Robison: Reevaluate our ranking system (Nov. 15)
Matthew Robison | Wednesday, November 14, 2012
Preseason rankings are garbage. Based on what voters may or may not have seen from backup players in limited time the previous season, recruiting classes and, to a limited extent, returning starters, preseason rankings are an arbitrary assortment of guesses and opinions.
But the worst thing about preseason rankings is that they matter – a lot. This happens in both college football and college basketball. Right now, the three undefeated teams in college football rank differently in the Associated Press poll and the BCS rankings. But both are swayed by where those teams started in the rankings. I hate to sound like a homer, but somehow I feel Notre Dame would not sit at No. 3 in both rankings had it started the year ahead of Kansas State.
In college basketball, the situation is less severe for postseason positioning, as there are no BCS debacles at the end of the year. But when tournament selection time begins, teams are evaluated on “quality wins” and even “quality losses.” Sometimes, the quality of those games is rated by how high the opponent was ranked at the time they played. If a team plays highly ranked teams early in the year, it can artificially boost their schedule strength if those teams falter during the regular season. Likewise, teams can play some high-quality opponents who don’t get noticed until they make a run late in the season.
On the flipside of the coin, teams – like Notre Dame – can have a preseason strength of schedule that doesn’t quite hold up as the season progresses. When teams start the season ranked highly in the polls, they can inflate their opponents’ perceived strength of schedule.
All in all, the entire system is a hoax. Rankings should be based on performance, not expectations. A team ranked in the middle of the top-25 with several strong wins early in the season can still find itself ranked behind a top-five team with wins only over cupcake opponents. So those preseason shots in the dark have implications for late-season rankings.
In my humble opinion, rankings should only come out after the first week of the season. Even though it’s tough to gauge exactly how good teams are after one or two games, high profile teams often play one another early on. Some examples are No. 8 Duke and No. 3 Kentucky squaring off and No. 21 Michigan State playing No. 7 Kansas on Tuesday. In a rankings-based-on-performance model, teams are rewarded for playing strong teams early in the year rather than rolling over the meek.
The rankings would then give an accurate reflection of relative strength, rather than expectations, which is exactly what they should do.
So my nonexistent vote is cast for doing away with the garbage that is preseason rankings, letting performance speak for itself and evaluating teams on a relative basis. That is what rankings are supposed to do, after all.
Contact Matthew Robison at firstname.lastname@example.org.
The views expressed in this Sports Authority are those of the author and not necessarily those of The Observer.