The problem with the Harvard chart is that it relies entirely on Pro Football Reference's Career AV stat, which has its own problems when it comes to valuing players (imo).
How to Value NFL Draft Picks | The Harvard College Sports Analysis Collective
The gist of the problem, to me, with the methodology of the Harvard chart is that it makes no adjustment for minimal level Career AVs. There should be some baseline Career AV subtracted from ALL players to make the analysis more meaningful in terms of gauging the picks. Using a Career AV of 0 as the floor distorts the analysis.
Look at the example of Josh McCown that the author uses to explain his methodolgy. McCown has a CAV of 15, Rodney Harrison in his example has a CAV of 74. According to his analysis Harrison is 5 times more valuable than McCown. In the real world of trying to win championships in the NFL, I would say Harrison is closer to 50 times more valuable.
His model overvalues back-up level CAV's (very minor contributors) because his floor is 0 CAV (a guy who never sees the field). This small starting assumption, to me, undermines his results. Yes, McCown is better than a guy who sees the field a few times and finishes with a CAV of 3, but not, imo, 5 times better. To me, both of those players fall into a much more closely comparable category of "non-factors."
Obviously, every metric is subjective in some way. And each annual draft field varies, making some charts more applicable than others depending on the year.
But count me in the camp that thinks the Pats did pretty well on the trade, especially considering the other first round trades and this year's glut of 2nd-tier (but good) talent. But I wouldn't say they fleeced or robbed anyone. As for a true judgment of the trade we will need to wait and see who you guys take with the picks and how those players and Patterson perform.
