Wins above replacement (WAR)

Wins above replacement compresses a player's overall contribution, batting, baserunning, fielding, and for pitchers, run prevention, into one number: the estimated number of additional wins that player provided compared with a "replacement-level" player, meaning a readily available minor-league or bench-level substitute. A player with 5.0 WAR is roughly five wins better over a season than what a team could get for free.

WAR is built by converting each component of play into runs relative to league average, applying a positional adjustment, converting the replacement-level baseline into runs, and then dividing the total by an estimated runs-per-win figure, typically around ten. It grew out of sabermetrics, the empirical study of baseball performance, as a way to compare players across positions and eras on one scale. Different publishers calculate WAR differently, especially for pitching, so values for the same player can vary between sources.

Teams and analysts use WAR for roster construction, salary arbitration arguments, and awards debate, since it accounts for defense and position scarcity in a way that traditional stats like batting average do not. The main pitfall is treating WAR as an exact figure rather than an estimate with real uncertainty, and comparing WAR values computed by different methodologies as if they were interchangeable. It has since been adapted, loosely, to other sports alongside metrics such as player efficiency rating.