Poisson score probability calculatorHERMES TOOLS · Sports data

The Poisson model turns each team's expected goals λ into the probability of scoring 0, 1, 2 goals and so on. P(k) = e⁻λ × λᵏ ÷ k!. The tool assumes the two teams score independently and multiplies the two to get each scoreline.

When you pick a match from this site, the tool uses the finished scores cached in this build, rates home and away attack and defence separately, then multiplies by the league's home and away scoring averages. A sample of n games gets weight n ÷ (n + 6); the rest shrinks toward the league average. Last season and injuries are not used. You can still adjust the values afterwards.

Example: with λ = 1 for both teams, 0–0 has a probability of e⁻² = 13.53%, and 1–1 is also 13.53%. Both teams scoring is (1 − e⁻¹)² = 39.96%. With λ = 0 for both, 0–0 is 100%.

Win / draw / loss totals go up to 40 goals per team. The grid shows 0–5 goals plus the probability outside it, instead of quietly folding the tail into 5. The model ignores red cards, injuries and dependence between the teams; it is a data reference, not a prediction.