Originally Posted by Pestilence
Yeah....the way you put it sounded better.
Well, I noticed you mentioned Carr and Leinart and I started looking at it deeper.
You have to have an "anchor point", some verifiable data that you can use as your baseline. That's the bottom group, where we have a preponderance of verifiable busts.
You don't even really have to discount Carr and Leinart at that point because you're not trying to figure out why they busted. You're looking at it as a somewhat linear percentage chance to bust, proportionally related to their score.