Sunday, October 01, 2006

The Problem With Statistics

Right now I'm watching what is probably the week's most interesting game: Chargers at Ravens. The Chargers, at 2-0, have the league's best defense (statistically), while the Ravens, at 3-0, have the second-best. During the pre-game hype, some talking hairdo dismissed San Diego's ranking, pointing out that the Chargers have beaten two lousy teams, the Raiders and the Titans: hus, the argument went, the Chargers' status atop the league's defenses. The implication was that the Ravens' defense is much more legitimate.

But guess what? The Ravens' three opponents--Tampa, Cleveland, and Oakland--have yet to win a game among them. The combined record of the opponents in today's "epic struggle"? 0-11. So, by what standard are we judging these defenses "for real"?

Of course, the argument can easily be made that that 0-11 record has more to do with the defensive strengths of the Chargers and Ravens than the offensive ineptitude of their opponents. It's an argument that has at least some merit (though I doubt that any of those teams will emerge as an offensive powerhouse). But, at the very least, we need to reserve judgement.

(Something similar happens with rankings of run and pass defenses: the team with the best pass defense is often the team against which teams run successfully. Likewise, much of the Steelers' recent success against the run comes from two things: their relative struggles against the pass, and the leads that the offense has built early in games, which force teams to the air. But, also, you don't run on the Steelers.)

0 Comments:

Post a Comment

<< Home