Did WAR Destroy the MVP Dialog?

On Wednesday, I wrote about considered one of my favourite subjects: The influence of sabermetrics on the apply and evaluation of baseball. Particularly, on this case: How MVP voters behave within the post-Fireplace Joe Morgan period. And for these of you who acquired to the top of that 2,000-word submit and didn’t really feel sated, there’s excellent news! This was not the query I really got down to reply once I began kicking the subject round.

Welcome to Half 2.

The very title of the MVP award invitations voters to contemplate the worth of a sure participant’s contributions. For practically 100 years, that was a difficult proposition. How do you weigh variations in place, in taking part in fashion, park elements, hitting versus pitching versus fielding versus baserunning? It’s sufficient to boggle the thoughts.

One of many earliest and most enduring tasks of the sabermetrics motion has been the pursuit of a catch-all metric that captures your entire image. To measure each participant’s manufacturing, put it in context, and spit out a single quantity that claims he was value X variety of runs or Y variety of wins. The implications of such a quantity are apparent for an enterprise comparable to figuring out probably the most worthwhile participant within the league in a given season.

Critics from days passed by — or critics from now, who’re so behind the instances they solely sound like they’re from days passed by — would wield WAR as a straw man in opposition to extra numerate analysts. “In case you imagine in WAR, why not simply vote the WAR leaderboard?”

The reply is that WAR, whereas revealed to the tenth of a win, is a mannequin constructed on assumptions that may not maintain to that degree of precision. It’s a very good start line, but it surely’s not the entire dialog, particularly in an in depth race.

And you may inform that WAR will not be settled science as a result of even the sabermetrics eggheads can’t agree on what WAR is. FanGraphs, clearly, has its personal WAR, and I exploit that as a result of I’m a very good firm man. However there are dozens of MVP voters, and plenty of hundreds of people that publish evaluation of baseball for public consumption. Sadly, a few of them get their data from different sources. (However we’re engaged on that, I promise.)

When folks discuss WAR, they’re both speaking about FanGraphs WAR, Baseball Reference WAR, or WARP, from Baseball Prospectus. (The “p” on the finish is for PECOTA, I believe.) Every makes its personal assumptions concerning the function of protection in pitcher analysis, simply to call one level of divergence. So whereas all three mainstream WARs normally arrive at comparable conclusions, they don’t arrive at the identical conclusion. Which is what you’d have to make “kind by WAR” a coherent award voting technique.

I’ve compiled a listing of each participant since 2000 to complete within the high 10 in MVP voting, and their WAR totals that yr for fWAR, bWAR, and WARP — 481 seasons in all. On just one event — Justin Turner in 2017 — did all three main flavors of WAR agree on a participant’s worth to inside a tenth of a win. So I really feel fairly assured that Turner was value precisely 5.6 WAR in 2017. Every part else is open to interpretation.

There’s one other complicating issue: All three flavors of WAR get up to date on occasion, as wins above substitute will not be some inviolable idea that was thought-about excellent when it was first willed to humanity by the gods. It’s a statistical mannequin, in spite of everything, and people should get up to date when new data emerges. So the WAR totals you see now won’t be precisely the identical as they have been when voters have been wanting on the numerous websites’ leaderboards again within the day.

So there are alternative ways of answering the query of how carefully WAR and the MVP vote line up. From a normative standpoint, you possibly can take a look at it because the voters getting the query proper, or as a examine on WAR — do the numbers match what we see with our eyes, or is that this math left to run amok?

From 2000 to 2023 (since we don’t have the MVP vote tallies from this yr but), there have been 134 particular person league leaders in WAR. That’s 24 seasons instances two leagues instances three WAR manufacturers, minus the 5 years (2000 to 2004) for which BP doesn’t have revealed WAR information.

Out of these 134 particular person participant seasons, the WAR chief — for every league and WAR sort — completed within the high 10 in MVP voting 120 instances. Listed below are the exceptions.

It ought to shock nobody that beginning pitchers are closely represented right here. MVP voters hate pitchers for causes that aren’t solely clear to me. A beginning pitcher can lead the league in WAR and never get the time of day from voters. Greinke, apparently, can lead the league in two completely different WARs and nonetheless be ignored.

On Wednesday, I dragged Max Nichols, the poor soul who forged the one dissenting MVP vote in opposition to Carl Yastrzemski in 1967. At present, I’d wish to reward one other singularly iconoclastic voter: Nick Piecoro of the Arizona Republic in 2018.

That yr, Christian Yelich ran away with the MVP vote, successful 29 first-place votes for a season by which he missed the Triple Crown by two dwelling runs and one RBI. He hit .326/.402/.598, which was not solely a paradigm-shifting breakout for Yelich himself, but additionally a transformative one for his staff. Yelich, who was traded from Miami to Milwaukee the earlier winter, led the retro Brewers to the no. 1 seed within the Nationwide League. Narratively, it was not in contrast to Yaz in 1967: an unbelievable particular person season on the heart of the league’s greatest staff degree story.

And but Jacob deGrom was so significantly better. This was the 10-9, 1.70 ERA season that spawned one million memes and ended up, someway, going underrated by voters. The Mets ace led the league in all three WAR classes and had greater than a win on Yelich in all three instances, as much as 2.6 WAR over the eventual MVP in accordance with B-Ref. And but he solely completed fifth.

Not each MVP debate has a proper and fallacious reply. This one did, and Piecoro was the one voter who discovered it.

A lot of the remainder of the ignored WAR leaders topped the desk in WARP, which is usually a bit stingy and infrequently spits out an surprising end result. Nonetheless, it’s a reminder that Soto was actually good even in a down yr in 2022, and McCann was massively underrated usually.

