| The stat monkeys vs the selection committees [message #1015687] |
Wed, 10 May 2006 14:10 |
|
The battle lines are forming...
"for predicting the outcome of tournament games...our [logistic
regression/Markov chain] method has beens significantly more successful
than...tournament seedings"
Kvam and Sokol
http://www2.isye.gatech.edu/~jsokol/ncaa.pdf
"The individuals whose quantitative ratings are used in the computer
ranking have never been given a clear objective to design their ratings
for and they are limited in the inputs that they can use (for example,
they can not use score or site of game). For all of these reasons, I am
advocating a boycott of the Bowl Championship Series by all
quantitative analysts."
H S. Stern (2006) "In Favor of A Quantitative Boycott of the Bowl
Championship Series", Journal of Quantitative Analysis in Sports: Vol.
2: No. 1, Article 4.
http://www.bepress.com/jqas/vol2/iss1/4/
|
|
|