Following last week’s PED news coming out of Miami, I was actually planning on this post being about baseball’s real drug problem: drunk driving. But that was before Jim Caple’s post on ESPN.com on Friday, where he critiqued the use of WAR by the more statistically inclined. Clearly, a statistical measure is much less important in the long run than the fact that a lot of players, managers and broadcasters are taking stupid risks with not only their lives, but also the lives of others. As a self professed stat head, I had some issues with Caple’s characterization of those who use WAR and how we use it.
My issues weren’t with Caple’s discussion of the drawbacks of WAR. It’s true that there is no definitive WAR figure. Baseball-Reference and FanGraphs not only have different versions that use different metrics, but they also put “replacement level” at a different place. Baseball-Reference has a team full of replacement level players winning about 52 games, while FanGraphs gives them 43 wins. On top of that, a good number of baseball teams have their own internal versions of WAR that are completely proprietary.
It’s also true that one of WAR’s strengths, the fact that it takes defense into the equation, also leads to one of its biggest problems: as defensive statistics are new and not based upon as clear cut of information as did a player get a hit or make an out, they are less reliable than offensive or pitching statistics. When you see a large swing between a position player’s rWAR and fWAR, it’s almost always because Baseball-Reference and FanGraphs judge their defense differently.
My issue with Caple’s post was that he argued that those of us who use advanced statistics don’t understand WAR’s flaws on the basis that we have issues with Miguel Cabrera winning the American League MVP over Mike Trout.
Let’s stop pretending that anyone thinks that Miguel Cabrera isn’t one of the best hitters in the game. But the fact is that he won the MVP because he won the Triple Crown, and that means he won the MVP because he led the league in RBIs. And the biggest reason he led the league in RBIs was not because he improved off of his 2011 effort, but because Austin Jackson raised his on base percentage 60 points and because Cabrera bats third in that line up.
The mental gymnastics that the people who voted for Cabrera went through were in competition for a medal. Some said he deserved credit for being willing to switch to third base when Prince Fielder signed, despite him being terrible there defensively and Mike Trout being an excellent defensive center fielder, a more demanding defensive position than the hot corner. Then there was the argument that Cabrera made the playoffs and the Trout did not. But Cabrera only made the playoffs because the Tigers played in the worst division in baseball, while Trout’s Angels finished with a better record despite playing in the very tough AL West.
Trout was better than Cabrera last year. WAR is not the reason why Trout was better than Cabrera, but is merely one (of many) pieces of evidence in his favor. Trout put up nearly identical offensive numbers while being much more valuable on the base paths and defensively. If the BBWAA wanted to award Cabrera the MVP based on meeting a neat but archaic set of criteria, that’s their prerogative. But just be honest about it.
And don’t tell the statistically inclined about the problems with WAR. We’re aware it needs refinement. We’re aware that defensive statistics need a lot of fine tuning. We’re aware that, eventually, there is probably going to have to be some agreement about how WAR is calculated. We’re aware that Miguel Cabrera is one of baseball’s best hitters. We don’t think that the MVP should just automatically go to whomever has the highest WAR. But in 2012, Mike Trout was the American League’s, and all of baseball’s, best player. With or without WAR.