The MVP finalists are in and you’re never gonna believe this, but they’re ordered almost exactly like WAR.
In the American League, the finalists are Mookie Betts (10.4), Mike Trout (9.8), and Jose Ramirez (8.0). They ranked 1-2-3 in Fangraphs WAR, or fWAR, which is not the kind of thing you’ll find on the back of a baseball card.
In the National League, the finalists are Christian Yelich (7.6), Javier Baez (6.3), and Nolan Arenado (5.6). They ranked 1-2-5 in Baseball-Reference WAR, or bWAR, which sounds like something that makes your eyes glaze over in a Stats 101 textbook.
The slavish devotion to WAR has poisoned not only postseason awards voting, but Hall of Fame voting, too, even though the very people who created and regularly update the various formulas comprising it routinely note that differences smaller than one to two wins should not be considered definitive.
And yet we’ve had entire discussions hijacked by it. TIM RAINES IS A HALL OF FAMER BECAUSE WAR DEMANDS IT. TROUT MUST WIN THE MVP AWARD EVERY SEASON BECAUSE WAR DEMANDS IT. JUSTIN VERLANDER WAS ROBBED IN THE 2016 CY YOUNG RACE AND WAR PROVES IT.
For the uninitiated, WAR stands for Wins Above Replacement, and it’s not a stat, per se. It’s not like batting average, which is computed with one fixed formula. To steal the language of ESPN analyst and former Blue Jays executive Keith Law, who wrote a book called Smart Baseball, it’s a construct.
The idea is that WAR combines a player’s various contributions – offense, defense, pitching, baserunning – and assigns them a number of wins relative to what a team could expect if an average Triple-A player started in his place. There are many different ways to arrive at this number – bWAR and fWAR differ in their lifetime assessment of Hall of Famer Nolan Ryan, for instance, by a whopping 22 wins – but the more detailed the data, the better a player can be evaluated.
I’m not a WAR denialist. I use it as part of my Hall of Fame evaluation, and I’ve referenced it in years when I’ve had MVP or Cy Young votes. It’s a useful tool. My problem is with WAR absolutists who vote based on differences that are statistically insignificant, or who ignore other means of evaluating a player. A 6.0-WAR player and an 8.0-WAR player might be the same; it’s not enough to stop there.
Take this year’s AL MVP race. Betts led the league in WAR, will win the award, and deserves it. Hard to argue with Trout or Ramirez, either, based on performance, though I’m certainly open to the idea of an MVP coming from a winning team, and not Trout’s sub-.500 Angels.
But there are other deserving candidates whose WAR won’t give them a sniff. Red Sox slugger J.D. Martinez is one of them. For five months, he flirted with the Triple Crown. Because he’s mostly a DH and no better than an average fielder when he does don a glove, he never had a chance, though his 5.9 fWAR is roughly in range of Ramirez.
Some advanced stats make a case for Martinez, such as his weighted runs created (170, 3rd), base-out runs added (73.36, 1st), win probability added (5.4, 3rd) and . . . there’s no need to lose ourselves any further in those weeds.
But there’s also the non-statistical measure of Martinez’s impact. The Hank Aaron Award means something. That teammates such as Betts considered his David Ortiz-like presence the key to the lineup means something. The way he impacted his teammates’ approach to launch angle and batting practice and video study means something.
WAR measures none of that, and there’s a safety in shrugging that if it can’t quantified, it’s not worth considering, just as there’s safety in submitting a WAR-approved ballot, content in the knowledge that no one on Twitter will yell at you.
If I had an MVP ballot this year, I would’ve ranked Martinez ahead of Trout. The fact that the Red Sox were the best team in baseball, with Martinez posting monster numbers in the heart of their lineup, has to mean something. And I know the counter-argument – swap Trout in his place and the Red Sox might be even better – but this isn’t a hypothetical exercise. Martinez actually produced there, impacting actual wins, and delivering an actual World Series.
Sometimes it feels like WAR tells us more about what a player should’ve done than what he actually did. Bill James ignited a fierce discussion on this front last year when he noted that WAR rated Aaron Judge and Jose Altuve as virtually equivalent, even though Altuve clearly had the better season (and, it should be noted, was rewarded with the MVP).
James’ point is that the “wins” in “wins above replacement” should relate to real-life ones. Others argue that we should only measure a player’s underlying contributions and not, say, penalize Bryce Harper for walking 130 times if the Nationals couldn’t drive him in on 110 of them. I’ll leave that discussion to the experts.
All I know is that anyone who starts and ends their evaluation with WAR isn’t forward-thinking. They’re lazy.