clock menu more-arrow no yes mobile

Filed under:

The Year of the Impossible MLB MVP Pick

Aaron Judge or José Altuve? Kris Bryant or Anthony Rendon or Giancarlo Stanton? The list goes on, and so do the conundrums: In 2017, both leagues present MVP logjams the likes of which we haven’t seen in decades, and for the first time in a long time, WAR can’t help.

An illustration of various major league MVP candidates Getty Images/Ringer illustration

This December will mark the fifth anniversary of my admittance to the Baseball Writers’ Association of America. Although I’m halfway to having a Hall of Fame vote, which active members earn after 10 years, I’ve never voted on a single-season award, which has no service-time threshold. Because only 30 BBWAA members vote on any one end-of-season award; because I belong to the BBWAA’s crowded New York chapter; and because non–Hall of Fame voting duties rotate randomly within chapters from year to year, my number has yet to be called.

The real benefit to being a BBWAA member is the access, not the input on awards, and in an era with a thorough statistical record and increasingly comprehensive value rankings, the subjective Cy Young or MVP opinions of a small sample of writers seem to matter less. Still, despite a creeping apathy about awards, I’ve felt a touch envious of colleagues who’ve had their names selected in years with especially contentious races, when one clearly deserving candidate (in my mind) was in danger of being snubbed by a subset of the writers that saw the race some other way. In those seasons, I’ve wished I could stand up and be counted in the Mike Trout camp—and, naturally, get a quick column out of explaining my righteous rationale.

Not this year. I still don’t have a vote, but this time I’m happy to have that burden lifted from me. This year’s award decisions—excluding Manager of the Year, which is always murky to me—are either obnoxiously easy or really, really hard.

Some of this year’s races have runaway winners. With apologies to Rhys Hoskins, Paul DeJong, Manuel Margot, and Austin Barnes (who could have a case, if you factor in framing), Cody Bellinger is the National League’s best rookie. In the American League, the decision is even easier: By FanGraphs WAR, Aaron Judge has had the sixth-best rookie season since 1901, and no other AL rookie this year is within five wins. NL Cy Young seems like a clear win for Max Scherzer.

Everything else is basically too close to call. Corey Kluber and Chris Sale are separated by a sliver in the AL Cy Young race—take the average of their WAR values at FanGraphs, Baseball-Reference, and Baseball Prospectus, and Kluber, who also leads in Tom Tango’s Cy Young Tracker, has the edge by six-tenths of a win—and because they’re both having such strong seasons, they’re also in the thick of the AL MVP race. That’s where this awards season really sets itself apart: In both leagues, we’re looking at MVP logjams the likes of which we haven’t seen for decades. And while in recent years BBWAA vote totals have been aligning increasingly closely with WAR, WAR won’t be of much help in guiding this season’s electorate.

Let’s start with the evidence that BBWAA vote totals have been dovetailing with WAR. The graph below, which was created with data provided by Dan Hirsch of The Baseball Gauge, shows the percentage of MVP point totals awarded to the top two, top five, and top 10 players by Baseball-Reference WAR in every season of the DH era. The straight lines reveal the general trends over time, while the dotted lines display rolling averages of results from the previous five years.

Although the results fluctuate widely from year to year, the overall trend is toward the top players by WAR amassing a larger share of the MVP points—as one would expect, given that WAR, which is growing in prominence, wasn’t a tool that voters could consult during past decades. The rolling averages for the top-10 and top-five groups are at all-time highs, and even the top-two group is on the upswing. In the past five years, the top 10 players by WAR have received about 67 percent of all MVP points, on average, while the top five have checked in at about 48 percent. In the first five years of this sample, those figures sat significantly lower, at 45 percent and 25 percent, respectively. (In the earliest years of BBWAA MVP voting, which began in 1931, voters actually matched up with WAR more closely than they did from the ’70s through the ’90s, probably because the pre-expansion player pool was small enough that the best players more easily stood out.)

