The Numbers Game: Why Everything You Know About Football is Wrong has been this summer’s best selling sports title. It is about to appear in an American edition and is being translated into Portuguese (for Brazil), Italian, Finnish and Japanese among other languages.
It was therefore a great pleasure to go down to London this week and talk to American co-author Chris Anderson about the book over lunch. He and David Sally wanted to write a serious piece of analysis, but also a book that would appeal to football fans.
The Numbers Game: Why Everything You Know About Football is Wrong has been this summer’s best selling sports title. It is about to appear in an American edition and is being translated into Portuguese (for Brazil), Italian, Finnish and Japanese among other languages.
It was therefore a great pleasure to go down to London this week and talk to American co-author Chris Anderson about the book over lunch. He and David Sally wanted to write a serious piece of analysis, but also a book that would appeal to football fans.
Hence, they used an intermediate level statistical technique: scatter plots and two variable linear regression. It’s something I used in my PhD thesis forty years ago. Econometrics has moved on, but this is a reasonably accessible technique and, in any case, the book is structured around a series of stories and anecdotes which make for easy reading.
There are many insights in the book and I look forward to discussing them in the context of relevant events over the next few months. However, I want to focus on what I think is possibly the key insight in the book: that football is a weakest link game. If the authors are right, the implications for strategies and tactics are considerable. Indeed, it is evident that Anderson has already broken into the closed, village world of English football and has visited the training grounds of top clubs (he is on a two year stay in London).
What they mean by football being a weakest link game is that ‘success is determined by whichever team makes the fewest mistakes, whether they are individual or collective.’ While football is a team game ‘it is one prone to being decided by sheer, staggering individual ineptitude.’ Perhaps those fans who identify scapegoat players are not so wrong after all (although whether they identify correctly is another question).
As the authors admit, a great player can make a rare mistake that decides the course of a game. Even so, ‘upgrading a weak link can help a club more than improving its best player.’ The results show that performance differences in weak links are 30 per cent more important when it comes to goal difference, and almost twice as important with regard to points per game.
An implication is that ‘teams who spend their summers lavishing millions on recruiting the latest superstar may have it all wrong.’ Who wins games and championships depends on ‘the incompetents at the heart of the defence or the miscommunicating clowns in midfield.’
So why did Real Madrid spend a lot of money on Gareth Bale, way beyond most estimates of his value as a footballer? As the authors state, superstars ‘provide the glamour, the genius, the moments of inspiration. They sell the shirts and fill the seats.’
As we have argued before, signing Bale was as much a commercial decision as a football one. It was a matter of prestige, a means of putting down a marker and saying ‘We are Real Madrid, the world’s top team, and we pay whatever it takes to get our man.’
One could also argue, as the authors would, that coaching or acquisition decisions are often not ‘efficient’ in an economics sense. As they make clear in, for example, their account of Everton, football is increasingly a numbers game. Quantification and numbers are taken seriously.
That does not mean that there is not a substantial element of chance which gives the game its appeal. Their data in the weakest link chapter explains about a third of the variation, which is a good result by social science standards. I remember running one of my first data sets as a MSc student and getting .3 as my best correlation. My instructor said, ‘welcome to social science.’.
Anderson was once a goalkeeper and one of their forecasts is that the gap between the salaries and transfer fees of strikers and defenders and goalkeepers will shrink significantly. As it is, goalkeepers are the cheapest position, although they can play a vital role in deciding the outcome of a game.
As a fan of goalkeepers and quality defenders, especially defensive midfielders, I hope they are right, but I don’t see much sign of change yet. But perhaps the lessons of their book will sink in over time. But, even today, football can be a very closed community.