Player Empirical Comparison and Optimization Test Algorithm (PECOTA)

Definition

Also a "backronym" for former Major Leaguer Bill Pecota, PECOTA is Baseball Prospectus' system for projecting player performance. The acronym stands for "Player Empirical Comparison and Optimization Test Algorithm." It was developed by Nate Silver in 2003, and Silver ran the projections - which were owned by Baseball Prospectus -- from '03-09, before ceding full responsibility to Baseball Prospectus itself.

PECOTA is perhaps best known for its use of comparable players. In predicting future performance, PECOTA uses historical comparisons of players with similar career paths.

PECOTA is used primarily for projecting the performance of individual players, but it can be applied to entire teams as well to project records. It uses predicted depth charts to establish the projected runs for and against each team, before gleaning a projected win total from that.

Obviously, no one is claiming that every prediction will come true, but PECOTA is widely regarded as one of the most accurate predictors in the industry.