download All Football App

Expected goals (xG) is among the most useful stats in football... and here's why

  /  autty

They squawked. They shouted. They put it down. And, at the end of it, they ran Expected Goals (xG) out of town with all the tolerance of a group of villagers towards Frankenstein's Monster.

xG is a new metric that has been introduced by Opta this season. It has been trotted out on Match of the Day since the first weekend of the campaign, but not many people seem to understand what it is.

Among them are Jeff Stelling and the panel on Sky Sports' Soccer Saturday. The host responded to claims from Arsene Wenger that Manchester City are not unstoppable. The French manager had used xG to make a point about his team's display against the Premier League leaders.

And firing back, Stelling ranted: 'The daftest thing he said this week, he quoted Expected Goals, you know the little stat that you see? This was expected to be a close match. He said: "Look at the expected goals. 0.7 for them and 0.6 for us".

'He's the first person I've ever heard take notice of expected goals, which has to be one of the most useless stats in the history of football.

'What does it tell you? The game has finished 3-1. Why do you show expected goals afterwards? It's absolute nonsense. If he really believes that… well, I don't think he can.'

The problem is that Stelling is fundamentally wrong. There are far more useless stats that are regularly used to analyse football matches. And xG is among the most useful.

Comparing shots and shots on target is like deciding who had the best dinner based on the number of plates used. There's no indication of how well cooked the meal - or in the case of this laboured metaphor, the chance at goal - was.

And another that gets trotted out regularly is possession. Again, that tells you very little. It simply measures the number of passes a team have played in a match.

What it does not do is talk about the quality of possession, where a team had the ball or in what direction passes were played. There's a reason Leicester won the Premier League with an average of just 43 per cent possession.

Compared to stats like these, simple numbers that have not really been analysed, xG is an iPhone being slapped down on the table among Nokia 3310s.

It is a really good metric for understanding just how well a team is actually playing. The name is something of a misnomer. Really, all it does is show the quality of chances that a team are making.

It takes in factors like the distance from the goal, the angle, whether it was a shot or header, the type of assist, if it was a rebound, if it was one on one and much more. Over 300,000 shots were analysed for Opta to put the metric together.

Think about it this way. You're stood behind a goal. For the entire first half your team pelts the net from close range only to be repelled by a goalkeeper embodying the spirit of Gordon Banks, Peter Shilton and David de Gea. It's capped off by a ludicrous stop from inches out.

And then in the second period you watch as an opposition player turn 40 yards out before pinging one straight into the top corner. It's the only goal of the game.

How do you feel when you leave the stadium? Aggrieved, right? Understandably, you'll feel your team were unlucky.

It's tough to quantify luck though, isn't it? So you go home and by the time you get there, you're calling for the manager's head. A loss is a loss after all.

What xG lets you do is look at a defeat and understand if it will always lead to losses. If your team has a high xG number, it suggests that over time you will revert to the average and start picking up victories.

Which is why, game to game, xG can often be flawed - it is such a small sample size. Over the course of a season, though, it does tend to even itself out.

Scoring a wonder goal is all well and good. The problem is it sometimes just papers over the cracks. Pajtim Kasami scored one of the best goals in Premier League history in a season when his club, Fulham, went down. This is because the chance of him scoring that particular goal on a regular basis is so statistically unlikely.

Whereas a match with a team scoring four tap-ins might not be as aesthetically beautiful but it does suggest they have the capacity to sustain it.

Which is why City, as explored elsewhere, are really good this campaign. They do not rely on goals from range. Gabriel Jesus has not scored any of his eight Premier League goals from outside the box.

So put down the pitchforks. xG is not here to destroy the game - just take analysis beyond 'he has to score that!'.

Wenger used to be known as Le Professeur but you don't have to be a genius to understand the line he trotted out. Stelling and Co would do well to give it a go.