On the eve of the 2012 UEFA Champions League final between Chelsea and Bayern Munich, Petr Cech, the Chelsea goalkeeper, was given a two-hour DVD of every penalty kick that Bayern had taken since 2007. It was enough to calculate the probability of which way the ball would be placed when a penalty was taken. Chelsea went on to win the match following a penalty shoot-out and thereafter Cech said, “Six penalties the right way and I saved three, so basically the homework was very good.”
This March, at the MIT Sloan Sports Analytics Conference in Boston, Michael Niemeyer, Bayern’s head of match analysis, said of Pep Guardiola, the famed former manager of Barcelona and the current Bayern boss, “As he came to Bayern, the first thing he said was: ‘The match analysis department is the most important department for me.’ The second thing was: ‘I see a big part of my work in the auditorium.’ The auditorium is the place where he has video sessions.”
Football, as a sport, has long held out against numbers. Heart, soul, and the desire to win have always been cherished and rightly so, but the theoretical dimension of the game has been tough to take into account. The two instances above show that things are finally changing. In the words of Chris Anderson and David Sally who wrote The Numbers Game: Why Everything You Know about Football is Wrong, “The datafication of life has started to infiltrate football”.
Charles Reep, a wing commander with the Royal Air Force, was the one who started it. In 1950, he logged data for a match involving Swindon Town. He is said to have noted 147 attacks by Swindon, extrapolated from there and concluded that close to 99 per cent of attacks were failures. He may have been right or wrong but he proved to be a pioneer, going on to log data for more than 2,000 matches.
Kick off
The real number crunching, however, took off only in the mid-1990s with the arrival of companies such as Opta and Prozone and their high-end technology. What is the percentage of goals scored from corners? What are the pass completion rates? How effective are throw-ins? What do teams do in the first few seconds after losing or winning the ball? What is the recovery time for a player between matches? Opta records something close to 1,500-2,000 events, of which the above is just a microcosm.
“Every step on a football pitch is measured now,” Roberto Martínez, the current Everton manager, told The Guardian. “We monitor each session with GPS and heart-rate profiles. From a physical point of view, the most significant stats are probably the number of sprints, the sprint distance and the amount of high-intensity efforts a player gets through. We look at these through the season and they give us a good indication of how fatigued a player is and the recovery he needs.”
As a discipline, analytics is slowly shaping how teams are built, how games are strategised, and how prospective players are scouted. Football now increasingly deals with abstracts. Every part of the pitch and every area off it is broken down and analysed. What it does is to destroy preconceived notions and eliminate experience bias. Someone who has been in the game all his life might not be able to spot everything. There is finally the realisation that metrics help you see the unseen. And this has been proven by Arsene Wenger, manager of Arsenal, who studied economics and mathematics and turned to a fitness ace to solve his problems; by Sally and Anderson, behavioural economist and professor of politics, respectively, who wrote The Numbers Game; and by the maths wizards whom Billy Beane appointed to revolutionise baseball.
Getting the numbers right
Simon Kuper, a Financial Times columnist and co-author of Soccernomics noted how Manchester City winning the English Premier League title had much to do with data. “The analysts persuaded the club’s then manager Roberto Mancini that the most dangerous corner kick is the inswinger, the ball that swings towards goal. Mancini had long argued (strictly from intuition) that outswingers were best. Eventually he capitulated and City scored 15 goals from corners, the most in the Premier League.” City won the League on goal difference.
However, much depends on how one uses the data. Technology is such that when one buys into it, he or she does so completely. There are no half measures. The trick lies in what to see and how to see. As Nate Silver, the master American psephologist who was once a baseball analyst, said, “Most of the data is just noise, as most of the universe is filled with empty space.”
In 2011, when Liverpool signed Andy Carroll for a whopping £35 million, it raised quite a few eyebrows. The then director of football Damien Comolli thought that Carroll, who had the highest conversion rate from crosses, in combination with Jordan Henderson and Stewart Downing, both of whom had the best passing numbers and a high percentage of winning the ball in the opponents’ penalty box, would bring in a goal rush. But he didn’t see the data that said that crosses are not always an effective way to score goals. The plan as a whole was a spectacular failure.
What this calls for is more refined metrics. No one thinks that intuition will ever go out of the game. The players are the ones who win you matches. But data, by being dispassionate, helps them maximise their talent. It is said that nothing has altered the way football has been played in almost a century except for the modern-day offside rule. In the years to come, analytics might be the next game-changer.
