Turning pro has spoiled stats game

At the end of last year this column took up the cudgels again in the GAA statistics battle, invoking Jorge Luis Borges in an effort to stimulate the debate.

(Because nothing stimulates debate like obscure South American writers — ed.)

Let me continue, please, before I get all up in your face with some Gabriel Garcia Marquez, yo. (Take the eye-rolling as implied — ed.)

Anyway, I happened to chat to a knowledgeable Gaelic football man recently and statistics came up.

“Stats lie,” said my pal. “I’ll give you a good example. At his peak David Beckham was the most successful passer in the Premiership, according to the stats.

“Do you know who was the second-best, according to the stats? Neil Lennon. The vast majority of Neil’s passes were five-metre lateral passes; Beckham must have had 25 assists a season with those great crosses.

“Possession, to me, is a very overrated statistic. They tell you little or nothing about the game, because what’s important is how you use that possession.

“Obviously there are some statistics that are immutable: if your free-kicker hits less than 75% of his kicks then you’ve got to get rid of him. He’s not good enough.

“But I don’t think you can analyse a game, certainly, from bald statistics.”

Fair enough.

But then something caught my eye a couple of weeks ago, a piece on Slate, the online magazine, about statistics in American sport.

American sport, as we all know, is the fons et origo of the obsession with statistics of all shapes and sizes, where the discovery of new categories of statistics makes people feel like stout Cortez on a peak in Darien. (Keats! I remember him from school — ed.) Very good.

In this piece, though, writer Jason Schwartz pointed out something very interesting that had nothing to do with percentages and completion rates and so forth.

Schwartz looked at basketball statistics and, in particular, a relatively obscure message board on the Yahoo website which had been set up to discuss such stats in great (translation: mind-numbing) detail over a dozen years ago.

What was interesting about that particular message board — which was regarded as the top place to show your love of hoop stats, nerds in their natural habitat, if you like — was that many of those posting their theories and findings had been hired by NBA teams fired by the full flush of the Moneyball craze.

As a direct result of that the spread of information, the invention of new categories, had stalled, as the last thing you want to do if you’re working for the Boston Celtics, for instance, is to share your new wonder-system for evaluating players with someone who works for the New York Knicks.

Interesting, no?

The more people get into sports statistics, the less information is disseminated, yet at the same time you have more chance of being hired as a sports statistician the more information you disseminate . . .

Forget Borges. Who knew sports stats were a natural home for Kafka?

(I’m dropping a fiver in the pretentious box even as we speak — ed.)

More in this section

Select your favourite newsletters and get the best of Irish Examiner delivered to your inbox


Saturday, December 5, 2020

  • 3
  • 8
  • 10
  • 20
  • 43
  • 44
  • 40

Full Lotto draw results »