Statistical Odds to Win it All
Pomeroy has used his Log5 formula on this Pythag numbers to determine the odds that each team has at winning it all. His ranking isn't too kind to us in the first place, ranking us better than only St. Mary's and Cornell of the teams left in the tournament. Accordingly, he gives us a 1.2 percent chance of winning it all. He explains that Ohio State's chances at winning it all look pretty good because they merely have to navigate Tennessee and UNI/MSU in the Midwest. FWIW, before Wisconsin lost yesterday, they had been his #3 ranked team for several weeks (which I think was wacko). Jeff Sagarin has constructed a similar chart using a computer model that simluates the Sweet 16 100,000,000 times. He gives Tennesse about 30-to-1 odds at winning it all. Which is about the same as Xavier, and lands us as the 10th most likely team. Clearly they believe we are a big underdog to beat Ohio State. But at this point of the year, I think a lot of those statistical models can be thrown out the window, and what's more important is how well a team is playing right now, and, more importantly, how a team matches up with their opponent. Vegas has opened us up as a 4.5 point underdog to OSU. Both Accuscore and Pomeroy are giving us about a 28% chance to beat Ohio State. That seems low to me. Hopefully Tennessee will prove them wrong.
(I think these things are pretty much pointless, but they are interesting to look it)
about 2 years ago
golfballs03
2 comments
0 recs |
Comments
I see the UT/OSU game as about 50/50
I could see pegging Ohio State with a 60% chance of winning, based on their talent, but 70-75% seems awfully high. I hope we prove them wrong.
by Incipient_Senescence on Mar 22, 2010 1:05 PM EDT reply actions
Such statistical models are more useful...
…in a situation like the NBA playoffs where the teams are playing a series of games.
70 or so possessions is not a lot.
...just apologize for not thanking me.























