SABR: Projecting MLB WARP for Prospects

Student Author(s)

Nathan Longfield

Faculty Mentor(s)

Professor Jill VanderStoep

Document Type


Event Date



One goal of Sabermetrics is the development of projection systems to predict future player success. Many advanced models have been developed, such as PECOTA from Baseball Prospectus and ZiPS from the Baseball Think Factory. Since teams want to invest, both monetarily and developmentally, into players who will return value, the ability to project how the players will perform is essential. Once players are in the majors for a few seasons they have established what their production level will most likely be. However, when a player is in the minors it is not as clear how they will perform due to differences in levels and leagues. Since prospects and players in their first years in the professionals demand lower pay and are under team control, the team can often garner greater value out of a player at a lower cost. Players are under team control for six years after achieving rookie status so it would be beneficial for a team to be able to project the player’s value over those years so they can try to acquire prospects that should succeed and provide maximum value with minimal costs. Using both advanced metrics and scouts’ rankings to combine the two schools of player evaluation, we developed models to project a player’s average WARP (Wins Above Replacement Player) over their first six seasons when under team control.

This document is currently not available here.