Lies, Damned Lies, and Statistics: formulas and formations in modern football

Stats can be used to ‘prove’ anything but football’s obsession with facts and figures threatens to kill all the individuality of the beautiful game

The mid afternoon sun burns fiercely as the sweating crowd in the Estadio Jalisco in Guadalajara rises as one seeking to know what happens next. The ball from Tostao is inch perfect, close enough to bring under control but dropping sufficiently short to tempt the man between the posts. The forward sees the keeper’s taken the bait: a bull to a red rag. He has two options: take it left, into the path of the stampeding custodian, or take it right, away from goal where the angle is tight. Being Pele he does neither: he doesn’t touch the ball at all.

He shapes as if to go left, waiting for the Uruguayan to commit, then leaves the ball to roll. The keeper’s been fooled, turning like an oil tanker in a storm. Taking it early, Pele wraps his foot round the ball, seeking the opposite corner. The defender on the line won’t stop it, already in a different post code as he flies in the wrong direction. But the net doesn’t bulge and there is no goal, the ball rolls tantalisingly wide. Shot off target.

Pele’s dummy against Uruguay in the semi final of the 1970 World Cup is an iconic moment, notwithstanding the fact it didn’t result in a goal. It displayed ingenuity and individual genius: out of the penalty box thinking. And yet, if you were to look at it in terms of the facts and figures what did the dummy achieve? In today’s numbers dominated game he would probably be criticised for not sliding it across goal, increasing the chances of ‘maximum conversion’. That is what football has become: a sport of spreadsheets rather than spontaneity. 

As Sean Ingle, the sports editor of Guardian Online writes: ‘nowadays almost everything in football is tracked and crunched. Prozone monitors players' movements every 10th of a second...Opta actually has a model that quantifies the percentage chance a shot has of going in’. Not he makes clear, is that necessarily a bad thing: ‘while some data is squirreled away by those whose livelihood depends on an edge – big clubs and professional gamblers – much of what remains would enhance our understanding of the game.’ And he’s right – within reason.

Data can open our eyes to hidden possibilities and its use within the game, where every team and player is looking for that edge is as understandable as it is inevitable. The problem comes however when those in charge, those coaching, become utterly obsessed with these numbers to where everything becomes regimented: less sporting activity more military operation.

Systems such as Pro-Zone can be a great asset but too many are becoming over-reliant, letting it cloud their instincts. Using it to constrain a team’s talents rather than enhance them. Today you’ll see a player attempting a defence splitting ball or trying that piece of magic that will convince a kid watching to take up the game being shrieked at by a manager for giving the ball away, not thinking about the pass completion rate.

An over reliance on data can also create self-fulfilling prophecies: stats not telling us anything new but rather reinforcing old assumptions. We are told how important set pieces are because of just how many goals they result in. Looking at the raw numbers, the technocrat coach becomes reluctant to attack in open play, afraid that more unpredictable, less regulated modes of attack are futile. A team should play for corners because that’s where the hard work of studying probabilities and percentages comes to the fore. So set pieces do become where people score a lot of their goals, because that’s where managers have focussed their attentions.  

It’s a situation where increasingly everything is looked at in black and white but without the context. ‘We deserved to win the game: we had 65% possession and 7 more corners, not to mention that our shot count and attempts on target were far higher’. Arguments popular in the modern day post match post mortem, however the ‘statistical analysis’ on which these protestations are formed often lack a wider perspective.     

In this instance the majority of the possession could have been going nowhere while the ball was rolled painfully slowly between the teams two centre half’s. From all those corners how many caused the sort of wild panic that spreads when the ball bounces around the six yard box like a loaded grenade? ‘Shots on target': the majority daisy cutters the keeper didn’t so much as have to get his knees dirty to deal with. Stats can be used to ‘prove' anything. And then there are stats to disprove that

.

In a world of diminishing resources, using data and information to improve efficiency and performance is important and is something that can be applied to the beautiful game however sport cannot be programmed like a computer. Sport is unpredictable, spontaneous and exciting: it’s why people watch but too many coaches in the 21^st century want to reduce games to a robotic series of set piece battles with troops deployed in position doing as they’re told. Individuality is frowned upon, little as possible left to chance. Mavericks and entertainers a dying breed, replaced by robotic drones that can run all day, are a standard 6 foot tall and while lacking in individual ability, put all the right ticks in the boxes the algorithm has spat out.

Those in the game need to remember to open their eyes, not hide behind pie charts and percentages. Watch the game, feel the game, not let it become a series of calculations devoid of all the imperceptible possibilities that create the drama. Memories of Pele live long for the magical moments of unpredictability rather than what results from the use of maximum likelihood estimators. Football evokes a passion and through all the measuring of stats and numbers we should never forget it is the conjurers of the game casting their exquisite spells that fuels it.