Sabermetrics: Revolutionizing Football?
As a football enthusiast and as a (at the time future) BIM student, I was excited to learn about the concept of sabermetrics during the last football season. Sabermetrics is the analysis of in-game gathered player data, essentially ruling out any emotion or subjectivity when it comes to judging a player’s performance. The system gathers statistics on key performance indicators – goals scored surprisingly not being one of them. Then, the analysis determines exactly which player would with best in the current team, using the current players’ data as well. What’s more, sabermetrics is able to cross-reference results from different leagues with statistical analyses to build a continent-wide ranking of football clubs. The purpose is that a club’s management does not have to invest heavily in a new player, or even do the effort of scouting a new player – all it has to do is let the system crunch some numbers and whoever comes out will be the right choice for the team. The exact methods of the system are kept secret, but for a simple example calculation, take a look at the video below (the explanation starts at 3:26).
The analysis method was invented to analyze baseball statistics, and sabermetrics rose to fame in 2002 when the Oakland Athletics heavily outperformed larger baseball teams despite their modest budget. It has recently been introduced in European football as well. English side Brentford F.C. adopted the method two years ago and was promoted immediately from the third to the second tier of English football, finishing at a respectable fifth place last season. F.C. Midtjylland of Denmark – whose holding company is owned by Brentford F.C. owner Matthew Benham – adopted the method last season and became Danish champions for the first time in their existence.
Midtjylland in particular applies an extreme form of sabermetrics. At this club, the management values data analysis over anything else. When they were at the top of the league towards the end of last season, the club’s chairman, Rasmus Ankersen, said:
“The model rating always, in any case, overrules the league position in evaluating our performance. The table is lying. This is what I keep telling the coaches: don’t think we’re good just because we lead the league. We’re good when our model says we’re good.”
Applying sabermetrics also allows Midtjylland to allocate its assets differently and more efficiently. For example, scouts no longer judge the actual quality of a player, but rather his personal and psychological fitness. Also, sabermetrics enables the club to formulate more precise and more effective training schemes. These aspects of sabermetrics are perhaps the most compelling. If data analysis can completely transform the way all aspects of football are managed, what does that imply for the future of football? Will data analysts and software engineers take charge? If so, this would make way for a revolution in European football. A revolution that is driven by data analysis rather than by money, for once.
I am curious whether this method will pay off. While I can see how data analysis can be applied directly to baseball, football is a whole different ball game, literally. Interaction between players and randomness are much more prevalent than in baseball, which is a relatively predictable sport. Besides, FIFA president Sepp Blatter, considered somewhat of an authority in the world of football, has stated that emotion and subjectivity are an essential part of football.
What do you think? Can data analysis successfully play a leading role in decision-making processes at major football clubs?