In the Great Baseball Analytics Wars, we’ve long known that the nerds won. Now it just feels like they are rubbing it in.
First came a regular season in which Tampa and Milwaukee eschewed the traditional concepts of a starting pitcher, either by starting a reliever and switching to a starter for the middle innings (a middler?), or by just running out a string of relievers to eat up nine innings. Analysts have long argued for the potential effectiveness of such a strategy, since it reduces the likelihood that the “starter” will be raked during his third time through the batting order, where evidence shows there is a noticeable decline in performance. Still, it was assumed that managers would be loath to go there. Imagine telling Jack Morris that he was going in for the second inning. He would have jammed a tobacco wad down Sparky Anderson’s throat.
The pro-analytics forces did suffer a setback of sorts in the World Series, when data-focused Dodgers manager Dave Roberts saw most of his moves blow up in his face like so many short-fused firecrackers, although he was beaten by a team in the Red Sox that is known to love the math, so that was probably a wash.
Now this week comes news that in the voting for the American League Most Valuable Player Award, Boston’s J.D. Martinez, despite almost hitting for the dang Triple Crown, is not among the three finalists. Those players are Boston’s Mookie Betts, Anaheim’s Mike Trout and Cleveland’s Jose Ramirez. (Although MLB uses the “finalist” term, they are the top three vote getters, since voting from the membership of the BBWAA is already completed.)
Frank Thomas called Martinez’s omission a “disgrace.” Harold Reynolds, another player-turned-analyst who has much contempt for the use of spreadsheets, called it “flat ridiculous.” There’s a decent chance he was shaking his fist at the nerds when he said it.
The explanation, though, is simple. Martinez was primarily a designated hitter. Despite his finishing second in the AL in home runs (43) and batting average (.330) and leading in RBI (130), Martinez was just ninth in the league in Wins Above Replacement (6.4). Betts was first (10.9), Trout second (10.2) and Ramirez tied for fourth (7.4). WAR, another stat that tends to make old-timers cranky, takes into account defence and base running, which is how someone like Martinez can have a remarkable offensive year at the plate and still not be within sniffing distance of his teammate Betts, an all-world outfielder who also stole 30 bases.
I should note, here, that I’m mostly joking about the nerds. I understand the value of analytics. I, for one, welcome our new nerd overlords. It’s just interesting that the MVP race has essentially become a proxy for highest WAR, and it shows how far attitudes about player value have come. Twenty years ago, the AL MVP award went to Juan Gonzalez of the Rangers in a landslide. Gonzalez had 45 home runs and a hilarious 157 RBI that season, and in 1998 that was all that needed to be said. But he was also 15th in the American League in WAR (4.9), owing primarily to the fact that he was a defensive liability but also because he didn’t draw many walks and he struck out a lot. The AL leader in WAR that season, Alex Rodriguez (8.5), finished ninth in the in the MVP voting. He was a much more complete player than Rodriguez, and this was back in his Seattle days when everyone still liked him, and yet he was an MVP afterthought because his power stats weren’t quite gaudy enough. He was the Mike Trout of 1998, but without the end-of-season recognition.
That the MVP voting now closely follows the WAR leaders is a fairly recent development. As recently as 2013, Miguel Cabrera won the AL award over Trout even though Trout was first in WAR and Cabrera was fourth.
But enough voters are now considering “value” in the way that analysts have long advocated: by accounting for everything a player does on the field, and not just the counting statistics that can be dramatically affected by a player’s home park and the teammates in the lineup around him. Where RBI totals used to largely reflect the MVP voting, with the exception of the odd leadoff hitter who snuck in there, now Ramirez is the lone AL finalist anywhere near the league leaders in RBI, and he was fourth. This despite the fact that it would not take you very long in an average bar, or even an average broadcast booth, to find someone who says that J.D. Martinez should be the AL MVP because he drove in more runs than anyone else.
But the best proof of how much the conversation around value has shifted in baseball will come next week when the final AL MVP totals are released. Oakland’s Khris Davis led the league in home runs with 48 and he was second behind Martinez with 123 RBI. But he’s a negative defender who strikes out a lot and doesn’t walk much, which means his WAR is a relatively paltry 2.9. Will the AL home run leader even make the top 10 in the MVP vote? Top 20?
Last year, with 43 home runs and 110 RBI, Davis finished 22nd in the voting.