Stats in football, a musing for both ‘sides’

[EXTERNAL SHOT. A tall tower, imposing, penetrating upwards into the dark and stormy night sky.]

[INT. An air conditioned room, atop the tower. Banks of computers line the curved walls, minions coding, plotting, blankets wrapped around them to protect them from the chilling freeze that the ventilation system and the black holes where their warm, loving hearts would be is creating. Next to each is a winding wisp of steam, rising from mugs of freshly brewed, artisanal coffee, brewed exclusively from beans selected by algorithm, stewed in the boiling mass of proper football mens’ tears. In the centre of the room is the frame of a goalpost, a lone shadowed figure leaning against it, menacing, the embodied metaphor of the encroaching threat of loveless calculations into the game of passion, the nation’s game]

The above is an extract from a documentary on the use of statistics in football, currently in some stage or other of production. There is a high percentage chance that the previous sentence is a lie. It is, however, a tongue in cheek representation of ‘stats nerds’ that have, over the last decade, found more and more work within football clubs. It’s also probably not useful in any way. I do get actually serious in the next paragraph, so keep reading.

Statistical analysts aren’t evil or threatening, and nor do they exhibit the kind of kernel qualities that these hammed up stereotypes would emerge from. The most common complaint about the use of statistics in football is that they simply can’t be used. The average punter on the street who watches games at the weekend doesn’t fear walking bespectacled asthma inhalers ripping the heart out of football, they fear their club screwing up and losing matches because of them.

They fear that a guy with a computer, probably of some fairly middle-class background who got picked last for the sport in PE, will look at a laptop and tell their club to spend a quarter of what they got for their star striker on Dejan Lovren. That these kinds of opinions are taken when they don’t even know if the stats are ‘right’, and that they’re overruling the opinions of those who’ve watched these – and many, many other – players in the flesh.

The underlying fears about underlying numbers, then, are about their efficacy and their influence. And these are two things that your average punter (as well as a good bunch of stats twitter) doesn’t know a lot about, to varying degrees. 

For obvious reasons, clubs are reluctant to reveal their hidden secrets, whether the heat in those decision rooms is coming from cigars and pies or cappuccinos and processors, and just what kind of processes those processors are doing. Of course, in reality, the heat in those decision rooms will be coming from a bit of both, as long as the pie doesn’t fall between the cracks in the keyboards and make a mess. 

Stats are also a useful new bogeyman, a scapegoat for bad transfer decisions. Few called for the firing of the Liverpool scouting department when they bought Andy Carroll for £35 million, or Chelsea when they spent £50 million on Fernando Torres to let that transfer happen. Yet people scoff and swear at the presence of data analysts. And it’s also not known how much data analysts had to do with the Lovren transfer, for example. An existing narrative of ‘stats transfers’ and ‘Rodgers transfers’ suggests that the stats picked Emre Can and Robert Firmino while it was the ‘old fashioned’ methods that picked up Lambert and Lovren. But we don’t know, that could all be a myth too.

With much of the statistical analysis in football either unavailable or incomprehensible, the relationship between stats and non-stats in the public sphere has to have a large dose of trust. But, at the moment there isn’t much evidence out there, glowing garish neon-yellow and bright silver, while there are still the horror stories of old. The punter on the street hasn’t yet come to realise that the Boo Radley bogeyman should be treated like any other human being. Stats and algorithms, the ones being used in the game way above our heads as they may be, are a tool; watching players, as time-consuming and also riddled with difficulties as it is, is a tool too.

Whether ‘believers’ like to admit it or not, the ringing of the stats bell, sandwich board across shoulders and bible in hand, can easily appear haughty and arrogant, particularly when discourse is limited to a grab-bag of words and acronyms on Twitter. The constant improvements – or ‘improvements’, because there will inevitably be yet another one soon – the unknown and the uncertainty, while simultaneously giving off the aura of the necessity of the approach, can take this use of stats out of the pubs and into the ivory towers.

The path that I see is a wide one, open to all. With increasing availability and quality of data, stattos will be able to apply their algorithms to more and more areas, stretching wider and deeper down the league structures of the world. They’ll be able to draw up a long list of candidates for a role the club are looking for. A meeting can take place, scouts giving their views on players they’ve seen or heard about, gradually crossing names off and whittling down the list. The ‘new breed’ of analysts will do further work with data and video on this shortlist, while the scouts will do their work going to games and putting feelers out. In some part of the future there will be people who’ll have skills in each, hybrids spanning the imaginary divide.

This is probably how it works at clubs at the moment. But, as mentioned above, we don’t know, and for understandable reasons, however unhelpful for the public this may be.

The new breed of analysts are only to be feared by those of the old breed who aren’t sufficiently good at their jobs. Statistics themselves are certainly not to be feared, only bad application of them, just like bad managers are to be feared, bad coaches, bad scouts, bad owners, bad players. Trust needs to be gained, somehow, whether through making the complex simple, the abstract applicable to real life, or something else. 

[Final scene. EXT. The generals of their respective armies face each other on the battlefield, bloody from the gruelling fight they have already been through this long, long day. Through dirt and sweat and suppressed fear on their brows, they look each other in the eye. Both, at the same moment, raise a hand, a signal to their armies. Stop. The war had gone on long enough and in that short second they understood each other for the first time. One dropped their broadsword, the other their more modern, lightweight blade of choice, the stained metal heavy in the dewy grass. Slowly, they take tentative steps towards each other, still wary – then embraced each other, as friends.]



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s