on February 11, 2019, 14:09:35, in reply to "Just deleted my response here because I think I figured it out."
i find the experience thing a little suspect. Experience can be a function of not recruiting well, big losses to attrition, big losses to injury, etc. All it means is you played players early. that may help some, but it is probably pretty marginal, it probably only applies in one snapshot season when all that experience lines up as a group of seniors, and i'm sure it doesn't correlate to outcomes as well as recruiting rankings. 
Previous Message
I was totally confused by this subthread but I think I get it now. You see my post as a defense of Illinois - "don't look at the Illinois ranking here because Connelly's numbers are bad bad bad!".
That's not what I'm saying. I mean, I guess it is what I'm saying, indirectly, since we'll hopefully Syracuse this year, but that's not why I posted this. Even Connelly says that these numbers aren't to be taken the same way as his in-season numbers. In-season he's trying to predict the results of the next few weeks; offseason he's simply trying to come up with a good way to predict a full season in advance.
And in my view, his "returning production" numbers still need tweaking. There should be a better way to predict a Louisville or a Syracuse (but I get that making those adjustments probably takes an Oklahoma from 4th to 33rd and that wouldn't make sense for the true purpose of doing this which is predicting the top-25 to the best of his math's ability).
53
