Poisson Match Predictor
Model match outcomes with Poisson distributions — scoreline probabilities, moneyline, spreads, and totals
Read the guide →| Line | Over % | Over Odds | Under % | Under Odds |
|---|---|---|---|---|
| 1.5 | 74.2% | -288 | 25.8% | +288 |
| 2.5 | 49.4% | +102 | 50.6% | -102 |
| 3.5 | 27.5% | +264 | 72.5% | -264 |
| 4.5 | 13.0% | +672 | 87.0% | -671 |
| 5.5 | 5.3% | +1800 | 94.7% | -1794 |
| Spread | Home Covers | Home Odds | Away Covers | Away Odds |
|---|---|---|---|---|
| -0.5 | 68.7% | -220 | 31.3% | +220 |
| -1.5 | 87.0% | -672 | 12.9% | +673 |
| -2.5 | 95.8% | -2282 | 4.2% | +2291 |
| +0.5 | 42.8% | +133 | 57.1% | -133 |
| +1.5 | 20.7% | +383 | 79.3% | -383 |
| +2.5 | 7.9% | +1164 | 92.1% | -1162 |
The Poisson distribution models the number of events occurring in a fixed interval, making it ideal for predicting match scores in sports like soccer, hockey, and baseball. Each team's scoring is treated as an independent Poisson process with its own expected rate (lambda).
How it works: Given expected goals for each team, the calculator builds a joint probability matrix where P(home=h, away=a) = P(home=h) × P(away=a). This independence assumption means each team's scoring is modeled separately, then combined.
Limitations: The independence assumption doesn't capture game-state effects — e.g., a trailing team may push harder in soccer, or score-dependent pitching changes in baseball. The model also ignores overtime/extra innings. Despite these limitations, the Poisson model is widely used in sports analytics as a strong baseline for match outcome probabilities.