Last updated:
Short answer: correct score prediction deals with probabilities spread across dozens of possible scorelines, so even the single most likely result usually carries only about 10–13%. The standard way to estimate them is to take each team's expected goals, turn them into goal-count probabilities with the Poisson distribution, and multiply the two teams' figures.
A match result has three outcomes: home win, draw, away win. A scoreline has far more: stopping at four goals per side already gives 25 combinations. The full 100% has to be shared among all of them, so no single scoreline can grow very large.
In a typical league match, with each side expected to score between one and two goals, the top scoreline – usually 1-0 or 1-1 – lands at roughly 10 to 13%. Only unusual fixtures push it higher: with both teams expected to score 0.8 goals, 0-0 rises to about 20%. Put plainly, the scoreline at the top of the list happens in roughly one match in eight.
The starting point is one number per team: how many goals it is expected to score in this match. It is usually built from chance quality – the expected goals (xG) figures covered in our xG guide – adjusted for venue and for the opposing defence. In the illustrative example below, the home side is expected to score 1.5 goals and the away side 1.0.
Goals are rare events scattered across 90 minutes. The Poisson distribution is the standard statistical tool for counting events like these: give it an average, and it tells you how often to expect 0, 1, 2 or 3 of them. The formula is P(k) = e^(−λ) × λ^k ÷ k!, where λ is the expected goals figure and k the number of goals. What matters is the shape: most of the probability sits close to the average and thins out quickly.
| Goals | Home (expected 1.5) | Away (expected 1.0) |
|---|---|---|
| 0 | 22.3% | 36.8% |
| 1 | 33.5% | 36.8% |
| 2 | 25.1% | 18.4% |
| 3 | 12.6% | 6.1% |
The basic version assumes that the two teams' goal counts are independent of each other. Under that assumption, the probability of a scoreline is simply the home figure multiplied by the away figure: 1-0 is 33.5% × 36.8% ≈ 12.3%.
| Score | Calculation | Probability |
|---|---|---|
| 1-0 | 33.5% × 36.8% | 12.3% |
| 1-1 | 33.5% × 36.8% | 12.3% |
| 2-0 | 25.1% × 36.8% | 9.2% |
| 2-1 | 25.1% × 36.8% | 9.2% |
| 0-0 | 22.3% × 36.8% | 8.2% |
| 0-1 | 22.3% × 36.8% | 8.2% |
| 1-2 | 33.5% × 18.4% | 6.2% |
| 2-2 | 25.1% × 18.4% | 4.6% |
| 3-0 | 12.6% × 36.8% | 4.6% |
| 3-1 | 12.6% × 36.8% | 4.6% |
The two most likely scorelines are tied at 12.3% each, so even with a clear home favourite, 1-1 is exactly as likely as 1-0. The ten rows together cover about 79.5%; roughly a fifth of the probability sits on results not shown.
Every other goal-based question is just a sum of cells in the same distribution. With the same 1.5 against 1.0 inputs:
Reading markets this way keeps them consistent: a very high BTTS expectation and 1-0 as the favourite scoreline do not fit together.
A correct score estimate is a distribution, not a single answer. Expected goals go in, the Poisson distribution turns them into goal counts, and multiplying the two teams' figures gives every scoreline a probability – the top one rarely above about one in eight. Read the whole table, keep its blind spots in mind, and use it to check that your views on result, goal total and BTTS agree.
Reading a match as a spread of scorelines rather than a single answer is part of the daily match analysis in the WinBall app.
This page is published for information only. WinBall is not a betting platform; it provides football analysis and predictions. 18+
Free match analyses and statistics every day. Available on iPhone and Android. 18+.
Stay one step ahead with professional football analyses and data-driven predictions.
Payments are handled securely in-app via the App Store and Google Play.
18+ · Our predictions are for information and analysis purposes only; they do not guarantee any outcome. This platform does not offer betting and does not direct users to any betting site. Illegal betting is a crime. Please act responsibly.
© 2026 CH Global Bilişim Teknolojileri Limited Şirketi. All rights reserved.