One major caveat: teams are static throughout the entire tournament in this analysis. Fairly obviously, a 10 seed making a run to the Final Four is probably going to be less of a dog in the Final Four than they'd be against that same team right now. A 1 seed in the FF is more likely to have simply met expectations, and won't see a similar market boost. So, favorites' odds are probably somewhat overstated, underdogs' odds somewhat understated. North Carolina is very overstated if Lawson is seriously hurt. This presumes healthy Lawson.
Tuesday, March 17, 2009
March Madness Bracket Odds
March madness bracket odds are over on the right finally, and it's my best estimation of a true market bracket. All the first round games' win percentages should roughly match up with Pinny as of a couple hours ago. Again, mostly I just tweaked high level efficiency stats to get as close as possible to deriving the 1st round line, and past two games' lines. Actual game lines could easily be off by a point or two in the next round, there's really no way to hit it completely on the head with a high level formula. But, it hopefully tempers some of the more major over/under expectations of KenPom vs. market. Using spread -> ML conversions also avoids the log5 underrating big faves problem.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment