Many academic disciplines within the humanities and social sciences saw a flowering of intellectual discourse in the post-war period throughout Europe that came from anxieties about the direction of the post-war world, were given a physical expression in the student revolts that occurred in Paris and Prague in 1968, then their intellectual expressions came in the linguistic and cultural ‘turns’ that many of those disciplines took throughout the 1970s with major works from the likes of Michel Foucault, Jacques Derrida, Luce Irigaray, Judith Butler and Hayden White all forming some of the key texts of the philosophical moments known as the cultural and linguistic turns. Similarly, football has also in the last quarter of the twentieth century and on into the twenty-first century experienced its own moment of crisis too perhaps, that could be described as resulting from what we might call it’s ‘statistical turn’.
As Jonathan Wilson notes introducing his magisterial review of football’s tactical development since the late nineteenth century, Inverting the Pyramid
the history of tactics, it seems, is the history of two inter-linked tensions: aesthetics versus results on one side and technique versus physique on the other… and beauty – or at least what fans prefer to watch – remains very much in the eye of the beholder.
These are tensions that the statistical turn, in some ways, try to negate – to find, through quantifying, the correct balance of the aesthetic as pragmatism (this is pragmatism now is both sporting and economic: no’ rational’ consumer, at least, should want to pay to watch football without a modicum of aesthetic sense) and for the right balance of technique and physique for the correct situation.
After all there is a symbiotic relationship between the statistical turn of football and the changing nature of tactics in the game. And it’s not just football of course. Indeed, football is a late comer to the statistical obsession of sport. Cricket has a long history of logging minutiae, as does baseball. Now even Gaelic football is having its moment of statistical awakening with the publication of Tactics not Passion. The obsession too in recent times with applying the theory of ‘Moneyball’ so ably utilised by the Oakland A’s in America’s Major League Baseball, to the game of football, and in particular in the Premier League is further proof of the desire to economise, to distill knowledge about players, positions and their stats into the most economical means of achieving the highest results in the game. In some respect’s there is a philosopher’s stone, alchemical aspect to this pursuit. Ostensibly, ‘moneyball’ should be applicable to not only other sports, but specifically to many different teams and produce the same results over and over. Ultimately, this is the crux of the matter: lies, damn lies and statistics.
The statistical turn in football is to be welcomed, but not all stats are either a) equally valid or b) equally useful. Oftentimes, it has been my impression that the statistics rolled out on television coverage of football tell you very little – they only give the impression of furnishing you with further analytical tools.
For instance, the obsession with tracking the number of kilometres a player runs, as though a huge number is meant to impress us. Certainly, I find it impressive to consider how much many players run over a sustained period, but that doesn’t necessarily tell you anything beyond the fact that these are highly-trained athletes of considerable fitness. But, frankly, this I already knew. Take Dimitar Berbatov for instance. I’d be willing to bet that Berba runs only a fraction of the amount of many strikers, or players full stop for that matter in the Premier League. So if you were to compile a league table of the average distance covered by all strikers in the Premier League at the end of the season (say of those who start a majority of matches) you’d probably find Berba quite low down on that particular list. However, if you were to then judge that against the number of goals scored and got an average of ground covered versus goals scored, I’d be willing to bet Berba’s would be amongst the most economical. He can often cut a lethargical figure on the football pitch – he doesn’t do much running but then that’s because he doesn’t have to, usually. His greatest skill is his sense of positioning firstly, and secondly his ability to turn a defender. It would tell you more to figure out the statistical likelihood of Berba turning a defender in a particular area of the final third of the pitch and thus encourage him to drift to that side and to have passes played to him on that side, thus using the available statistics to bolster performance and (you hope) his likelihood of scoring.
To carry on with the kilometres-run example, the stat is further only really properly useful if you find out how many of those a player runs with the ball in his possession (if a full- back, attacking midfielder, or striker) or how little running he needs to do to either retain, get or pass the ball if a central defender, defensive midfielder etc. By knowing this, it will be possible to figure out if the player can be doing less running and retaining the same amount of possession, being more economical with their own fitness levels and so forth. Of course, this is the kind of statistical breakdown that clubs are increasingly engaged in, and their knowledge will get better with the greater back data-sets they have to work with. The longer they continue to count all of the minutiae the more historical knowledge about a player (or group of player)’s habits they will be able to draw on to inform the tactical decisions of the team. But the statistical data presented on television coverage seems to me to exist more to obscure knowledge and to give the impression of an unknowable, ironically hard-to-quantify special insight that only those on the telly have available to them. Their job, like the jargon of the pseudo-intellectual, is to obscure rather than inform.
Encouragingly, I’m not the only person to think so. In an interview with ESPN in 2011, Chris Anderson of Soccer by the Numbers said that:
I personally think teams and fans should ignore stats like distance covered, or things like possession or pass completion percentages. I don’t think they currently tell us all that much. Instead, I think stats for particular areas of the field or in particular situations are very interesting. So, for example, I think it would be quite telling to hear what a player’s or team’s pass completion percentage in the final third are, or when making forward as opposed to lateral or backward passes. I think fans should demand from you guys at ESPN (and not just you, of course) that particular stats (like pass completion) be put in perspective. How much better or worse is someone performing relative to players in similar positions or situations?
The reality is a lot more boring. The question of soccer’s lack of “discrete events” like those in baseball (pitching, batting) to analyse was addressed by Leeds University professor Dr Bill Gerrard in an influential paper for the International Journal of Sport Finance titled Is the Moneyball Approach Transferable to Complex Invasion Team Sports? His thinking broadly sums up the convention view of the potential of soccer analytics:
Replicating Moneyball and Scully’s pay-and-performance analysis in invasion team sports such as the various codes of football, field and ice hockey, and basketball is much more problematic. Invasion team sports are much more complex and hence the separability of individual player contributions is considerably more difficult.
Most people writing on the subject place huge emphasis, rightly, on the work of Simon Kuper and Stefan Szymanski first published in 2009, titled Why England Lose in Europe and Soccernomics in the US (no doubt to capitalise on the success of Levitt and Dubner’s Freakonomics). The book has gone into a second, updated edition now – certainly there are aspects of it that are revelatory, if one some counts it also proffers common sense advice. Whether the failure of Liverpool since its takeover by US owners obsessed with the ‘Moneyball’ model, the statistical turn in football has little to offer or if the statistical turn is still in its infancy and so the wrong things are being noticed; whether there is to be some compromise between the view of the game as a randomised mess, a text without an adequate theory and something wholly quantifiable one thing is absolutely certain: football’s statistical moment has come and there will forever be the game before its statistical turn and the game after that moment.