If you have ever looked at a post-match graphic and thought that the team with 65% possession should have won, you have fallen into one of football's most persistent traps. Possession statistics are everywhere. They are on the screen at half-time, they fill pre-match analysis segments, and they are the first thing many people check when reviewing a game they missed.

But here is the uncomfortable truth: possession is a remarkably poor predictor of match outcomes. And if you are using it as a major input for your scoreline predictions, it is probably hurting your results more than helping them.

The possession myth

The idea that more possession equals more goals has its roots in the Barcelona era under Pep Guardiola. That team dominated games through relentless ball retention and won almost everything. The conclusion many people drew was straightforward: control the ball, control the game, win the match.

The problem is that Barcelona did not win because they had lots of possession. They won because they had Messi, Xavi, and Iniesta executing a system that happened to involve lots of possession. The possession was a byproduct of their quality and approach, not the cause of their success.

In the Premier League, the correlation between possession percentage and match outcome is surprisingly weak. Teams regularly win with less than 40% possession. Teams regularly lose while holding 60%+ of the ball. The stat on its own tells you very little about who scored more goals.

Why possession misleads

Possession without penetration

Having the ball does not mean you are doing anything dangerous with it. A team can pass the ball sideways across the back four for minutes at a time, racking up possession percentage without ever threatening the goal. This kind of sterile possession is common when a team is struggling to break down a well-organised defence.

When you see a team averaging 60% possession, ask yourself: are they creating chances with that ball, or are they recycling it in safe areas? Expected goals (xG) is a far better indicator of whether a team is actually threatening. A team with 40% possession and 2.5 xG is more dangerous than a team with 65% possession and 0.8 xG.

Deliberate low possession

Some teams deliberately concede possession as a tactical choice. Counter-attacking teams want the opposition to have the ball so they can exploit the spaces left behind when the opponent pushes forward. Teams like this often have low possession averages but strong goal-scoring records because they are lethal on the break.

If you see a team averaging 42% possession and assume they are struggling, you might be completely wrong. They could be executing a plan perfectly. Underdog results often come from teams who cede possession happily and then punish the opposition with clinical counter-attacks.

Game-state effects

Possession stats are heavily influenced by the scoreline. A team that goes 1-0 up early will often sit deeper and allow the opposition more of the ball, focusing on defensive shape and counter-attacks. The final possession stat might show 55-45 in favour of the losing team, but that does not mean they controlled the game - it means they were chasing it.

This is particularly misleading when you look at season averages. A team that frequently takes early leads will have a deflated possession average because they spend lots of time defending leads. That does not mean they lack ball control - it means they are winning.

When possession stats are actually useful

All of that said, possession is not completely useless. There are specific situations where it adds genuine value to your predictions:

Identifying tactical matchups

When two teams with very different possession profiles meet, the stat helps you anticipate the type of game you will see. A possession-heavy team against a counter-attacking team often produces a specific pattern: one side dominating territory while the other looks to hit on the break. These matches tend to be lower-scoring than the average fixture, because the high-possession team controls the tempo but struggles to create clear chances, while the counter-attacking team has fewer but more direct opportunities.

This is useful for picking your scoreline. If you expect a possession-vs-counter match, scorelines like 1-0, 0-1, and 1-1 become more likely. The open, end-to-end 3-2 is much less probable.

Spotting style changes

If a team's possession average suddenly shifts significantly - say from 55% to 45% over several weeks - that tells you something about a tactical change. Maybe a new manager has come in, or the existing manager has changed approach. Tracking these shifts can alert you to changes before they show up in the results table.

Contextualising other stats

Possession is most useful when combined with other metrics. A team with 60% possession and a high number of shots on target is genuinely dominant. A team with 60% possession and few shots is just keeping the ball without purpose. The possession stat gains meaning when you layer it with chance creation data.

Better stats for your predictions

If possession is unreliable on its own, what should you be looking at instead? Here are the statistics that correlate more strongly with match outcomes:

Expected goals (xG) - measures the quality of chances created, not just the quantity
Shots on target - a simple but effective proxy for attacking threat
Big chances created - opportunities where a player is expected to score
Goals conceded from open play - strips out penalties and set pieces to show defensive quality
Progressive passes and carries - better than raw possession for measuring whether a team advances the ball dangerously

These metrics give you a much clearer picture of how a team actually performs. Combined with form analysis, they form a solid foundation for prediction decisions that possession alone cannot provide.

A practical example

Imagine you are predicting a match between Team A (averaging 58% possession, 1.4 xG per game) and Team B (averaging 44% possession, 1.8 xG per game). If you looked only at possession, you might favour Team A. But Team B creates better chances per game despite having less of the ball. They are the more dangerous side.

Now add context: Team A is at home (which tends to boost possession slightly and also performance). Team B is a strong counter-attacking side that does not mind travelling without the ball.

A reasonable prediction here might be a close game - perhaps 1-1 or 1-2 to Team B. The possession stat alone would have pointed you towards Team A, but the underlying quality of chances tells a different story. This is exactly why looking beyond surface-level stats matters so much.

The bottom line

Possession is one of the most visible statistics in football and one of the least useful for predicting results. It tells you who had the ball, not who did anything dangerous with it. It is distorted by game state, tactical approach, and the quality gap between teams.

Do not ignore it entirely - it has some value for understanding the likely pattern of a match. But never use it as a primary input for your scoreline predictions. Combine it with xG, shots on target, and form analysis, and you will have a much more accurate picture of what is likely to happen. The best predictors are the ones who look past the obvious numbers and dig into what actually drives goals.

Keep reading

Want to use data more effectively? Read Expected Goals Explained: What xG Means for Your Predictions or The Most Common Premier League Scores (and How to Use Them). For a broader approach, try How to Pick the Right Scoreline for Every Match.

The Role of Possession Stats in Score Predictions