In other words, BBWAA awards voters have been getting better, if by “better” we mean “more likely to vote for players with higher WARs.” Considering the less telling stats that many writers used to rely on, it’s not a stretch to express it that way; WAR can’t capture everything that happens on the field, and it needn’t be the sole criterion by which one appraises players, but it is the most accurate single measure we have. If each league had a most-qualified candidate whose WAR was way higher than that of the next name on the list, we could conclude, based on recent voting patterns, that the writers would be more likely to pick him than they would have been at any point in the past.

This year, though, those clearly most-qualified candidates don’t exist. Judge, the AL leader, tops second-place Sale by hundredths of a point of FanGraphs WAR, with José Altuve and Kluber not far behind. In the NL, Kris Bryant leads Anthony Rendon and Joey Votto by less than a tenth of a point, with three more players (Charlie Blackmon, Giancarlo Stanton, and Scherzer) within one win. Granted, there’s more than one way to calculate WAR, and the Baseball-Reference and Baseball Prospectus flavors often differ from FanGraphs’, but whichever method you choose, the top of the leaderboard is abnormally bunched up.

The graph below, built with data supplied by FanGraphs’ Jeff Zimmerman, shows the combined number of players within 1.0 or 1.5 FanGraphs WAR of their league leader (including the leader) in each season since BBWAA MVP voting began in 1931. The blue bars represent the number of players within 1.0 WAR of a league leader—close enough that the difference is likely within WAR’s margin of error—while the combined blue and red bars encompass every player within 1.5 WAR of the leader.

The 2017 season’s total of 10 players within 1.0 WAR of a league leader would be tied with 2010’s for the second most during this span, behind only 1988’s 11. This year marks the first time in the BBWAA voting era that both leagues have had at least four players within 1.0 WAR of the league leader. And the headache for voters goes even deeper than that: This season’s total of 18 players within 1.5 WAR of a league leader—a group that also includes Mike Trout in the AL and [deep breath] Tommy Pham, Paul Goldschmidt, Corey Seager, Nolan Arenado, Justin Turner, Stephen Strasburg, and Zack Cozart in the NL—is tied with 1988 for the most of all time. In 1988, 15 of those 18 were NL players; Andy Van Slyke, the nominal NL WAR leader, finished fourth in MVP voting behind Kirk Gibson (third in WAR), Darryl Strawberry (tied for 10th), and Kevin McReynolds (tied for 24th). This year is a little less lopsided than that, so there are no easy answers in either league.

Just as more and more BBWAA voters have come to view WAR as a good guide to value—or at least, have come to value players the same way WAR does—WAR has ceased to be of much use in identifying the most deserving MVPs. Statistically speaking, there simply aren’t any obvious MVP favorites this season.

With WAR declining to wholeheartedly endorse any candidate, one might fairly default to the team-dependent factors that sabermetrically minded analysts tend to bristle at when they’re used to justify a vote that goes against the stats. But even that barely helps here. Of the 18 players in question, 13 play for presumptive playoff teams.

Even if you eliminate the members of non-playoff teams—the amazing-as-always Trout and Votto, the literally-out-of-left-field Pham, the riveting, historically hot-streaking Stanton, who’s giving Babe Ruth and Roger Maris a run for their money (and whom my colleague Michael Baumann reasonably selected in his awards column this week), and Cozart—that leaves a lot of statistically indistinguishable options. How do you decide between Bryant, the reigning NL MVP and best player on a division champion; Rendon, the perennially underrated, excellent-at-everything engine of the NL East–winning Nationals; and Blackmon, the late-blooming star who’s anchored the Rockies’ otherwise weak lineup and, in defiance of history, driven in 100 runs despite batting almost exclusively out of the leadoff spot? And even if you exclude Kluber and Sale from the AL MVP race—on the dubious grounds that there’s already an award exclusively for pitchers, so they shouldn’t get greedy—how do you decide between Judge and Altuve?

