Enter how many goals you expect each team to score and the calculator turns it into probabilities for every scoreline, the match result, over/under 2.5 and both teams to score. This is the same statistical model most odds compilers start from.
Expected goals is the average number of goals a team would score against this opponent over many repeats. 1.5 means a bit more than a goal on a typical day. Estimates from the last 5 to 10 matches work fine.
The Poisson distribution answers one question: if a team scores at an average rate of λ goals per match, what is the chance it scores exactly 0, 1, 2 or 3 today? Goals in football are rare, roughly independent events, which is exactly the situation the distribution was built for. Feed it one number per team and it produces a full probability for every scoreline.
The formula is P(k) = e−λ λk / k!. With λ = 1.5, that gives a 22.3% chance of no goals, 33.5% for exactly one, 25.1% for two and 12.6% for three. Multiply the home and away distributions together and you get the full score matrix this calculator shows.
The output is only as good as the two numbers you feed it. The quick method: take the team's average goals scored over the last 6 to 10 matches, then adjust for the opponent. A simple, robust recipe used by odds compilers for decades:
Home λ = home team's attack strength × away team's defence weakness × league average home goals
Worked example: the league averages 1.50 home goals per match. The home side scores 20% more than an average home team (attack 1.2) and the visitors concede 10% more than average (defence 1.1). Home λ = 1.2 × 1.1 × 1.50 = 1.98.
If you already follow xG numbers, a team's average xG for and against over the recent sample is a better input than raw goals, because it strips out finishing luck. Our Poisson distribution guide walks through the full attack/defence method with a season's worth of numbers.
Three known blind spots, all worth respecting before you bet off the output:
Low-scoring draws are underrated. Plain Poisson treats the two teams as independent, but real matches see slightly more 0-0 and 1-1 results than independence predicts, because teams settle for what they have. Professional models apply a Dixon-Coles correction for exactly this. Treat the calculator's 0-0 and 1-1 numbers as a floor, not a ceiling.
Game state does not exist. A red card, an early goal or a must-win situation changes scoring rates mid-match. The model prices the match as if both teams play the same way for 90 minutes.
Your inputs carry all the error. A difference of 0.2 in λ moves the over 2.5 probability by roughly 5 percentage points. Two sensible people can disagree on the right λ by more than that, so compare the output against the market rather than trusting it in isolation. Run the bookmaker's odds through the no-vig calculator first and compare fair probability against Poisson probability.
What numbers do I enter into a Poisson calculator?
One expected-goals figure per team: the average number of goals you believe each side would score against this opponent. Recent scoring averages adjusted for the opposition, or average xG over the last 6 to 10 matches, are the standard inputs.
How accurate is Poisson for football betting?
It is a solid baseline. It prices over/under and correct score markets well for typical scorelines, but it slightly underrates low-scoring draws like 0-0 and 1-1 and it knows nothing about lineups, red cards or motivation. Professional models start from Poisson and add corrections.
Why does the calculator show fair odds next to each probability?
Fair odds are 1 divided by the probability. If the calculator says over 2.5 goals is 54%, fair odds are 1.85. A bookmaker price above that number suggests value, if you trust your inputs.
What is a typical expected goals value?
Across Europe's top leagues, home teams average roughly 1.5 goals and away teams roughly 1.2. A strong home favourite might justify 2.2 or more, a weak away side facing them 0.7 or less.
Does BetBot use Poisson in its predictions?
The bot's daily picks combine odds movement, form, standings and head-to-head data, and its goals-market logic is built on the same expected-goals thinking this calculator demonstrates. You can see the resulting picks graded on the track record page.
Daily football tips with 15%+ edge over bookmaker odds. Free, no signup, updated every morning.
Today's picks →