Here's an email that I got just last Saturday from a buddy of mine, who is a subscriber to Kenpom's site:
"Actually it's not just math. Pomeroy looks at the numbers and makes adjustments for any outliers to his model.
In Kentucky's case, their defensive efficiency has been so much better than any other defense in the Kenpom era, he's had to use a huge adjustment number to make them fit the model.
You have to subscribe to see the raw numbers. Kentucky's raw defensive number is some where in the mid 70's. Pomeroy has it adjusted to 82.5 or somewhere around there. The previous lowest raw defensive efficiency was around 84. That 7 or 7.5 swing in the raw to the adjusted efficiency is twice as large as any adjustment he's ever made.
Basically Kentucky has proven to be so much better than the next best defensive team historically according to Kenpom model that it's made him nervous that something isn't right with his model. He doesn't believe that 75 is real.
The Sagarin model uses strictly math, and it has Kentucky as at least a 3 point favorite over every other team in the country on a neutral court and only Virginia as a favorite in a game played on their own court."
This post was edited on 1/24 5:16 PM by J.C.D.M.