Andy Dawson: All football statistics are worthless*

Over the past couple of days, I’ve spent more than a little time gawping at heated discussions on Twitter between football writers arguing over the pros and cons of xG – the latest opinion-dividing stat that is supposedly aimed at giving a more precise reading of a team’s performance in front of goal. You know, as opposed to actual goals, that you can see with your actual eyes.

You’ve probably seen the xG score appear in the corner of your screen while you watch Match of the Day, in among all the rest of the stats that crop up post-match. Here’s my tip – free yourself from xG and the tyranny of its brethren. Ignore ALL stats, for they are completely worthless.

* Apart from match results and the league table – obviously. Oh, and the average player ratings in Football Manager. You’ll never get anywhere playing that game without those. But all the other ones can f**k off.

Possession percentages? Meaningless. The fact that a team has had 80% possession in the first half of a match looks impressive but if they’ve failed to encroach into their opposition’s final third, what does any of it stand for?

https://twitter.com/profanityswan/status/933279454150963200

https://twitter.com/profanityswan/status/933279624708198400

https://twitter.com/profanityswan/status/933280913399705600

Four-fifths possession sounds like total domination but what if they’ve been up against a team happy to soak up pressure and look to score from one or two counter-attacks over that 45-minute period?

The beauty of football is that it is simple and complicated at the same time.

You can generate reams of statistics about what has happened on the pitch, but each match is an individual drama with its own unique biology. Your side’s key midfielder might get an early red card or your left back could be having an absolute mare, tormented by a right-sided attacker who is having a stormer.

That’s what affects the outcomes of games – stats are just a bit of self-aggrandising window dressing, pretending to explain everything but in reality telling us nothing of actual value.

It all reminds me of this hilarious clip from Michael Palin and Terry Jones’ ‘Ripping Yarns’ about over-egging the football pudding…

https://youtu.be/lpuJBtcMMp8

This ludicrous new xG caper is just the latest piece of numerical snake oil peddled by a self-serving statistics industry that has got its claws into the game over the past few years. It gives every goalscoring chance in a match an ‘expected goal value’ (EGV). But the EGV of a shot assumes it is being taken by someone of average ability in the league, so right away the whole thing is rendered null and void.

It doesn’t differentiate between a chance by, let’s say, Ryan Shawcross, swinging a boot at a world class keeper like David de Gea from ten yards out following a corner, compared to Aguero banging one at a teenage keeper who’s come on as a sub for his debut.

I had a minor vent about xG on Twitter myself earlier today and received a reply which said:

‘Since football’s such a low scoring game, chance plays a massive role in the results – so match results and league tables are actually a really bad indicator of how a team is doing.’

Bollocks to that, frankly. The team is doing how they’re doing according to their results – they’re winning or losing matches. Those results have a direct effect on the pressure that is put on the players every time they take to the field.

Any other abstract conceptual stuff above and beyond that is guff. Conjuring up some flawed expected goals number means s**t all else, and there’s no tangible player confidence rating that you can factor into an xG figure.

For crying out loud, Sunderland beat Liverpool 1-0 in 2009 with the goal going in off a beach ball. The whole statistics industry should have been put in a burning bin that very evening.

I can see why statistics play such an important role in other, lesser sports such as cricket, tennis, baseball and American football. They’re all ‘short burst’ sports, where rigid, fairly repetitive plays make up the game – it makes more sense to quantify and analyse outcomes in those situations.

Football isn’t that. A football match and indeed a team’s season, is a living, breathing entity.

The increasing focus on pointless football statistics is a disease caused by saturation coverage of the game. If you’ve got hours of airtime to fill pre and post-match or you’ve got 24-hour TV and radio stations mainly devoted to football, you need something, anything to fill that time.

Just sit down for the kick off, watch the actual football, and go and put the kettle on when they start running through the numbers at half and full time, because YOU know what you’ve seen with your eyes better than some stat-spewing computer.