Guide

Most likely exact score: why it is rarely above 15 %

Why the most likely exact score rarely tops 15 %: goals spread across many outcomes, shown with a worked example and a simple way to read it.

5 min read19 November 2026by AI Football Predictor

You open a match page and see that the most likely score is 1-0, at 12 %. Your first reaction may be: 'Only 12 %? That looks tiny for the favourite.' Then you see 1-1 at 11 % and 2-0 at 11 %, and the whole thing starts to look like a shrug.

It is not a shrug. It is how football works. In this guide we explain why the most likely exact score almost never goes above 15 %, even when one team is clearly stronger, and how to read that number properly.

A single score is a thin slice of the pie

A football match can end in many different scores: 0-0, 1-0, 1-1, 2-1, 2-0, 3-1, 2-2 and so on. Each one gets its own slice of the total probability, and all the slices together must add up to 100 %.

Compare that with the question 'who wins?'. There are only three answers (home win, draw, away win), so the slices are big. With exact scores, the same 100 % is cut into dozens of pieces. The more outcomes there are, the smaller each one becomes.

  • Result market: 3 possible outcomes, so the top one is often between 40 % and 60 %.
  • Exact score: dozens of realistic outcomes, so the top one is usually around 9 % to 14 %.

A 12 % score is therefore not a weak signal. It is simply one slice among many.

Goals are spread out, not concentrated

Football is a low-scoring sport. A typical top-flight match has roughly two and a half goals in total, give or take depending on the league. But 'about 2.5 goals' does not mean 'exactly 2 or 3 goals every time'. In practice you see 0-0 draws, 1-0 wins, 4-2 shootouts, and everything in between.

Statisticians often describe goal counts with a Poisson distribution. The idea is simple: if you know the average number of goals a team is expected to score (its expected goals, or xG), you can estimate how likely it is to score 0, 1, 2, 3 or more. Even for a team with an average of 1.8 goals, scoring exactly 1 or exactly 2 are both common, and 0 or 3 still happen regularly.

Now combine two teams. Each has its own spread of possible goal counts, and the exact score is the combination of both. The probabilities of the two spreads are multiplied together, which makes every single combination smaller again.

AIFP
These numbers, for every match todayProbabilities, expected goals and checked reliability, free, on the site.
See the probabilities

A worked example

Let us take an invented match as an example. The home side has an expected goals figure of 1.8 and the away side 0.9. Using a simple Poisson model, the most likely scores come out roughly like this (rounded, for illustration only):

Score (home-away)Approximate probability
1-012.1 %
2-010.9 %
1-110.9 %
2-19.8 %
0-06.7 %
3-06.5 %
0-16.0 %
3-15.9 %

Look at what happens. The home side is a clear favourite: in this example it wins roughly 59 % of the time, the draw sits near 23 % and the away side near 19 %. Yet the best single score, 1-0, only reaches about 12 %.

Why? Because the home win is shared between 1-0, 2-0, 2-1, 3-0, 3-1, 4-0 and many other scores. The result is strong, but it is split into pieces. The most likely score matches the most likely result here (a home win), but its probability is far smaller than the probability of the result itself.

Why 15 % is a practical ceiling

Can you find a match where the top score goes higher? Rarely. Two things would be needed at the same time: a very predictable number of goals for both teams, and a very small spread around it.

Even in extreme examples the cap is low. Imagine a huge favourite with 3.0 expected goals against a team with 0.5. With the same simple model, 2-0 and 3-0 each come out at roughly 13 to 14 %. That is already an unusually lopsided match, and it still stays under 15 %.

At the other end, in a tight match between two cautious sides with about 1 expected goal each, 1-1 and 0-0 and 1-0 all share similar slices, usually between 10 % and 14 % at best.

  • Low-scoring matches: a few scores (0-0, 1-0, 0-1, 1-1) share the probability, but none dominates.
  • High-scoring matches: there are even more possible scores, so each slice shrinks further.
  • Lopsided matches: one team scoring many goals widens the range of plausible scorelines.

This is why, in real model output, you will see most top scores between roughly 9 % and 15 %. Above that is exceptional.

How to read the number correctly

The right question is not 'is 12 % high enough?' but 'compared with what?'. Here is a better way to look at it.

  • Compare the top score with the second and third. If 1-0 is at 12 % and 1-1 is at 11 %, the model is saying the match is open and many outcomes are close.
  • Look at the result probabilities first. The chance of a home win, draw or away win tells you who is favoured. The exact score only adds detail.
  • Check the expected goals. xG shows the underlying balance of chances and helps you understand why a score is leading.
  • Remember that the other 88 % exists. A score at 12 % means that about 88 times in 100 similar situations, another score happens.

Common misunderstandings

'A low percentage means the model is unsure'

Not at all. The model can be very clear about who is stronger and still show a modest percentage for the best score, simply because of the maths of splitting the outcomes.

'The most likely score should happen most of the time'

It should not. At 12 %, it happens about once in eight similar matches. That is perfectly consistent with it being the most likely score.

'Exact scores are a safe way to read a match'

They are the opposite. Exact scores are the hardest thing to estimate, which is exactly why the numbers stay small. If you ever consider gambling on football, remember that statistics are not guarantees, that gambling is for adults only and that it carries real risks.

The takeaway

The most likely exact score rarely goes above 15 % because football produces many plausible scorelines and the probability has to be shared between all of them. A figure around 10 % to 13 % is normal, healthy and perfectly informative when you compare it with the neighbouring scores and with the result probabilities.

AI Football Predictor publishes these numbers (result probabilities, expected goals and the most likely score) for every match, so you can see the full picture and not just the headline score. Next time you see 1-0 at 12 %, you will know it is not a weak signal, just a slice of a well-divided pie.

faq

FAQ

Is 12 % a low probability for the most likely score?

No. Exact scores are split among dozens of possible outcomes, so even the best one usually sits between about 9 % and 15 %. A 12 % figure is a normal value, not a sign of weakness.

Can the most likely score ever be above 15 %?

It is rare. It would require a very predictable match with a narrow spread of goals. Even a heavy favourite against a very weak side typically produces top scores of around 13 to 14 % in a simple model.

Does the most likely score tell me who will win?

On our pages the most likely score always matches the most likely result, but the result probability is the better number to read. A score of 1-0 at 12 % sits inside a home win probability that may be much higher, such as 50 % or more.

The probability of every match, free

Today's most likely matches, live, team sheets and proof of reliability, free on the site.