IvanKaramazov
Footballguy
Yeah, I don't really blame the analytics guys for doubting Allen. The problem is that should have recognized that their model was never designed to evaluate a player like this. They're used to evaluating guys who played in major programs, with solid teammates around them, against solid opponents, and with the benefit of solid coaching. Allen never had any of those things. For example, if you're looking at a QB coming out of Ohio State or LSU or UCLA or someplace like, and you spot some mechanical problems, you figure the school's coaching staff have seen the same problems that you're seeing and have done everything they can to fix them. The fact that those problems are still there is a little worrisome. You can't really say the same of a guy coming out of Wyoming.FWIW, I think Allen is definitely a major outlier so it’s not like people didn’t have reason to doubt him.
In other words, you're taking a model that does a pretty good job with players from big time programs and applying it to a guy from a third tier program but with out-of-this-world raw athleticism. That sort of player doesn't come around very often, and people who pride themselves on their mathematical acumen ought to have realized that the normal metrics don't automatically carry over. This is a well-known problem with forecasting in general -- it's always dangerous to make forecasts outside the realm of the data you used to create your model in the first place.
This is just anecdotal with no actual numbers to back it up, but I think this is wildly exaggerated. Every QB in every game throws interceptable passes, gets sacked, etc. People grossly overreact to those plays when Allen makes them because that's the narrative. I'd be surprised if it turned out that he was making "he'd like to have that play back" plays any more than other top-flight QBs.The boneheaded plays? Well, still too many of those,