Clutchness is one way; Judge hasn’t hit well in high-leverage situations. But if the BBWAA doesn’t draw your name, and your editor doesn’t make you declare one way or another (thanks, Mallory Rubin), you don’t have to decide. That may be kind of a copout, since someone has to win. But because neither Judge nor Altuve clearly deserves to beat the other, it’s also true that neither clearly deserves to lose. Would a tie be too much to ask? It has happened before. And regardless of which one wins, the fact that they’re so evenly matched is one of the most beautiful things about baseball.

In some ways, Altuve and Judge are the rich man’s Juan Pierre and Adam Dunn. Earlier this decade, Pierre and Dunn called it quits after 14-year careers. Stylistically, at least, the onetime teammates represented the epitomes of dramatically disparate styles—Pierre the slap-hitting speedster, and Dunn the sluggish slugger, bigger by eight inches and 105 pounds. Pierre played a premium position, center field, while Dunn played the positions where he could do the least damage defensively—left field, first base, and DH. Pierre, who batted .295 lifetime, occasionally led the league in hits, sacrifice bunts, and steals and times caught stealing; Dunn, who batted .237, led in walks and strikeouts. Pierre hit 18 homers; Dunn, who hit 19 in his partial rookie year, finished with 462.

It would be hard to design two players who were less alike. Yet in a kind of cosmic joke, the pair produced exactly the same value. At Baseball-Reference, the pages of Pierre and Dunn display identical career wins above replacement totals, down to the first decimal point. In 2015, I argued that they were the perfect examples of why we need WAR; without an all-inclusive framework that could objectively weigh the impact of playing time, defense, baserunning, and divergent approaches to generating runs at the plate, it was too easy to be swayed disproportionately by certain stats. During the portions of their careers that predated widely used win-value metrics, Dunn and Pierre turned into totems for fans who subscribed to certain styles of play, but in retrospect, we can weight their contributions and find that they balanced the scales.

That’s where we are with Altuve and Judge, who are worth much more than Pierre and Dunn ever were but arrive at their WAR scores via similarly roundabout routes. Altuve is about to win his third batting title and lead his league in hits for the fourth consecutive season. Judge is batting .280-something, but he’s also leading his league in home runs, having broken the rookie record. Judge walks twice as often as Altuve; Altuve strikes out significantly less than half as often as Judge. Altuve steals bases; Judge risks stealing only once in a while. Altuve plays middle infield; Judge plays corner outfield. And, most salient of all, Judge is a giant, the tallest position player in the sport, while Altuve is tiny, tied for the shortest. Judge used to be doubted because he’s so big; Altuve used to be doubted because he’s so small. They’re both the best player in the American League, at least when Mike Trout misses several weeks.

I normally trust technology, but I have a crisis of confidence when Statcast says that these pitches are almost exactly the same height off the ground:

And when we see Altuve and Judge juxtaposed, as we did for the first time in a head-to-head (or head-to-shoulder) series in May, we go through the three stages of acceptance.

First, we come to terms with the fact that these humanoids have the same number of limbs, spikes, and jockstraps, and must therefore belong to the same species and sex. Then, we talk ourselves into believing that they play the same sport, in the same highest-level league. Lastly—and this is the hardest—we realize that they are both better than the hundreds of elite athletes whose bodies would be between them if we ordered all MLB players by size.

Aaron Judge, Jose Altuve, and Salvador Perez Photo by Brace Hemmelgarn/Minnesota Twins/Getty Images

Judge and Altuve are almost caricatures of the cliché that baseball is the sport that permits stars of all sizes. We get it, gods of baseball—you could have proved the point with, say, Bryant and Dustin Pedroia. Judge and Altuve are overkill.

Together, the two straddle baseball’s current crossroads. Altuve, who hits for plenty of power himself, is still in some respects a throwback to the sport’s put-it-in-play past. Judge is a herald of the three true outcomes, the embodiment of baseball’s highest-ever home run rate, and the perfect advertisement for exit velocity. It makes no sense that they’ve overlapped like this, but there they are. One of them will probably be MVP, in this year of wonderful, impossible awards choices. Just please don’t make me pick.