I started counting on my fingers …
November 11, 2010 § Leave a comment
An interesting after-effect of the takeover of Liverpool by New England Sports Ventures has been a susurrus of chit-chat about sabermetrics. Depending on your point of view, it’s either the future of football, or a pernicious (read: foreign) idea that has absolutely no application in football. A tiny nadir for the second perspective can be found in this piece from Mike Calvin’s stunningly trite and weirdly free-form blog. The piece handily synecdochises itself thusly:
Football, unlike baseball or cricket, is not a statistics-driven game.
Football fans operate on blind faith, ritual optimism, and desperate hope.
Now, the second is just a fancy way of saying that football fans are stupid. But as for the first: to say that a sport is statistics-driven is to imply that the statistics are a vital component of the game, or that without them the game would be diminished. Yet even a cursory consideration of baseball shows that this isn’t the case. It can be played to a high level or watched with great enjoyment with nothing more complicated than a basic understanding of the rules, and it doesn’t matter if you spend the seventh-inning stretch trying to put salt on your slugging percentage.
If you were to watch just one game, however, you would take away an exceptionally narrow perspective on the abilities of the players involved. This is self-evident, and true of any sport. (For instance, you might buy John Jensen on the strength of this). So this is where a range of information comes in handy: one goal in one game can be part of a run of one goal in forty, or forty in forty.
It’s clear that baseball is more straightforwardly stattable (definitely a word) than football, in that it comprises a series of easily quantified events that remain constant across games. Each game is divided into discrete instances, all of which begin with the ball in the pitcher’s hand and end with one of a relatively small number of outcomes: walk, strike-out, on-base, error, and so on. These instances are the raw data, which can then be crunched into trends and averages and percentages and equip you to turn to your friends and say “That Babe Ruth was a handy player”. Then sabermetrics emerged from the realisation that traditional stats might be misleading – that a team’s batting average doesn’t help much when calculating their likely runs, for instance – and so deeper number-crunching was needed.
Even pre-sabermetrics, baseball was significantly more ‘statted’ than football. This might be due to football containing fewer similarly discrete instances (penalty shoot-outs aside). A football match ebbs and flows, and so the statistics that dominate footballing discourse tend to be those that result in something tangible. Most obvious and crucial is the goal (and its good friend the assist), then comes the tallying of throw-ins, corners, fouls, yellow and red cards, and so on. Each of which tells a focused, highly specific and limited story.
Take Laurent Koscielny, who – despite being the very image of a Wengeresque centre-half, all charmingly affable interceptions and slightly questionable positioning – has two red cards in the Premier League this season. Karl Henry has one, while Nigel de Jong has none, and yet red cards are often held up as a measure of a player’s, and a team’s,”dirtiness”. Statistics may not lie, but they may not always be saying what they’re taken to say.
Which is not to say that a baseball statistic considered in isolation is any more informative – that’s why they are so many, to allow them to talk to one another – but to note that football statistics tend to be presented without any context or depth. In the case of Luckless Laurent, both his dismissals have come in the late stages of a game Arsenal were chasing, for non-violent offences; a considered analysis of red cards might incorporate time remaining, the state of the game, nature of offence and so on. And so a considered, deeper analysis of Koscielny as opposed to, say, Karl Henry, might provide some context for the fact that Henry has half the reds. (Coming soon: Twisted Blood’s comparative red card assessment index.)
The fact that baseball statistics can, if used correctly, present complex and so illuminating information regarding a baseball player’s capabilities doesn’t mean that the sport is statistics-driven, just that it is statistics-rich. The game is the game, and is distinct to the analysis of the game. But for those who need to deal with the how and the why of wins and losses, which means being aware of the good and the bad of players, it is inescapable: the more you know, the more you know. Nobody wants statistics to drive, and they wouldn’t be able to anyway. It might not be too bad an idea to let them help with the navigating.
PS. Apologies to any more baseball-literate readers if my terminology is a touch out, or my characterisation slightly odd. It’s not so long ago that I thought Fielder’s Choice was a brand of tobacco.