I did need to spotlight Markakis as a result of, on the threat of being unkind, he’s the final individual I anticipated to guide the league in something, ever. He led the league in bWAR in 2008 by having, in some respects, a typical Markakis season: He hit round .300, with about 20 homers and 40-odd doubles, and performed good protection in proper. What was uncommon is that he additionally walked a career-high 99 instances, which elevated his OBP to .406, and DRS credited him with 22 runs saved above common, which is a ridiculous quantity. That was sufficient to spice up a really, superb season to 7.4 bWAR (and 6.1 fWAR, so it’s not like B-Ref was an outlier right here), and in a weak season that topped the AL.

This was my favourite second in researching this piece. Discovering out Markakis led the league in WAR as soon as was like discovering out Cesar Tovar was really higher than Yastrzemski in 1967. Suffice it to say, the voters didn’t discover. Not solely did Markakis miss out on the highest 10 in MVP balloting, he didn’t get a single vote from anybody.

Now that we’ve seen the WAR leaders who didn’t get a lot love from the MVP voters, let’s take a look at the opposite aspect of the equation: MVPs who didn’t lead the league in any model of WAR. Since 2000, there have been 18 such instances. (By the way, that listing consists of the previous two Twins MVPs and the previous three Phillies MVPs, so in instances the place voters dislike WAR they apparently love sarcasm and smoked and/or cured meats.)

That group of 18 additionally consists of each MVPs in 2000 and 9 of the 20 MVPs between 2000 and 2009. That inflection level would appear to carry with the historic proliferation of superior stats. The Fireplace Joe Morgan period resulted in 2008, and the early 2010s have been to the anti-intellectual voter what the Hundred Days have been to Napoleon. These included the Jack Morris Corridor of Fame debate and the Miguel Cabrera vs. Mike Trout MVP campaigns of 2012 and 2013.

However current historical past doesn’t line up with the WAR leaderboards both. From 2020 to 2022, 4 out of six MVPs — José Abreu, Freddie Freeman, Bryce Harper, and Paul Goldschmidt — did not high any WAR leaderboard. After all, all of these races have been arduous instances: 2020 due to the 60-game season, 2021 and 2022 due to a diffuse and pitcher-dominated WAR leaderboard.

Certainly, many of those supposedly WAR-deficient MVPs solely missed out on league management by fractions of a win. So let’s slim that group to MVPs who gained in years the place a special participant led the league in at the very least two completely different WAR classes by at the very least a win in every case.

This listing doesn’t embody instances just like the 2000 AL race or each races in 2006, the place it’s clear {that a} purely WAR-based voter wouldn’t have given the MVP to the participant who gained in actual life, however the high of the leaderboard is shut sufficient that it was much less clear which participant ought to’ve gained. (I had forgotten, for example, how good Jason Giambi was in 2001 and Carlos Beltrán was in 2006.)

Each considered one of these instances has a powerful narrative foundation. Cabrera gained the Triple Crown in 2012. Donaldson, Altuve, and Yelich turned up-and-coming groups into juggernauts. No person respects pitchers or A-Rod, and for some motive Trout turned a proto-culture struggle volleyball within the waning days of the game’s battle over empirics.

From 2000 to 2004, FanGraphs and Baseball Reference WAR had the identical chief 9 instances out of 10 possibilities. That may have one thing to do with methodological convergence, but it surely most likely has extra to do with Barry Bonds and Alex Rodriguez simply dragging everybody else within the league throughout that interval. The voters — most of whom had most likely by no means heard of WAR on the time — gave the consensus WAR chief the MVP 5 instances out of these 9 possibilities, and voted him second on two different events.

Within the 19 seasons that adopted, one participant led the league in all three WARs on 15 events. The voters agreed 10 instances.

In three of these instances — Trout in 2013 and 2015 and Decide in 2017 — the consensus WAR chief got here second. The opposite two concerned pitchers: deGrom in 2018 and Shane Bieber in 2020, and whereas neither gained the MVP, they’re the 2 most up-to-date full-time pitchers to complete within the high 5.

Has WAR turned MVP voting right into a leaderboard-reading train, as critics predicted? Probably not. Due to the three publishers’ refined variations in methodology, I’m unsure that it might probably. When there’s consensus, the voters normally observe go well with (once more, until there’s a pitcher concerned). However that was principally the case 20 years in the past, when on-base proportion was cutting-edge.

If MVP voting has change into predictable within the 2020s, and if it’s the results of pressures and improvements introduced on by sabermetrics and on-line media, I don’t assume it’s as a result of everyone seems to be simply blindly following WAR. We’re all getting higher data now, and other people with clubhouse entry and award votes are higher outfitted to make use of that data. Whether or not that results in conformity is of secondary significance to the standard of the evaluation.

Even within the Nineteen Sixties, there was a threat of being brigaded for expressing an unpopular opinion — simply ask the ill-fated Max Nichols. And concern of the dogpile has solely grown immeasurably since then. It’s an incentive to not disagree with the consensus, certain, but it surely’s a fair larger incentive to get your information straight. Certainly no one would place a league-average utilityman over an 11-WAR slugger on an MVP poll in 2024. (With that mentioned, nothing would make me happier than discovering out, in per week’s time, that somebody voted for Matt Vierling over Decide for AL MVP. I’ll throw an precise social gathering if that occurs, and also you’re all invited.)

So is it that voters not have the braveness of their convictions, or is it simply that they’ve higher convictions as of late?