0% | 20 / 50 | 0.3315 | 50 |
| Serie A | seriea-elo-v1 | 56.0% | 28 / 50 | 0.2844 | 50 |
| Bundesliga | bundesliga-elo-v1 | 53.3% | 24 / 45 | 0.2890 | 45 |
Spread Brier uses the model's cover probability (0.25 ≈ coin flip, lower is better). Same auto-settled data as above, published unfiltered.
No spread settlements yet
No first-half winner settlements yet
No first-half / team total settlements yet
Perfect calibration
When we say '70% to win,' how often do those teams actually win? Points hugging the dashed line mean the model's confidence is honest. Blue = beats prediction (under-confident), red = worse than prediction (over-confident), green = within ±5%.
Moneyline model · 1893 settled games (voids excluded)