[bdeg]

I'm working my way through this...

"The R^2 was just 0.35, which isn’t very high, but I’m not sure if you can come up with a formula to make it any higher. "

That's terrible.

"Whether or not this formula is perfect isn’t that important — for what we’re looking for, with hundreds of QBs, something that’s generally correct is all we need. We don’t need to know specifically if Troy Aikman was better than Jeff Garcia, but just that both were better than Danny Kanell and Bobby Hoying."

Guy sets the bar pretty high, huh?
If he wanted it to make sense, why is he simply devaluing each QB's play next year by 5% continually? He should run a regression and find out exactly what impact every age has on a QB's change in efficiency.--I didn't read far enough, it sounds like he did this? not clear
I haven't gotten to the methods yet, but I think this thing is flawed from the start. I think you HAVE to find a way to account for the offense a guy has around him. see Philip Rivers

That whole part about draft value is just dumb. It doesn't make sense at all and he just made it up and stuck it in there. At least he admits it, though. Shockingly, according to their final equation they found draft value to have the MOST impact. I think that's just because the rest of it is so flawed. Since their formulas don't line up, obviously NFL teams will do a decent job of guessing how players will do.