I understand why people don't like this because it's easy to think of times when recruiting rankings over or underrated a player. But in general the rankings give you an idea of who is on the roster. A team loaded with 5 star talent, all else being equal, will beat the team loaded with 2 star talent. It's not everything, but it probably gives you some information. Not football, but our expectations for BYU basketball performance next year are higher with AJ Dybantsa signed than had he not signed.

It seems that for some people, using a predictive model based on recruit rankings and prior year performance violates some sense of fairness. I can't argue with people's values. If you think it is just inherently unfair to include recruiting rankings regardless of whether it predicts anything then that's your opinion and I respect it.

However, if your argument is that these rankings don't actually predict anything then I disagree. I'm reasonably sure this ESPN team is not adding variables to their model that have no predictive power. And I'm confident that they aren't adding recruiting rankings to the model in an attempt to pump up the SEC or reduce the number of Big12 teams in the playoff.