You obviously missed the point. Miami has wins over then #4 Pomeroy Duke, on the road, then #7 Florida, on the road, and then #23 Illinois. Plus dropping 80 on UVA is magic for your offensive efficiency ranking. That more than counter balances the EKU game. And EKU ranks 71st on Pomeroy, so it's not like they lost to a bad teaam.
So if all of this is correct, I'm assuming that when UL beats Duke our strength of schedule goes up and UVA's goes down.
It depends on margin of victory. Since Louisville is the favorite, they could win and suffer a slight decrease in their rating. If Louisville wins be a decent margin, like 10 or more, it would give a slight nudge to UK's SOS. Our performance against Alabama and Virgina's against Boston College will have a more profound impact on their respective ratings.
Does non-hysterical, dispassionate, discussion about the details of a computer formula bother you? Nobody is freaking out or claiming bias. Just a simple question about something that didn't seem logical. Since you obviously "don't care", why open the thread and comment?
Why would he be embarrassed? Because you don't understand how the model (which is far from perfect, but overall pretty solid) works? He's even said to take early and mid season numbers with a huge grain of salt. I'm sure once UVA plays those teams and the model has more data it can better evaluate UVA. You know, just as the author states and just as any computer model works.
Considering it's the Pomeroy rankings, how efficient each team is on offense and defense will have more effect on the rankings than anything else.
