Let me explain the numbers:
Pythag is how good you are. A team with a Pythag of .750 would expect to beat an average team on a neutral field 75% of the time. Unlike Percentile and Sosmov, Pythag is designed to be predictive.
Percentile is how well you’ve done against your schedule relative to a generic group of teams with a Pythag of .900. You might ask why FSU isn’t in the 100th percentile. It is because a team with a Pythag of .900 would go undefeated against FSU’s schedule 5.41% of the time. I am putting half of those teams above FSU and half below. The same is true for the other 5 teams, all at X-1. For the record, Percentile is operating under the assumption that all six teams win their remaining games.
Sosmov is how well you’ve done against your schedule considering MoV and SoS. Clever, right? Sosmov is descriptive in terms of how well you’ve done against your schedule, and does a good job of explaining why the committee seems to be socking FSU in the shorts.
Percentile only cares about wins and losses. Pythag cares about every play of every game (although it is not taking tOSU’s quarterback issues into account. Sorry.) Sosmov is strictly a formula, although it uses the Pythag’s of the opposition to create the SoS.
So, what to make of these numbers? Alabama is clearly #1. Oregon’s claim on #2 is reasonably solid. Baylor’s claims that they belong in the top 4 are much weaker than I was expecting. FSU is garbage, save for the fact that they are undefeated (and perhaps slightly better than Baylor). What I find interesting is the cases of Ohio St. and TCU. Now, I’ll admit tOSU is hurt by the fact that this guy is their starting QB:
However, if Ohio St. smashes Wisconsin, I’d expect them to end up ahead of TCU in my numbers. More likely, they’ll lose and leave a debate between TCU and Baylor. TCU is a bit better than Baylor, and has been more dominant against their schedule. As a general rule, I could care less about H2H and would give TCU the nod.
Enough preamble, here are the numbers: