Every football match prediction starts with one number: how many goals each team is expected to score. The Poisson distribution is the math that turns that single expected-goals figure into a full set of scorelines and win probabilities.

A team’s expected goals (often written xG or λ, lambda) is not just their season average. We adjust for the opponent’s defensive strength, home advantage, rest days, and recent form. A weak attack playing an even weaker defence can have a higher expected-goals value than its raw average suggests.

Once you have λ_home and λ_away, the Poisson formula gives the probability of any exact scoreline. For example, if a team’s expected goals is 1.5, the model assigns roughly a 22% chance of exactly 1 goal and a 34% chance of exactly 2 goals.

Every possible scoreline (0-0, 1-0, 2-1, and so on) has a probability. Sum all scorelines where the home team scores more, and you get the home-win probability. Do the same for draws and away wins. That three-way split is the 1X2 market.

This is why two matches with the same favourite can still feel different. If one expected-goals pair produces many tight scorelines, the draw probability rises; if it produces one-sided scorelines, the favourite’s win probability rises even though the "pick" looks the same.

Poisson assumes goals are independent events, which is a simplification. Red cards, weather, and game state (a team chasing a goal late) change the flow in ways the base model does not capture. Good prediction systems layer those signals on top rather than trusting Poisson alone.

That said, over hundreds of matches the Poisson framework is a strong, calibrated baseline — which is why it remains the foundation of most serious football prediction tools.

A plain-language guide to the Poisson model behind football score predictions: expected goals, the 1X2 matrix, and why it beats gut feeling.

How Poisson Distribution Predicts Football Scores

Expected goals (xG) measures the quality of a team’s chances, not just how many goals they actually scored. A tap-in has a high xG; a long-range effort that squeaks past the post has a low one. Sum them up and you get a fairer picture of how a match really went.

A team can win 2-0 while creating only two big chances all game. Another can lose 0-1 while peppering the goal. The final score tells you the result; xG tells you the performance. Over a season, xG is a much better predictor of future results than raw goals.

If a team consistently outperforms its xG (scoring more than the chances deserved), a regression is often coming. If a team underperforms its xG (finishing poorly), it may be due to get luckier. Sudden gaps between xG and actual goals are early warning signs.

In our predictions, xG sits alongside form, schedule, and opponent strength as one input to expected goals — never the whole story, but an important check against misleading scorelines.

Expected goals, or xG, explained simply: what it measures, why it beats raw shots and scorelines, and how to read it before you bet or predict.

预测方法博客

理解每一个预测背后的逻辑

How Poisson Distribution Predicts Football Scores

What Is xG (Expected Goals) and How to Use It