Wisdom of the Mock Draft Crowds: Group Analysis Outperforms Individual Expertise

My method was a little different, but as it turns out even closer to the real results. Healy used an average of all the results. I used the mock drafts of others as a sort of draft board, and then actually simulated drafts using those rankings (as a result, things like positive outliers tend to drive draft position, such as Blake Bortles going higher than the average ranking). Using the scoring system employed, my second level mock draft scored a 149, as almost half (15) of the first round picks were selected within 2 spots of where they were ranked in my exercise. In contrast, that score, which represents the distance from projection to real results for all picks, ranged from 162 to 230 for all individual mock drafts reviewed.

There are reasons to think that the wisdom of crowds might not apply to the mock draft industry, namely that it is not entirely independent and diverse. While early mocks may be independent, there is no doubt that knowledge becomes centralized, and norms begin to get established. Still, there is enough diversity. In the 12 I looked at, there were 50 different players predicted for the first round, so while the group may have missed on some, they got most of them identified.

My main fear of mock drafts, with individual competition to be accurate, is a fear of being different. However, I do think that in those spots where there was true universal consensus (i.e., Jadeveon Clowney as first pick, Greg Robinson 2nd) it was not misguided. Maybe mocks are not so misguided after all. Rather than competing, everyone should do their own, and then in the end, the “wisdom of the crowds” does a pretty good job of identifying real front office behavior.