The stats that tell us most about what has happened in Premier League matches

What can we learn from stats? Which ones tend to align with the result and which ones do not? Michael Mauboussin wrote in a book called the Success Equation: “Statistics are widely used in a range of fields. But rarely do the people who use statistics stop and ask how useful they really are.” Let us try to.

After a match has finished we already know whether a team have won, drawn or lost. But did they deserve to? As Jurgen Klopp has said: “You can win with one lucky strike.”

A single match can never tell us much, but over a series of games we might be able to build up a picture of whether a team’s results have been consistent with their play – and thus whether those results are likely to continue.

I studied the score and a range of stats from every Premier League game played in the last six seasons, 2015-16 to 2020-21. I was interested in the general relationships between those things. How often did the team with most of a given stat also score most goals? Did they win more often than they lost – and, if so, how much more often? Were some stats more closely associated with the result than others?

I put my findings in the graphic below. Start in the middle. The team who were caught offside more often won hardly any more contests than they lost – they won in 39 per cent of games and lost in 38 per cent.

Being flagged offside is something that can happen when a team attack, and a team have to attack to score. But when they are flagged offside that attack ends. Play restarts with a free kick for their opponents. An offside tells us next to nothing about the threat carried in other attacks.

A corner is not the end of an attack but the start of a new phase in one. It can be more revealing. The team who took more corners won in 42 per cent of games and lost in 34 per cent.

Possession told us only a tiny bit more. The team with more possession won in 43 per cent of games and lost in 34 per cent.

I looked at some other measures of ball retention – completed passes and passing accuracy. The figures were similar. This should not be surprising. They are different measurements of the same thing. Passing accuracy is highly correlated with completed passes, which are highly correlated with the time spent in possession.

The strongest relationship between a stat and the score was for shots. This should not be surprising either. The team who took more shots won in 53 per cent of games and lost in 24 per cent. The association with the result was better for some types of shot than for others. For example, teams with more shots from outside the box won less often than teams with more shots from inside the box, who won less often than teams with more shots on target.

What about expected goals, you might ask. Teams with more expected goals won as often as teams with more shots on target.

All of the stats I have discussed so far were associated with the result in a positive way. The team with more of a stat won more often than they lost. Some other stats were associated with the result in a negative way. The team with more of those stats lost more often than they won. The first type of stat was about attacking. The second type of stat is about defending.

The team who took more goal kicks lost in 40 per cent of games and won in 36 per cent. So did the team who committed more fouls.

Stronger than the relationship between the score and fouls, however, was the relationship between the score and bookings – ten points for each yellow and 25 points for each red. Yellow and red cards are shown for the most serious fouls. They are signs that a team were having to defend desperately. And when a player is sent off his teammates have to carry on without him.

You might be surprised nonetheless to discover that defeats for teams with more bookings points were more common than victories for teams with more corners or possession. They lost in 45 per cent of games and won in 31 per cent. I looked at other scoring systems used in betting markets – one point for each yellow and two points for each red, also one point for every card. The figures were similar.

One thing you have probably noticed by now is that although some stats were associated with the result more closely than others, positively or negatively, none of them was associated that closely.

Football is a simple game. So said Bill Shankly. And Ron Greenwood. Even Che Guevara. The team who score more goals win. Before a team can score they have to shoot, and before they can shoot they will nearly always have to pass the ball. Teams with the better passing and shooting stats win more often than they lose, but not all that much more often.

Football is a game that makes some sense but not perfect sense. Teams who take more shots win twice as often as they lose, but only twice as often. Because many games are drawn they win only just over half of all games.

Stats can help us give results some context. Knowing both we will understand more than we did when we knew only one. But even then there will be a lot that we still do not understand.

Not got a bet365 account? Sign up today and get up to £100 in bet credits

Up to £100 in Bet Credits for new customers at bet365. Min deposit £5. Bet Credits available for use upon settlement of bets to value of qualifying deposit. Min odds, bet and payment method exclusions apply. Returns exclude Bet Credits stake. Time limits and T&Cs apply.

CLAIM OFFER HERE

MORE FREE BETS

Today's top sports betting stories

Follow us on Twitter @racingpostsport