Comments on: Do stats tell the story?

By: TheScout

TheScout — Thu, 13 Mar 2008 15:06:34 +0000

From what I’ve read Opta has multiple spotters at each match that compile their statistics. Most sports stats services use a similar system where people are assigned to track specific things during the match and focus solely on those things.

You don’t have a situation where one person is trying to track passes, tackles, shots etc for both sides in an entire match. That would be a near impossible task for one person.

By: sam

sam — Wed, 12 Mar 2008 11:54:28 +0000

I wonder how to found statistics in a given match? If it is by collecting tally for every shoot and pass, I think the validity and releability of the stats will be under question mark?? So if you have an answer please tell me.

By: Kartik

Kartik — Mon, 10 Mar 2008 19:46:39 +0000

Bill Beane as you mentioned is running the Quakes, or partially running the team. He’s hard at work from what I have been told (a source fairly close to him) devising similar scouting tools.

By: Darin

Darin — Mon, 10 Mar 2008 17:50:45 +0000

Billy Beane of the Oakland A’s baseball club developed a statistical system to find good performing players who are much cheaper to buy than the Jeters and A-Rods of the world, and it has been received well. He is somewhat involved with San Jose of MLS (not to mention an EPL fan), and has been supposedly trying to develop a similar system to find quality value players for MLS’ very limited salary restrictions.

I don’t know if he can even do such a thing for football like he did for baseball, but if he can it would be of great benefit for teams all across the globe. I’m sure it’s been attempted before.

By: jm

jm — Mon, 10 Mar 2008 12:35:55 +0000

Kartik,

Your argument is misdirected. It has absolutely no bearing on the “usefulness of stats.”

Rather, your argument points out the uselessness of some uninterpreted statistics. Well, that’s trivial. Statistics without context, without interpretation, are often mere counters. They are records of events. You cannot directly infer from a record of events to an evaluation of those events. That’s why you cannot infer from “Vinny Testeverde is 6th all time in passing yards” (or whatever he is) to “Vinny Testeverde is the 6th best QB of all time.” That’s just simply not what those statistics mean.

But of course they are not useless! If we do interpret them correctly, they will tell us a lot of information. Statistics are not the domain of losers. They are the tools that successful teams in all reaches of sport use to analyze the game from a scientific perspective, dealing with trends over time. They allow you to synthesize a large number of individual events into broader trends, trends that might go unnoticed to the human eye when it is only observing the individual events themselves. They simply are not the sort of thing which can be categorically useless. It’s a question of understanding their scope, and their limitations, and applying them thus.