Friday, April 13, 2012

WAR: What is it good for?

Please forgive the title, but I couldn't resist. Today's topic is Wins Above Replacement, or WAR. I've been using WAR on this site as a nice shorthand to compare player seasons, but it's still unfamiliar to a lot of fans. So I'm going to reprint the entry about WAR from my new book, The Book of Baseball Literacy: 3rd Edition.


WAR is fairly new sabermetric statistic that aims to quantify a player’s contribution to his team in the field, at the plate, and on the bases compared to that of a generic “replacement player”—meaning a typical player readily available from the minor leagues (which is not the same as an average major leaguer). Like Win Shares, WAR is complicated to calculate, and even among those websites that do make the effort—including Fangraphs.com, Baseball-Reference.com, and Baseball Prospectus—they each make slightly different calculations, so the numbers for any given player don’t exactly match up.

One thing to realize with WAR is that the calculation literally translates to the number of wins a player is responsible for above and beyond the contribution of a replacement player. For example, in 2010, Josh Hamilton registered a WAR of 8.7 (according to Fangraphs), the highest in baseball. That means his bat, glove, and legs personally delivered 8.7 wins to the Rangers more than an average minor league outfielder.

Does that seem low to you? At first glance, it does to me. I mean, Hamilton was the best player in the league that year, and he was only responsible for 8.7 wins? At second glance, however, it starts to make sense. Because of the nature of baseball, a team filled with Triple-A replacement players would probably win at least 40 games. A team filled with Josh Hamiltons would probably lose at least 40 games. The 82 games in between are what separate major league–quality teams from minor league–quality teams, and there are 25 players on the roster all contributing to the club in one way or another. In that context—25 players diving up 82 wins, or an average of 3.28 per player—Hamilton’s 8.7 makes more sense. WAR suggests that he was between two and three times as good as the average player on his team, and that seems about right.

It’s possible to register a negative WAR, but in practice, players who do usually don’t last long in the majors because teams will be quick to replace them with a cheaper, younger alternative from the minors. (However, some veterans do keep their jobs despite scoring poorly; Melky Cabrera, Carlos Lee, and Adam Lind all scored negative WARs in 2010, but still have major league jobs because teams are hoping against hope that they’ll return to their former glory.)

WAR is an exciting new statistic that is accepted in the sabermetric world but has a long way to go before casual fans can believe in it. But it’ll get there.

No comments:

Post a Comment