Well, they've gone and done it again. They've invented another omnibus number that's supposed to describe a baseball player's performance. The new metric? "Win values," introduced here and further explained here by Dave Cameron.
Essentially the method is this: Add batting runs and fielding runs, throw in a positional adjustment, divide by 10 -- because 10 runs equals one win -- and
voila, you've got "win value" (or "value wins," depending on which page you're looking at). Makes sense to me, and I suspect the results are at least as good as "wins above replacement player" (WARP) and "win shares".
Of course, the comparisons aren't so simple. The "win values" method doesn't consider timing -- a solo home run counts exactly the same as a grand slam -- while "win shares" does. Win Value doesn't consider an outfielder's throwing arm or a middle infielder's ability to turn the double play (though I understand they're working on both). As far as I know, none of these methods (or any other) consider baserunning other than stolen-base attempts. And, of course, nobody knows what to do with the catchers' defense.
We're getting better all the time, though, and I suspect that "win value" is now the best we've got.
What does the method say about 2008? It says the MVP voters had a pretty good year. National League MVP
Albert Pujols led the majors with 9.0 value wins; American League MVP
Dustin Pedroia finished second in the AL with 6.6 value wins (behind only
Grady Sizemore's 7.0). You might recall that I argued for
Joe Mauer to get serious consideration, and he finished fourth in the balloting, which now seems about right, because Mauer was fourth in the AL with 5.7 value wins. What
doesn't seem right?
Justin Morneau finished 28th in the league with 3.1 value wins -- dead even with
Kurt Suzuki and
Curtis Granderson -- and somehow finished second in the MVP balloting. Oh, and speaking of overrated first basemen,
Ryan Howard finished second in
his MVP balloting … and 32nd in his league with 3.4 value wins.
One can certainly argue with the method. I wonder, though, how likely it is that a player who comes in 28th or 32nd, according to this method, really was the second-most valuable player in his league. I'm not an expert or anything. But it does strike me as somewhat unlikely.
(If you want to play around with value wins,
here's a good start.)
Rob Neyer
