Stats, Tackle, Jaap Stam

I am a big fan of Simon Kuper. His writing on football is among the best and he seems to be fascinated with the statistics and economics of football, just like I am (or at least think I am). In his latest column for FT, he wrote about the data revolution in football. To my surprise, he mentioned that Jaap Stam was let go by Sir Alex Ferguson in August 2001 not because of his autobiography but because of statistics. The statistics said that Stam was tackling less, therefore he must be in decline. SAF sold him and Stam showed that he is still a world class defender the years after that.

Upon reading that my mind went to Frank de Boer, Stam’s defensive partner in the Dutch national team. When he started as coach of Ajax, during one training session Frank de Boer hammered the two rules of defending: never tackle and never turn around to block the ball. The “do not tackle” is of utmost importance and I understand why. It is much better to defend by positioning yourself properly so that the opponent can not get the ball. A tackle is the last effort when everything else fails. And a tackle is very risky. A slight misjudgment and you can cause a foul, resulting in a free kick or penalty, with probably a yellow card or even a red card from the ref. Why risk it? Having said that, Frank de Boer has made a tackle with this perfect timing.

Frank de Boer learnt all this at Ajax, where he excelled as a defender. Jaap Stam on the other hand was a hard and tough defender. At the age of 22 he was still playing in the Dutch second division, while Frank de Boer at that age was already representing Holland. However, during the late 90s their partnership blossomed and at 1998 they hit their peak.

When it comes to tackling, England is obsessed with it. A good defender is one that can tackle all the time and tackles are dished out left right center. Stam became a good fit for Manchester United as he was a hard defender who tackled first and did the rest later. Nemanja Vidic is like that too.

I’m sure that playing with Oranje taught Stam some lessons. That positioning is more important and that tackling should only be done when everything else failed. He took that with him to Manchester and it resulted in less tackles. A statistician who did not study the background and development of Stam concluded that because he tackled less he is in decline. Sir Alex Ferguson listened to this statistician and he has then called it his biggest regret.

This then goes back to the data revolution in football. There is no doubt that data is very important. But more important than that is what is behind the data. Those quirks are not easily deducted from data. An assist is a pass that led to a goal. Two players can have 3 assists but the quality of those can vary wildly. Xavi gives an assist with a through ball in space that liberates a striker, while Walcott gives an assist by running very hard and crossing the ball in front of the goal and hopes for the best. The differentiation of these assists are not (yet) visible in the football data that we can see (maybe it’s there in ProZone, but not yet on the apps or chalkboards that I have access to). Therefore it is hard to judge which player is better if we just look at the number of assist.

So even though I am a bit obsessed with football statistics, I am also very cautious with it. Some players don’t score goals, don’t give assists, don’t run much and don’t make tackles yet are very important to a team. Football is more than statistics, yet the statistics is very important to the game. If only I can break that code…