Sunday night, and I mean late Sunday night, Yahoo!’s Jeff Passan published a long, rambling column about the deficiencies – some real, some supposed – of Wins Above Replacement. Shortly, FanGraphs’ Dave Cameron responded via Twitter, which began a “discussion” that would last well into the night. As a charter member of the Jeff Passan Fan Club and the Dave Cameron Booster Society, I didn’t get into the middle of things …
If you want to read about a bunch of the outstanding issues, I heartily recommend both pieces (and you can skip the Twitter interplay in the middle; Twitter’s really not the place for informed, insightful analysis, which is why I do sometimes get frustrated when readers expect that sort of thing in 140 characters).
Article continues below ...
Since they’ve done all the hard work, I’m not going to inject myself into the debate here. Not really. I do want to riff for a moment on this passage from Passan’s piece, though:
… And as much as Cameron has warned that "we don’t encourage you to use any version of WAR as the be-all, end-all of analysis," FanGraphs sorts both its pitching and hitting leaderboards by WAR. When Cameron writes his annual trade value column, he uses projected WAR. It’s not the be-all, end-all, but FanGraphs subtly suggests its readers use it as such.
No, I don’t believe FanGraphs is suggesting any such thing. First, they’re not actually leaderboards. I mean, they are. But they’re really just landing pages. Unless you want all your hitters and pitchers sorted by their last name when you got a page full of players’ statistics, you have to pick some statistic to start with. When I was a lad, the Sunday newspaper listed all the batters in order of their batting averages.
No sorting. You got them by batting average, and you liked it.
But you gotta choose something to start with, whether you’re a newspaper in 1984 or a Website in 2014. In 1984 they chose batting average, because a lot of people still figured that was the single best measure of a player’s value.
But even then, I don’t think most people thought it was the be-all, end-all. If they had, the Most Valuable Player Awards would have gone to the batting titlists every year. Which they didn’t. They just figured it was probably the single best measure, maybe just a little better than Runs Batted In.
So I’ll ask Jeff (and you) this … If not WAR, then what should be the default setting when you land on that page? Batting average? RBI? OPS, perhaps? I don’t think anyone at FanGraphs is trying to say WAR is the be-all and the end-all and all that. Rather, among all the measures we’ve got, WAR remains the single best at measuring a player’s overall value. I’m the best writer in my house right now. Doesn’t mean I’m perfect, and I’m sure as hell not the be-all, end-all of anything.
Seriously, though. If you had to grab nine players right now, straight off a leaderboard, for your new franchise, which leaderboard are you going to use?
Correct! You win the kewpie doll today.
I don’t believe in WAR as a be-all, end-all for the simple reason that I don’t believe in be-all, end-alls at all. Everything’s just another data point, whether we’re talking about people or history or politics or policy or (yes) baseball. But WAR’s a really useful data point, isn’t it. You might not believe that Alex Gordon is really one of the two or three best players in the American League – and yeah, I think the defense thing is a little weird, too – but isn’t this a discussion worth having? Without WAR, would Ben Zobrist have gotten his just due? I don’t think of WAR as a conversation-ender and I don’t know anyone smart who does … but it’s one hell of a conversation-starter, don’t you think? When you’re casting about for MVP candidates, shouldn’t you start by scanning the WAR leaders? I would submit that if you don’t, then you’re doing it wrong.
When Troy Tulowitzki went down, I began to cast about for other MVP candidates. When Andrew McCutchen went down, I cast about for other MVP candidates.
Except then McCutchen came back after just a couple of weeks. And the Pirates started winning again. And Monday night I saw this:
… which got me to wondering if McCutchen’s two missing weeks might seem inconsequentional next to his otherwise-big numbers and the Pirates’ impressive (so far) stretch run. So I checked the landing page.
McCutchen ranks just sixth in the National League … but it’s a strong sixth, and it doesn’t take much imagination at all to make him your most valuable player in the National League. McCutchen’s got 5.3 FanGraphs WAR (fWAR), just 0.5 wins behind Jonathan Lucroy in the top spot. But 0.5 is nothing in this context. Nobody with half a sabermetric brain would pretend it is. Also, McCutchen’s essentially tied at 5.3 WAR with Jhonny Peralta and Jason Heyward, both of whom are credited with roughly 30 more fielding runs than McCutchen. But that doesn’t seem quite right, does it? This is where we get to use our own judgment, regress those fielding metrics just a bit, and suddenly McCutchen’s fourth on our list rather than sixth … and just a smidge behind Lucroy, Giancarlo Stanton, and Hunter Pence.
And now we get to have a really tremendous discussion about those four, along with Peralta and maybe Buster Posey, too. Now we get to really dive in and consider those fielding metrics and clutch hitting and team leadership and the metaphysical meaning of “value,” anyway.
But you know, McCutchen’s got a lower batting average than his teammate Josh Harrison, he’s got many fewer homers and RBI than Justin Upton, and he’s scored many fewer runs than Anthony Rendon. Without WAR, I don’t know that I would have appreciated how well McCutchen has played this season.
Be-all and end-all? Of course not. But if you’re ignoring it and you’ve got an MVP ballot, I’ll bet you miss somebody.
p.s. and speaking of missing somebody, yes I’m aware of Clayton Kershaw, who belongs in there somewhere, too. Just look at that WAR!