On Sunday, after 1500 games and nearly 27 years in charge, Alex Ferguson ended his reign as British football’s most successful manager.
When he was appointed to the helm of Manchester United in 1986, the latest Apple gadget looked like this…
…mobile phones looked like this…
…and the internet looked like this…
Since then, technological change has moved faster than Sir Alex sprinting towards the Fourth Official to protest an offside decision. And with it, the creation and use of information has risen exponentially.
Sport has led the way in showing how data can be exploited to make better predictions and improve both team and individual performance. Unsurprising, really, for a sector that, by its very nature, produces ream upon ream of statistics. Stadiums and arenas may still be the domain of the fast, the big, the skilful and the charismatic. But slowly, a new breed of sporting hero has emerged from the training grounds and changing rooms, gripping not bats, balls and other traditional tools of the sporting trade, but spreadsheets and analytics.
The most famous example is Moneyball’s Billie Beane (immortalised on screen by Brad Pitt). Beane has been General Manager of the Oakland Athletics baseball team since 1997 and has used the ‘sabermetrics’ data produced by the Society of American Baseball Research to ensure his low-wage team has sustained competitiveness against richer rivals. In 2002, the A’s became the only baseball team to have won 20 consecutive games.
A major innovator of Moneyball approaches in the UK has been David Moyes, Ferguson’s replacement at Old Trafford. Everton’s manager of 11 years was one of the first leading coaches in the game to formally use the database under-pinning Football Manager (yes, the video game) to identify both playing and backroom recruitment prospects. Similarly, in his previous role in charge at Preston North End, the former centre-half was one of the few non-Premiership managers to make use of sports data in this way, working with one of the field’s pioneers, ProZone.
At Everton, there was a full-time team of three data analysts working alongside Moyes, still using ProZone, and feeding their findings directly back to the players and their manager, to direct training and tactics. Other teams have followed suit with the application of data science. Arsenal are particularly effective with GPS technologies. And Opta has come to rival ProZone, working with other football teams, and in other sports. But it is at Goodison Park that an analytics-led approach has borne the most success.
For sure, the trophy cabinet has remained under-used but compared to their big-spending rivals, Everton have continually punched above their weight. ‘Pound-for-pound‘, they have arguably been the Premier League’s best team in recent years. In this 2012-13 season, Moyes spent just over 19 million on new players, the seventh lowest spend of all the top-flight teams. They finished sixth. The seventh highest spenders, QPR, it pains me greatly to say, ended the campaign at the bottom. Our manager famously struggles to organise information and use technology. The biggest spenders, Chelsea, whose transfer fees have been more than five times that of Everton’s, ended up only third. A consideration of wage bills reveals a similar pattern. Everton’s is currently the tenth lowest.
The blue half of Merseyside’s experience with David Moyes since 2002 tells us at least three important things about successfully exploiting huge amounts of data in sport, and in other activities.
One, it has to be committed to for the long-term. Applying quality analysis to ever greater datasets will not bring overnight results but it will bring significant improvements over the years.
Two, the analyses and their conclusions need to take place as close as possible in both time and place to the ‘frontline’. Everton’s analysts are based at the training ground, have access to the manager, full buy-in from the players; some of the findings are available directly after a match.
Three, context is everything. Whatever the statistics are telling you, you still need a human brain to parse true meaning. The statistics for a new youth scheme graduate used only against lower-league opposition in the early rounds of the Capital One Cup should be judged differently to those of the week-in, week-out journeyman battling in the league and European competitions.
Ultimately, data alone will not ensure success. In the context of getting 11 men to do the right thing at 3 pm on a Saturday, the tables and graphs would be as nothing without a manager’s leadership and the traditions and cultures of a football club. But gathered and exploited correctly, it can certainly give you that cutting, perhaps leading, edge.