Skip to content
Independent research on sports-betting software · No sportsbook partnerships
PRICE WATCH

Tool · Positive-EV betting

Poisson calculator: soccer score and match odds from expected goals

Free Poisson calculator for soccer betting: enter each team's expected goals to get every scoreline's probability, home/draw/away, over/under 2.5, both teams to score and fair odds.

Updated Nov 9, 2026Facts checked Nov 9, 2026

See the software we recommend for this
On this page

A Poisson calculator turns each team's expected goals into the probability of every possible score, then adds those up into match odds: home, draw, away, over/under 2.5 goals and both teams to score. Enter the two expected-goals figures below to see the probabilities and the fair odds for each market, with no margin.

Poisson calculator

Enter each team's expected goals. Optionally add a sportsbook's price to see its expected value against the model.
Sportsbook prices (optional)

Scoreline probabilities

Home goals down, away goals across. Most likely score highlighted.

Home goals012345
06.77.44.11.50.40.1
110.811.86.52.40.70.1
28.69.55.21.90.50.1
34.65.02.81.00.30.1
41.82.01.10.40.10.0
50.60.60.40.10.00.0

Most likely: 1–1 at 11.8%.

MarketChanceFair oddsEV
Home win49.0%2.04 (+104)—
Draw24.9%4.02 (+302)—
Away win26.2%3.82 (+282)—
Over 2.5 goals50.6%1.97 (−103)—
Under 2.5 goals49.4%2.03 (+103)—
Both teams score: yes53.2%1.88 (−114)—
Both teams score: no46.8%2.14 (+114)—
Assumptions and limitations
  • Each team's goals follow a Poisson distribution with the expected goals you enter, independently of the other team. Scores up to 10–10 are counted.
  • Plain Poisson tends to underestimate draws and low scores, because real goals are not fully independent. Treat small edges on draws and unders with caution.
  • Fair odds are 1 ÷ probability, with no sportsbook margin. EV = probability × decimal odds − 1, against these inputs only.
  • The output is only as good as the expected-goals numbers. Nothing you enter is stored or sent anywhere.

Key takeaways

  • Input: expected goals for each team. The output can only be as good as those two numbers.
  • Output: a scoreline grid plus fair odds for 1X2, totals and both teams to score.
  • Compare the fair odds with a sportsbook's price. A price longer than the fair odds is +EV if your inputs are right.
  • Plain Poisson underrates draws and 0–0 / 1–1 scores; treat close calls with caution.

How the Poisson calculator works

The Poisson distribution gives the chance of an event happening exactly k times when it happens λ times on average:

P(k goals) = λ^k × e^−λ ÷ k!

The calculator applies it to each team separately, using its expected goals as λ, then multiplies the two to get each scoreline. A 2–1 home win is P(home scores 2) × P(away scores 1). Summing the right cells gives each market:

  • Home win: every cell where the home team scores more.
  • Draw: the diagonal (0–0, 1–1, 2–2 …).
  • Over 2.5: every cell with three or more total goals.
  • Both teams to score: every cell where neither team is on zero.

Fair odds are 1 ÷ probability, the price with no sportsbook margin.

Poisson betting example

Home team expected goals 1.6, away team 1.1.

Scoreline probabilities (home goals down, away goals across):

0123
06.7%7.4%4.1%1.5%
110.8%11.8%6.5%2.4%
28.6%9.5%5.2%1.9%
34.6%5.0%2.8%1.0%

The most likely single score is 1–1 at 11.8%, then 1–0 at 10.8%. Adding up the whole grid:

MarketProbabilityFair decimal oddsFair American odds
Home win49.0%2.04+104
Draw24.9%4.02+302
Away win26.2%3.82+282
Over 2.5 goals50.6%1.97−103
Under 2.5 goals49.4%2.03+103
Both teams to score: yes53.2%1.88−114

If a sportsbook offered the home win at +120 (2.20), its expected value against these inputs would be 0.490 × 2.20 − 1 ≈ +7.7%. Check any price the same way with the expected value calculator.

Where expected goals come from

The calculator doesn't predict anything by itself; it reshapes two numbers you give it. Common sources:

  • Your own model: team attack and defence ratings from xG data, adjusted for venue and opponent.
  • The market: a sharp book's total and handicap lines imply each team's goal expectancy. Backing that out and running it through Poisson lets you price markets the book is slower to move, such as correct score or both teams to score.

The catch: where Poisson goes wrong

Real goals aren't fully independent. Teams that lead sit back, and teams that trail push, so plain Poisson tends to underestimate draws and low scores and slightly overestimate scorelines like 1–0. Red cards, rotation and weather aren't in the formula either. The usual fix in serious models is the Dixon–Coles adjustment to the low-score cells. Treat a small edge on a draw or under price with suspicion.

Bottom line

A Poisson calculator converts expected goals into scoreline probabilities and fair odds for every main soccer market. It's a fast way to price a match and spot prices worth a closer look, as long as your expected-goals inputs are better than the market's.

Frequently asked questions

What is a Poisson calculator in betting?

A tool that uses the Poisson distribution to turn each team's expected goals into the probability of every scoreline, and from those, the probability and fair odds of results like home win, draw, over 2.5 goals or both teams to score.

How does the Poisson distribution predict soccer scores?

It treats each team's goals as independent random events with a known average rate. The chance of a team scoring exactly k goals is λ^k × e^−λ ÷ k!, where λ is its expected goals. Multiplying the two teams' chances gives each scoreline's probability.

Where do the expected goals numbers come from?

From a model or data you trust: team xG averages adjusted for opponent and venue, or the goal expectancy implied by a sharp market's totals and handicap lines. The calculator's output is only as good as that input.

Is Poisson accurate for soccer?

It's a reasonable first approximation but it tends to underestimate draws and low-scoring results, because goals in real matches aren't fully independent. Serious models adjust for that, for example with the Dixon–Coles correction.

About this page

Editorial checks
Checked against the sources cited on this page. We correct it when a source changes or a reader reports an error.
Facts last checked
Nov 9, 2026

How we research and label evidence: our methodology.