Brady's final season at Michigan:
62.8% completions, 7.6 yards/attempt, 20 TDs, 6 INTs
I'm not sure what isn't elite about that.
Tom Brady's final season: 2600 yards, 20 TDs, 6 INTs, 7.6 ypa, 142 ratingElite by NFL standards? Sure, but by college standards? Only if you have an extremely generous definition of elite. That season is far closer to Kyle Boller than it is to Brees, Rivers, Roethlisberger, or Manning. For instance, here are some of the other elite seasons:
Philip Rivers- 4500/34/7, 9.3 ypa, 170 rating
Drew Brees- 3900/25/12, 7.2 ypa, 133 rating
Peyton Manning- 3800/36/11, 8.0 ypa, 148 rating
Roethlisberger- 4500/37/10, 9.1 ypa, 166 rating
Aaron Rodgers- 2600/24/8, 8.1 ypa, 154 rating
Tom Brady's final season was essentially identical to your average David Greene, Chris Leak, or Jason Campbell season (excluding Campbell's senior year, which was way better than anything Brady ever did). As I said, it was far closer to Kyle Boller than Philip Rivers. If that fits your definition of elite, then I agree that you'll be hard-pressed to find an elite NFL QB who was not also an elite college QB... just because it'll be hard to find *ANY* NFL QB who was not an elite college QB, because your definition of elite is so broad.
You brought up some good examples with Favre, Warner, Green, and Hasselbeck. I'd say the latter two were never among the top 5 QBs in the league at any point in their careers, but they were certainly better in the NFL than their college careers would have lead you to believe. As for Warner, I'm not sure he fits the discussion because, as you pointed out, he basically never played in college. In other words, he never had a chance to prove what he could do. So we don't know if he would've been a great college QB, a bum, or somewhere in between. It's not totally uncommon to see players like this wildly exceed expectations in the NFL. Guys like Drew Bennett, Matt Cassel, and Willie Parker did not excel in college either, most likely because their coaches never let them get on the field. In every case the player sat on the bench behind an (arguably) inferior talent. Call it "operator error." Not all college coaches are able to accurately assess the talent on their roster.
It's one thing for Matt Cassel to sit on the bench behind two Heisman trophy winners. It's another thing entirely for Kurt Warner to spend 3 years on the bench at
the University of Northern Idaho. I would submit that the two really aren't all that comparable.
Favre is the most interesting example because his career was so pedestrian in college and so prolific in the NFL. I'll fall back on the same excuse that applies to Green and Hasselbeck: I never saw them play in college, so I have no way of saying what they were like as players back then. As I said before, of all the college QBs that I have seen, none of the ones who were highly flawed in college went on to become standout pros. On the flipside, I can remember countless examples of NFL franchises trying to "coach up" physical specimens like Tarvaris Jackson, Charlie Whitehurst, and Kyle Boller. It hasn't worked.
Again, just an opinion. Nobody has to agree with me, but I think there's merit to what I'm saying. I think a QB will tell you everything you need to know about him by the end of his college career. You just need to pay attention and put the puzzle together. It's not always easy (I certainly didn't expect Freeman to become the best QB from that class), but some players are so spectacular that you almost know they'll succeed (Luck, Bradford) and others are so flawed that you almost know they'll fail (Locker).
Look, you know I like you and I love the work you do, but I really just think you're falling prey to a bundle of cognitive biases right here. Most prominent is the recency bias, but I see some selection bias at work, some confirmation bias (Tom Brady was "elite" his senior year and Kyle Boller was "obviously flawed", despite the fact that Boller put up incredibly similar yardage, TDs, INTs, YPA, and Rating totals), some hindsight bias, etc. Unless you have some compelling narrative for how flawed QBs could have succeeded in the past, but are no longer able to, then I think Occam's Razor suggests that we should assume that the current lack of flawed QBs who are excelling in the pros is most likely just a result of sample size and statistical noise.