Wins Above Replacement (WAR) should never stand alone as the deciding factor in giving player awards. First of all, it is an imperfect system. Second, I shouldn’t even say system when there are actually several systems of calculating WAR. And each system – the key ones being from Fangraphs, Baseball Reference and Baseball Prospectus – has its own flaws. Therefore, since there isn’t a consensus on which system to use, and there is a general recognition of the shortcomings of each, any attempt to use WAR as an “absolute” will fail.
Nonetheless, if someone compares players within one system – I prefer Fangraphs’ method – they can usually get a pretty accurate idea of who the better players are. And whether you choose Fangraphs or one of the other two, it’s still a better way to judge than only looking at a player’s BA, RBI and Fielding%.
I came over from the Dark Side of statistics a few years back now. Let me be clear what I mean when I say that: I don’t mean that fans who don’t know what WAR is or who praise RBI are terrible people… rather, they’re in the “Dark”, like I was for the first 25 years of my fanhood. These days, there are many, many excellent stats that let us see a deeper and more accurate picture of what players are doing on the field, and WAR is an aggregate of some of the better ones.
With that in mind, I was genuinely surprised when I realized that the 2015 ROY and MVP finalists were very closely matched to Fangraphs’ WAR (fWAR). Not only that, but even though the pitchers chosen for Cy Young contention were tied directly to ERA, they were also pretty close to fWAR standings. And if Sonny Gray (14-7) is any indication, pitching “Wins” were not of paramount importance in the decision-making. Perhaps most surprisingly, only two of the six MVP finalists came from playoff teams. (Somewhere, Brian Kenney is cheering.) All of the above is particularly true in the NL.
Check this out:
NL Rookie of the Year Finalists (with fWAR rank):
Kris Bryant (1), Matt Duffy (2), Jung-Ho Kang (3)
NL Cy Young Award Finalists (with fWAR rank):
Clayton Kershaw (1), Jake Arrieta (2), Zach Greinke (4)
NL MVP Award Finalists (with fWAR rank):
Bryce Harper (1), Paul Goldschmidt (2), Joey Votto (3)
The National League nominees are essentially the top fWAR players, with the one exception of Zach Greinke, who overcame his 4th place fWAR with a league-leading ERA.
Does this mean that I think the writers who vote for NL MVP are all fans of WAR? Definitely not. But I do think there are a couple of things in play here: first, most of the time, top MVP candidates have numbers that impress both traditional and advanced metrics fans, so there’s bound to be some overlap. Secondly, though, while the writers might not believe in WAR, I would bet that a great number have come to believe in some of the more recently favored stats that make up the components of WAR (or similar stats). Even a shift in focus from BA to OBP and SLG will get you closer to WAR, and if any of the voters are also looking at fielding stats that go deeper than Fielding%, that too would bring them closer to the world of WAR.
And I think it’s good if they don’t begin and end with WAR. In fact, I wouldn’t mind if WAR fell a bit to the wayside and voters focused on things like wRC+, OPS+, UZR, wSB and ERA-. Baseball is too complex to boil it down to an absolute, single number. But WAR can still be an excellent starting point, and if the awards voting starts to reflect that, then I believe baseball has made a positive step forward.