Forty-yard dash, Wonderlic Test, Heisman Trophy — these are mere trickles of the wave of information that teams use to judge prospects for the NFL draft. Franchises spend all year evaluating players in hopes of landing the next Cam Newton or Adrian Peterson.
But Wharton professor Cade Massey of the Operations and Information Management Department will tell you that for all this work, NFL teams are just as likely to draft a bust as they are to stumble upon a gem.
A study he co-authored with Richard Thaler of the University of Chicago Booth School of Business found that teams overestimate their skill in drafting compared to other teams. In fact, the study found that differences between teams’ draft outcomes are almost completely chance-driven.
“I’m not saying these guys are picking out of a hat,” Massey said. “Because they clearly know what they’re doing. Top picks do better than lower picks, so there is skill involved. It’s just there are no observable differences in skill across teams; they are kind of all equally skilled. And that means that differences in outcomes are 95-plus percent chance.”
Massey has researched aspects of the NFL draft ranging from teams’ overconfidence in their abilities to the true value of draft picks.
And although Massey’s work deals with uncertainties and probability, NFL teams could learn a few concrete lessons about player value from his research. Massey equates value with quantitative measures like starts and pro bowls.
One of these lessons is the idea that the value of first-round picks actually increases as the first round progresses, mostly due to the high salary of top picks. According to this model, the first overall pick is actually the least valuable pick in the first round.
This leads to one of Massey’s key conclusions: Teams should look to trade down in the draft order. Massey finds that teams that trade down to acquire more picks actually increase the overall value of their drafts.
“The No. 1 prescription that comes out of this research is that teams should trade down,” he said. “The most sophisticated teams, it’s not that they’re never trading or that they’re never placing a bet on a high pick, it’s that they do that infrequently. Mostly they’re going to trade down, mostly they’re going to stockpile these picks, then every now and then they’re going to say, ‘We’ve got to have that guy.’ And they’ll use some of those extra picks they’ve stockpiled in order to go out and get that guy.”
There is a method to Massey’s madness. Look at the 49ers or Patriots, for example, who often trade back and stock up on picks, and then compare them to the Raiders, who rarely surrender a top-10 draft choice. The relative success of the draft picks on these franchises says it all.
Massey’s work is part of a larger movement in sports towards big data and analytics. Yet unlike the Oakland A’s from the movie “Moneyball,” Massey noted that some NFL teams are eschewing new analytical methods.
Given widespread accusations about South Carolina defensive star Jadeveon Clowney’s lack of motivation and commitment, one would think that NFL teams have developed ways of using cutting-edge technology to evaluate players based on character.
“We’re not there yet,” Massey said. “Everyone kind of has the same data now on how fast he is, how far he can jump. This is kind of the frontier in terms of where player evaluation is.”
Please note All comments are eligible for publication in The Daily Pennsylvanian.