Never mind Jeff Stelling’s derision, expected goals – xG – is here to stay

Paul MacInnes on 22 November 2017

Jeff Stelling is known for being a cheery sort. The most exasperated he ever gets is when Hartlepool are drubbed and, even then, he usually bears it with the fatalism of the long-suffering supporter. This week, however, the Sky Sports presenter has starred in a much-shared clip in which he goes on an actual rant. And the subject? Expected goals.

In the run-up to their coverage of the north London derby, Stelling observed to his Soccer Saturday pundits that Arsène Wenger had done a daft thing. “He quoted expected goals!” Stelling said to an audible “No!” from one of his guests (perhaps Phil Thompson). “He’s the first person I’ve ever heard take any notice of expected goals, which must be the most useless stat in the history of football! What does it tell you? The game’s finished 3-1, why do you show expected goals afterwards?”

Distance from goal

Was it a long-range chance or an absolute sitter?

Was it a one-on-one chance?

Did the player have just the keeper to beat?

The assist

Was it a cross, a long ball, a through-ball, a cut-back?

The angle of the shot

Was it a tight angle or did the player have both sides of the goal to aim at?

The type of chance

Did the chance fall at the player's feet or was it a header?

In what passage of play did it happen?

For example: open play, direct free-kick, corner kick? 

Had the player just beaten an opponent?

This helps determine whether the player would have been under pressure

Was it a rebound?

This helps determine how prepared the player would be before taking the shot

Stelling’s remarks prompted much laughter in the studio and, after Soccer AM took a grab of the video, it got a lot of chuckles online, too. The sole exception perhaps being among the football analytics community, who were of the mind that expected goals is not actually the most useless stat in the history of football. That’s possession.

The basic concept of expected goals is simple. It answers the question: should a player have scored that chance? Or, when you add all the chances together, should a team have won that match? Defining each chance according to lots of different factors and then comparing that chance with similar moments in past matches, the outcome is a percentage: that chance, one time out of 10, would be a goal.

The question of whether this stat is useful to you depends on what you want from a game. A lot of people, increasingly known as “real football people”, don’t need a stat to tell them what they can see with their own eyes. Others, the analytics crowd or “football nerds”, think it provides context. Especially if you didn’t actually watch the game.

Dig down into what makes up an expected goals statistic (commonly known as “xG”) and things can get more complicated. There are lots of different takes on xG, compiled by everyone from the stats gurus Opta to individual fans. While they all might include the same basic information – location of the chance, where the ball had previously arrived from – each will have their own idiosyncrasies. Some might weigh the fact the chance fell to a striker’s foot rather than their head as being more likely to be a goal, but what happens when that head is Andy Carroll’s? There are lots of unknowns in any model and, when you put them all together, that might make the ultimate figure less reliable rather than more.

There is a bit of Old Man Yells at Cloud about Stelling’s outburst and that’s probably on purpose. Soccer Saturday is a bunch of old pros watching television, the ultimate “real football men” and perhaps the diametric opposite of nerds plotting outcomes on graphs. Its audience, meanwhile, are those people who turn on the telly to get their football information on a Saturday afternoon. They are unlikely to be keen followers of the latest trends in statistical analysis. On the other hand they might be more likely to see football as a simple game that doesn’t need to become any more complicated. Stelling is just giving his audience what they want.

There will likely be more gotchas of this kind in the months and seasons ahead (and whatever you think of xG, the subtext that Wenger only mentioned it because he lost to Manchester City is surely spot-on). But the broader point is that xG got a mention on Soccer Saturday. It also features regularly, if briefly, on Match of the Day.

Expected goals is a stat that’s here to stay and in just a few years time will be part of the furniture. The question is whether it is the end of the quantification of football or just the beginning.

Latest Football News