Michigan State, Texas, and that Dumb Old Simple Rating System

The thing about the simple rating system is that it is a predictive measure. Nothing’s perfect, by the way, in that regard, and if you find it, shoot me an email and we’ll split the profits. In college, there are fewer games, less interconnectivity between conferences, and the rosters churn quite a bit each season. As a result, one blowout, and some skewed non-conference results that may not represent the average performance, can make it very approximate.

Still, what of the teams like Texas and Michigan State, where we have the underachiever and the team that just knows how to win. They are ranked very similarly in the AP Poll this year, Michigan State at #13 just ahead of Texas #15. What if we looked at teams that fit each profile?

I looked at all teams since 1998 where the ranking in the final AP Poll and in the final SRS rankings were disparate. Most, as it turns out, are pretty close and the two rankings match up. I found all teams that were ranked at least 5 spots higher in one of the two systems, and had at least 3 points difference between where they ranked in SRS and where the team with the same AP ranking finished in SRS. (I did this to measure magnitude, the difference between being #16 and #25 may not be as large as #2 and #8).

Then, I looked at the next year for all teams in those groups who were ranked in the next year’s preseason Top 25 poll, to compare how they finished a year later. Here are the SRS Stars (like Texas) versus the Ranking Stars (like Michigan State).

By way of explanation, “similar” means the team finished within 3 spots of their preseason rank in the AP poll, slightly better/worse means within 4 to 8 spots, way better or worse were those teams finishing 9 to 15 spots different from their ranking, and Booms/Bombs were the ones who were at least 16 spots different. For example, a team that was preseason #4 but finished #22 would be a Bomb.

The SRS stars prevailed. More likely to have a way better ranking, with 3 versus 0. Far less likely to bomb, with 3 vs 8. It’s not perfect by any means, but it does mean something. Michigan State last year was one of the “slightly betters”, so maybe they can do it again. I’m skeptical, of course, that the 8-1 record in close games over the last two years will continue.

[photo via US Presswire]