Tuesday, April 14, 2009

NFL Schedules

The "strength of schedule" metric might be the worst possible way for trying to determine the actual difficulty a team will face in the next season. The Steelers earned a bye last year with the "toughest" schedule in the league- true, a number of the teams they played really were very good, but the issue is much more complex than just the winning percentage of the opponents from last year.
For example, the Steelers' schedule was considered so brutal in part because they played the Patriots, who had a winning percentage of 1.000. But the Patriots were largely toothless without Tom Brady, staggering to a 10-6 record and missing the playoffs (which, by the way, was dumb, and they should have gotten in over the Chargers). But injuries aren't the total package, either.
To a certain extent, the performance of a team in the previous season is also a function of their schedule. A team that really wasn't very good but got 10 or 11 wins due to a cupcake schedule will artificially inflate the strength of schedule of another team. This is unavoidable, at least in terms of standard statistical evaluation.
I think there is a more promising way to deal with things like this: certain statheads, the football equivalent of sabermetricians, swear by a metric called "Pythagorean wins." Baseball also uses this formula. I know exactly what the baseball formula is, I'm not sure if football tweaks it, but it basically goes like this (for baseball): first, you square the number of runs scored by a team. Then you subtract the number of runs allowed, squared. Then, you take the square root of the difference. It looks very similar to the Pythagorean theorem of geometry (A squared plus B squared equals C squared), only you subtract, you don't add. Apparently, for reasons well beyond my own comprehension, this is considered a more accurate evaluation of a team's performance than simple wins and losses. It's similar in a way to the way that people have calculated the best and worst NBA champions on the basis of their plus-minus number. I think a Lakers team was the worst, with a plus minus of about -4.5. But I digress.
In any case, don't believe the hype. Just because the Steelers have one of the easiest schedules according to winning percentages does not mean they're a shoo-in for the playoffs. I especially think it's worth remembering that the Ravens lost two very, very, very close regular season games to the Steelers (plus, obviously, the AFC Championship Game), which probably causes a faulty read on their actual quality. But they did lose their defensive coordinator, strong safety, inside linebacker, and- uh- I actually can't remember the other position and am too lazy to look it up.

But, if you take anything from this, it ought to be this: in the immortal words of that master of the spoken word, Flava Flav, "Don't believe the hype." And, yes, Chuck D also said that, but Flava Flav was funnier for my purposes, and I'm willing to slightly distort the picture for the sake of comedy.

No comments:

Post a Comment