|
A Log5 update
Prev Elite8 Final4 Final Champ
1S Duke 24.3 79.0 61.8 43.7 31.9
1E Kentucky 4.6 80.4 49.5 22.7 13.4
1W Syracuse 6.1 66.9 39.4 24.7 11.3
2MW Ohio St. 5.3 70.6 49.0 26.1 11.2
2W Kansas St. 3.4 62.3 32.0 18.8 7.9
2E West Virginia 3.8 66.4 33.3 13.4 7.1
3S Baylor 2.7 68.1 21.8 10.1 4.9
6W Xavier 0.6 37.7 14.9 7.0 2.2
5W Butler 0.5 33.1 13.8 6.4 1.9
4S Purdue 0.4 21.0 10.4 4.2 1.8
5MW Michigan St. 0.5 51.0 18.9 6.5 1.7
9MW Northern Iowa 0.1 49.0 17.7 5.9 1.5
6MW Tennessee 0.2 29.4 14.3 4.7 1.2
11E Washington 0.4 33.6 11.7 3.1 1.1
10S St. Mary's 0.1 31.9 6.0 1.8 0.6
12E Cornell 0.006 19.6 5.6 1.1 0.3
This is a log5 table. It's explained here. The "Prev" column indicates a team's chance of winning the title as estimated before the tournament began.
Now that the log5 probabilities have been recalculated after the first two rounds, the results correspond remarkably well with the brackets' seeding lines. The 33 teams that were eliminated owned about 47 percent of "championship share" at the outset of the tournament. That’s been redistributed to the 16 survivors, with Duke and Kentucky gaining the most. (Though actually in relative terms Cornell gained the most, seeing their chance of a title improve 50-fold from a week ago.)
Outside of the lingering issue with Duke, there don’t appear to be any obvious problems with the underlying ratings that are used in the calculations. Based on Purdue’s tournament performance, their rating seems reasonably close to being an accurate Hummel-less approximation. Michigan State’s overall strength undoubtedly takes a hit with the loss of Kalin Lucas, but it’s possible that the Spartans were underrated at full strength. They were so inconsistent over the past month that it’s hard to know what to believe.
Taking the national pulse as we enter week two, it appears Kentucky will be the public’s clear favorite, with Syracuse a not-so-distant second. The latter assumes that Arinze Onuaku is able to contribute in the round of 16. According to our computations, the chance of a Syracuse-Kentucky championship game is merely 5.6 percent. Of course, Duke has a significantly higher rating than either. However, even if Coach K is seen forking over wads of cash to Brian Zoubek today and the Blue Devils are subsequently removed from the tournament, the chance of a 'Cuse-UK match-up only increases to eight percent. Basically, it’s more unlikely than most think right now.
Keep in mind that while both the Wildcats and Orange completely overwhelmed their second-round opponents, Wake Forest and Gonzaga were among the weakest at-larges in the field. Even though Duke’s win over Cal was not as dominant, it mirrored earlier outcomes when the Bears faced Kansas and Syracuse. At this point, it does seem a little crazy to assert that Duke doesn’t have a realistic chance at winning the title, even with the recent history of tournament disappointment.
In addition, don’t rule out Ohio State from the championship equation. Regional balance took a hit the past four days. The Midwest went from being hailed as the strongest region to now being the weakest. Ohio State may not have looked as good as Kentucky, Syracuse or Duke over the past two games, but they need only negotiate games against Tennessee and either Northern Iowa or a Lucas-less Michigan State to make it to Indianapolis. This explains why the Buckeyes’ chance of getting there is at least comparable to each of the remaining one-seeds, each of whom have a much more difficult obstacle potentially lurking in the regional finals should they advance that far.
This isn’t to ignore the possibilities that exist for Kansas State, West Virginia, and Baylor at this point. They are clearly closer to the long shot category, but not so much so that their fans shouldn’t be investigating air fares to Indy.
There are a few other scenarios you may be interested in. The chance of none of the one-seeds making it to the Final Four is about 12 percent, which is identical to the chance of all three of the one-seeds making it there. The chance of a team with a nine-seed or greater getting to Indy is a surprisingly manageable 36 percent. The point here is that there are plenty of opportunities for more wild things to happen in the upcoming week.
Ken Pomeroy is an author of Basketball Prospectus.
You can contact Ken by clicking here or click here to see Ken's other articles.
| 
0 comments have been left for this article. 
Bubbles of their Own Making: Villanova, Temp...