Let me start by saying that Rabberousr did excellent work with his presentation of the Massey-Thaler information and it was one of the best things I've read this off-season. I am largely in agreement with him regarding the correct philosophy regarding the draft. I believe that you want to maximize premium picks and agree that the Cowboys have, apart from last year's trade, generally failed to do so since Jimmy Johnson departed. While the Jason Garrett era actually shows a slight upturn in the Cowboys respect for their premium picks, they are still 1 down from where they should be and that is not a good thing.
However, I have an issue with the discussion as a whole. Now, I may be off-- I'm beginning to get the feeling that I'm missing something. I feel like people are unwilling to engage me on the Massey-Thaler stuff or read what I'm saying about it. Perhaps I am being inane. Perhaps I'm missing something obvious. I am certainly ignorant as I have not yet read their work (though, with school out, it might make a nice summer project).
I think, however, that I get the gist of it, and that Ginge159's summary on rabble's front page article did, in fact, encapsulate what they are saying really well:
They are all able to predict how players will pan out to some degree (hence, 1st round picks are better than 7th rounders on average), but no team has been significantly better at it than the others.
I’m struggling for an analogy, but here’s one: it’s like a bunch of kids who keep getting 70% on a multiple choice maths paper where there are 4 options. It’s not pure luck, or they’d all get 25% each time. Instead, they are all equally good at the subject, and any variation is due to chance.
I haven't been able to get anyone to confirm that this is the thrust of the study, but Ginge's comment has a solid 8 recs (one of which is mine) and approval from Dr. P, Foyesboys, and (less directly, but I believe) rabblerousr himself-- in short, people on both sides of our little divide. I bring this up because, after thinking about it, I'm pretty sure this is false. Not a false representation of the study, but false in fact.
No matter how many picks they studied over the years, I'm pretty sure that many GM/Coaching regimes do not last long enough to provide a really good sample size. Now, I'm sure the variation in what teams do is small, but so is the variation between a .250 hitter and a .300 hitter, one of whom is a JAG at best and the other is an all star. Similarly the variation in abilities between athletes is strikingly small -- how worked up were we about Aaron Donald's height and arm length? Donald's height is first percentile, which means 99% of the people at the combine who play his position are taller than him. So he's a midget, right? Ra'shede Hageman, possibly too tall for DT at the other end of the chart, is 6'6" to Donald's 6'1". Aaron Donald is 94% of Hageman's height, yet we treat him as if he's miniscule and Hageman maybe too gigantic.
Why do I bring this up? because in areas of tight competition, small differences are magnified. and it's nonsense to say there aren't noticeable differences in teams' ability to evaluate talent. There's a glaring difference between the regimes of the Jerry Jones era of the Cowboys. If you actually look at the success of their drafts, the Johnson, Gailey (believe it or not), Parcells, and Garrett regimes are clearly better NFL talent evaluators than the Switzer, Campo, and Phillips groups were. Given that the scouting departments were largely similar across many of those coaching changes, it may be that some coaches ignored their own department more than was good for them. But I've seen way too much difference in the drafting ability of my own team to simply believe that everyone, as Ginge159 summarized, gets a 70.
Now, I believe Massey-Thaler's point was that the variations in evaluation are smaller than the variations in general value for the pick, and so far so good. But the quantum leap people are making from that to "don't EVER trade up" seems to be ignoring a couple of things.
1) as I stated, there are *real* differences in talent evaluation.
2) Even if there aren't real differences in evaluation, there are differences in needs.
Now, everyone always says "you don't draft for need" and then these very same people usually complain that we took a TE in the second round while there were positions of need still on the board. The fact is that the draft, while generally a long term game, is about improving *your* team, not someone else's. If there's a player on your board that has a 50% shot of repairing a glaring, hard to fill, hole on your team then that 50% shot might well be worth giving up 40% and 30% shots at less important or less urgent positions.
Clearly this is a losing strategy long term, but sometimes you need to take a risk and make a leap. There's a concept in flight called "getting behind the power curve". The basic idea is that an engine takes time to respond and if you throttle back too far or wait too long to bring the power back up, you will put the plane in a position where it simply does not have enough power to recover from slowing into a stall and crashing. I believe the draft is analogous to this.
If you draft, or otherwise manage your team poorly enough, you put yourself in situations where you need to take risky shots like the Lawrence trade to field a competitive team because you just cut your 1st ballot HoF pass rusher. The problem is that you run a very real risk of the pick not panning out, which will leave you with even more holes down the road because you spent (in this case) two picks on him. Joey Galloway and Roy Williams are fine examples of players that left the Cowboys premium pick poor. Each of those situations put the team behind in drafting and put them behind a power curve, creating needs that had to be addressed as other positions of less need at the time of the trade aged and failed.
The one thing that gives hope about the Lawrence pick is the difference in attitude towards it. There is no hint of the overconfidence that M-T describe. The Mo Claiborne pick provides a wonderful contrast here. No one came away from the Lawrence pick saying he was the best prospect since Ware or that he was the best player in the draft. They simply, flatly said, "this was a need pick." They made no pretense of justifying it. They admitted to overpaying. they simply said that the things Lawrence could bring to the team were worth the risk because the entire defense is based on pressuring the QB and they didn't have a pressure player. There's a hint of similarity there-- Ryan said he needed two press corners to make his defense work-- so I understand criticism that this is more of the same. I can only point to the much humbler attitude this time and say "I think they get it, they're just trying to claw their way out from behind the power curve."