I think a lot of the problem stems from the fact that these metrics are scrutinized all season long when they are only really designed to give an accurate result after the season is complete. Early results with a higher level of variance end up skewing the expectations and thus the final selections. When the talking heads get their hands on these numbers half way through the season and decree that x conference will only get y bids, it seems to be a bit of a self fulfilling prophecy. I don't know what the solution is, but the current selection process leaves a lot to be desired.