Sure, we have to take into account the lesser talent level that Keenum faced in college.
That's why when I evaluated him, I did these things:
1. I evaluated games he played against better defenses, like Alabama, UCLA, S. Miss, Penn St., Miss St., Oregon, TCU, E. Carolina (the years they had good defense under Skip Holtz, who had been with major teams like ND, Fla St, S. Car., and won multi C-USA championships.)
2. I evaluated games he played against common opponents in the same year with drafted QBs like Luck, RG III, Weeden, Ponder, etc.)
3. I evaluated his O-line play; for example, against UCLA in 2011, he was under pressure some 24-25 times as compared to 4 by Luck.
UH played UCLA early in the year, while Stanford played UCLA late, when the Bruins suffered some injuries in the secondary.
On the details, I check to see:
1. How fast does the ball leaves his hand on a 10-yd route, an intermediate route between 15-22 yards, a deep route of 30-40 yards as compared to other QBs from the same formation, running the same or a similar play.
(In his Fr year, the Cougars did run a small but significant number of plays out of the traditional I formation with 2 backs and a TE. They did run a few bootleg out of the PA just like the Texans have been doing.)
In the spread, Keenum got the ball out a hair faster than RG III, and much quicker than Weeden, for example.
2. How good is his presnap read. Can he reads defense and find the open receiver against that particular defense.
Does he recognize a potential blitzer and counter the D-call, making the aggressors pay for their aggressiveness.
3. On the post-snap read, can he recognize the disguises of the defense so as not to make stupid mistakes like Schaub has been doing on a consistent basis.
4. Does he understand his own protection scheme; for example, when the QB has max-protect (let say 7 blockers against 4), he can afford to stay in the pocket longer, and not having happy feet, or panic too soon.
When he has one-on-one protection, it means that his five weapons would also have single coverage (plus a single safety deep), the QB wants to find the mis-match .
There are a whole lot more nuances than just those things.
I took Bill Walsh's manual for things that he wants when evaluating a QB, and apply it, basically.
I rated Keenum more highly than Kolb or Ware (at the same moment in time before their respective draft period.)
So, there you have it; a few things in my thought process when I try to evaluate a QB for the draft.