Sunday, March 13, 2011

Bracket Lesson #2: "Clutch" vs "Overrated"

I'm going to copy & paste what I said when writing this post last year:

"I'm always fascinated when I hear the talking heads on ESPN complaining about how the Selection Committee is supposed to choose the 34 best teams for at-large bids. They're not, and they tell us that every year. They're selecting the 34 best resumes. There's a difference. And this is an opportunity for people filling out their brackets, knowing that there are steals to be had because some teams are seeded better than they would be if we were rating teams by how good they were."

What I said there is that there is a difference between the teams with the best resumes and the best teams. The Selection Committee doesn't take into account the scores of games. The fact is that beating a team by 30 points is a lot more impressive than beating a team by 1. More importantly, there really isn't a difference between beating a team by 1 and losing to the same team by 1. There is so much luck involved in a game that a close win or loss has to do mostly with luck. We hear from television talking heads about "clutch" play, but that's mostly bogus. The difference between a clutch and un-clutch player isn't that big. This has been borne out again and again (particularly in some great posts on Basketball Prospectus analyzing the over-obsession with the media about the close losses by the Miami Heat this year) - there simply is no correlation between records in close games and future performance. If a team is 10-1 in games decided by 5 points or less, it's not because they shoot 91% from the field in clutch moments.

Obviously the Selection Committee should consider whether you win or lose your games by 1 point. Sports are about who wins, after all. But when you're filling out your bracket you want to pick teams that are better than their seed and avoid teams worse than their seed. And the best way to figure that out, if you haven't watched a basketball game all year, is to look at the Sagarin ratings.

The Sagarin ELO_CHESS is the best measure of resume quality. The Sagarin PREDICTOR is arguably the best measure of team quality (many would argue that the Pomeroy ratings are actually better, but the PREDICTOR and Pomeroy ratings are always nearly identical). If a team has a PREDICTOR much better than their ELO_CHESS then they are better than their seed, and vice versa.

Below is a list of teams that are "overrated" (far superior ELO_CHESS to PREDICTOR) and "underrated" (far superior PREDICTOR to ELO_CHESS). I'm leaving out the 14, 15 and 16 seeds. Historically, you should pay closest attention to the "overrated" list. Those teams almost always under-perform their seed in the NCAA Tournament. Remember, this list is purely what one can conclude from the Sagarin ratings. I have my own opinions on which teams are overrated and underrated, and you'll need to tune in later tonight and tomorrow for those posts:

"Underrated" teams:

"Overrated" teams:
Notre Dame
St. John's
Texas A&M
Old Dominion
Penn State

No comments: