NFL.com

I always do my mock drafts a little differently from the traditional method, in that my team-prospect pairings are based on optimizing for wins in the upcoming season -- in a structured way. Here's a quick look at my process: I take the best available information about free agency, which right now is admittedly not much more than what we know about player contracts, to predict players' market value. I then check free agents' projected market value against the anticipated salary-cap space for each team. Finally, I add potential free agents or drafted players to different teams and identify which individuals increase teams' projected win total the most. (My player model leverages historical pre-draft player data to forecast NFL performance.) I'm starting this process early again this year, which is cool because I am excited to track how things change with free agency and as the draft process evolves. A little housekeeping: I used computer vision-derived measurements and stats in my player evaluations, and all of my speed, burst and advance tracking notes are based on on-field game speed metrics.