Before I get to today’s games, let me cover some questions about the Massey-Peabody tracking. I’ll be referencing this sheet.
Rufus asked me which numbers I was using:
If he were nowhere near a computer, that question would make sense. If he had access to one, it would have been easy enough to do the math and verify that yes, I am using the posted win probabilities. As I’ve said before, I cannot assume any else’s regressions. However, that isn’t my main concern. This is:
So… let’s look at the Alamo and Arizona bowls:
What we have here is a mean, which is being average with a median. That’s… problematic. I am straight up never going to do that. The margin between -1 and 0 is rather tiny, but for Massey-Peabody regression purposes it appears that it would be treated the same as a difference between -2.5 and -3.5.
Finally, there was one other question:
Absolutely. FWIW, no one ever does this, but if someone posted their opinion as a formula, such as (internal moneyline X, Pinnacle/CRIS moneyline Y, z%X, q%Y) I’d be happy to apply that. Now to today’s games:
Massey-Peabody is currently down $2,704.
|North Carolina St.||0||13583||32600||0|
ESPN’s FPI is currently down $991:
|North Carolina St.||0||22933||55040||0|
As for Ed Feng…
|% of Bankroll||-84.52%|