MoneyBall – Arsenal and the use of Statistics By Supertuur
The movie MoneyBall is about a baseball team, The Oakland Athletics, which uses statistical models to buy undervalued players in the market and analyse the game to get them more wins. This approach was needed as the Oakland A’s only had a limited budget for player transfers. The first year it was used the Oakland A’s managed to win 20 games in a row, which broke an American League record. Can the use of statistical model, which is now common in baseball, be used in football which is more complex.
The signing of Gabriel Paulista was the first time that Arsene Wenger mentioned something about the use of statistics and the purchase of a player. He mentioned to the press that statistically Gabriel looked good. The data was then used with scout reports of actually seeing Paulista play.
So where did Arsenal get their statistical data from? In December 2012, Arsenal spent £2.165 million to acquire StatDNA, a US based company to help them with the selection process of buying players. StatDNA staff analyse every game for 10-20 hours to get the most in depth data available. It is this now in-house statistical data that we use to get a more accurate assessment of a player’s performance.
Today an article in the Guardian talked about Brentford who playing for a play-off position in the championship league. Brentford a team with not much funds relies heavily on their youth academy and the use of statistical data. The owner of Brentford is Mattthew Benham, a hedge fund manager and professional gambler. He is also the owner of Midtjyland – a small Danish club now leading the Danish League by 6 points. Midtjyland is where former Arsenal player Kristoffer Olsson is plays and he was identified by their statistical software as a player with good potential.
Every game and every player is assessed based upon certain key performance indicators which, over the long term, the club believes are more indicative of success. They use data from E4talent, which tracks shots in the “danger zone” – an area that stretches from the six-yard box to the edge of the penalty area. 77% of all Premier League goals are scored in this area.
They also focus on set pieces where they scored 15 goals from 17 games, the second in Europe after Atletico Madrid. Interestingly Arsenal are the highest in the Premier League with 15 in 26 games.
Data is also used to give players a half time indication on how they performed against certain metrics. It also helps the coach with the press afterwards. As we can see in Midtjylands success, their approach seems to pay off.
Benham has made fortunes betting on football, this is not done by taking a punt. He has developed statistical models and exploited inefficiencies in bookmaker prices. It looks like data driven approaches are finding their ways into football clubs. What is good is that Arsenal seems to be ahead in this area, we need to if we want to outperform our richer competitors.