Mock 2018 NFL Draft
Published:
By Max Rotblut, Shekhar Shah, Sam Negus, Josh Leffel
The NFL Draft is one of the most important events during the football season. Teams have a chance to infuse their roster with college football’s most talented players. Draft day strategy has become a large part of front office operations, and draft picks are an extremely valuable resource not just for selecting college players, but using as leverage in trades between teams as well. The NFL Draft’s popularity made its way to the media, where insiders, experts and fans all create mock drafts, where they input picks for each based on what they think might happen in the draft.
Over the past few years, teams have started utilizing these mock drafts as a “wisdom of the crowd” strategy, where they are aggregating hundreds of mock drafts and rankings to come up with a quantifiable consensus order and desirability of each player in these mock drafts. Using expert mock drafts in tandem with consensus rankings of top prospects, we applied a statistical approach that evaluates how often one player is ranked ahead of another across all mocks and uses this evaluation to assign a value to each player. This value accounts for both talent (rank on Consensus Big Board) and whether or not the player is a QB. Using these results, we simulated 100,000 versions of the first round of the draft where players were chosen randomly based on these values and their availability. This approach allowed us to not only calculate average draft positions, but also the likelihood that a player would be available at any given pick.
This project demonstrates how data-driven models can enhance draft preparation by quantifying uncertainty and simulating possible outcomes. Rather than replacing scouting or expert judgment, this approach provides a structured, objective way to evaluate scenarios and make more informed draft-day decisions.