The Fascinating and Still Unsettled NL MVP Race

With five days remaining in the 2021 regular season, it’s abundantly clear that there won’t be much clarity offered in the National League Most Valuable Player race. Yes, Bryce Harper’s Phillies still have a mathematical shot at a postseason spot per our Playoff Odds, unlike Fernando Tatis Jr.’s Padres and Juan Soto’s Nationals, but not everybody is of the belief that an MVP needs to hail from a postseason-bound team or even a contender.

From a practical standpoint, it’s usually the case that an MVP does hail from such a team; in the Wild Card era (1995 onward), 42 of 52 (80.8%) have done so. The tendency shows an upward trend, the degree of which depends upon where one sets the cutoff. For example, three out of 18 MVPs from 1995-2003 missed the postseason, and likewise three of 18 from 2004-12, but four of 16 from 2013 onward; it’s just as accurate to say that from 1995-2004, four of 20 missed the playoffs, dipping to two of 20 from 2005-14 and then four of 12 since. Either way, all-time greats Larry Walker (1997), Barry Bonds (2001 and ’04), Albert Pujols (2008), Alex Rodriguez (2003) and Mike Trout (2016 and ’19) account for the vast majority of those exceptions, with Ryan Howard (2006), Harper (2015), and Giancarlo Stanton (2017) rounding out the group. That Rodriguez, Stanton, and Trout have doubled the all-time total of MVPs who have won while hailing from sub-.500 teams — a list that previously included only Ernie Banks (1958 and ’59), Andre Dawson (1987), and Cal Ripken Jr. (1991) — is perhaps the more notable trend, with Shohei Ohtani likely to increase that count this year. Effectively, that’s a green light for Soto’s late entry into the race, and also worth pointing out with regards to Tatis, as the Padres slipped to 78-79 with Tuesday night’s loss to the Dodgers.

From a practical standpoint, it’s also true that the notion of value is extensively tied to the things that can be measured via Wins Above Replacement. As old friend Eno Sarris noted at The Athletic (in an article on the value of Ohtani’s roster spot that’s well worth a read), in the past 14 years, only two MVP winners were not in their league’s top three by FanGraphs’ WAR, namely Jimmy Rollins in 2007, and Justin Verlander in ’11.

All of this is a preamble to my guided tour through the NL race (the format of which owes a debt to Craig Edwards’ 2020 treatment) in an effort to figure out for whom I’d cast my ballot, if I had one. While I’m a member of the BBWAA, as part of the New York chapter I’m lower on the depth chart for that honor than Ohtani is on the Angels’ list of emergency catchers. While I’ve opined as to whom I might vote for at various junctures via my weekly chats, I come to this exercise with no particular predilection towards the cases of any of the aforementioned trio, or any other candidate. It’s a long season, and I’ve kept an open mind as the shape of the race has continued to evolve.

Here, for example, is the NL WAR top 10 through the end of May:

Despite serving stints on the injured list due to his left shoulder (April 6-16) and a COVID-19 infection (May 11-19) and therefore falling below the plate appearance threshold to qualify for the batting title, Tatis had already established a presence in the race thanks to his offense. It didn’t hurt that the Padres were just half a game out of first place in the NL West, having gone 8-1 during his second absence as part of a 16-3 roll that included 10 of the 14 days during which the Padres had a share of the division lead. Harper, whose total of 38 games to that juncture matched that of Tatis, had missed time for a variety of ailments to that point, most notably a beaning-related bruised left forearm that sent him to the IL on May 25; to this point he’d hit for a 135 wRC+ with 1.1 WAR. Soto, who missed 10 games in late April due to a left shoulder strain, was puttering along with a meager .392 SLG, 110 wRC+, and 0.4 WAR.

On July 10, Acuña — then the NL WAR leader — tore his right ACL, upending the race just as the All-Star break arrived:

Harper’s trajectory hadn’t changed much, as he was hitting for a 137 WRC+ with 2.0 WAR, but Soto had improved to a .445 SLG, 128 wRC+, and 1.8 WAR. Finally unimpeded by injuries significant enough to sideline them for multiple games, both soon caught fire; from the start of the second half through the end of August, Harper hit for an NL-best 207 wRC+ (.348/.486/.723) with an NL-high 2.9 WAR, while Soto ranked second in both categories with marks of 196 and 2.5. Tatis, who missed the first half of August due to his balky shoulder and scuffled upon returning as an outfielder, in an interesting but short-lived experiment, still hit for a 162 wRC+ with 1.4 WAR in this span. He did that despite playing in just 28 games, 15 fewer than Harper and 13 fewer than Soto. That was enough to keep him in the WAR lead, with the other two now in the top 10:

At that point, the Padres were 72-62, a half game ahead of the Reds for the second NL Wild Card spot. The Phillies, who had been just 51-51 at the trade deadline were up to 68-64, two games behind the Braves in the NL East race. The Nationals were burnt toast, having traded Turner — who had generally been overlooked in the discussion despite his consistent presence on the WAR leaderboard — and Max Scherzer to the Dodgers on July 30. To these eyes, even with Harper closing the gap, Tatis appeared to be the favorite heading into the final month.

September has been Soto’s month. Through Monday (the cutoff for all the stats that follow, making this a snapshot in time rather than a definitive end-of-season roundup), the 22-year-old has put up video-game numbers this month, hitting .430/.590/.744 for a 239 wRC+ and 2.4 WAR, with a 27.6% walk rate and 6.6% strikeout rate. Harper has actually outslugged him (.346/.495/.756, 214 wRC+, 1.7 WAR) while helping to keep the Phillies alive. Turner has acquitted himself well (156 wRC+, 1.3 WAR), while Tatis has been fine (137 wRC+, 0.8 WAR), but even with Frazier and Machado hitting at a similar level, they’ve been unable to drag the Padres back to relevance.

So here’s the full-season leaderboard as of Monday:

And here’s a breakdown that includes multiple flavors of WAR and defensive metrics, with nods to win probability as well:

In considering the two flavors of WAR alongside straight WPA, Context Neutral Wins (WPA/LI), Soto, Harper, and Tatis do stand out above the pack, though it’s not unreasonable to consider elevating Crawford on the basis of his OAA and WPA-flavored stats as well as Goldschmidt via WPA. Throw in Turner, the race’s Steady Eddie, and I think we have a reasonable top six, with apologies to Reynolds and the rest.

Distilling each of these candidates’ pros and cons to 150 words, give or take:

Juan Soto

Soto leads in both flavors of WAR — narrowly in FanGraphs’ version, and by a wider margin in Baseball Reference’s version — thanks in part to much-improved defense in his first full season as a right fielder. He’s in the midst of an historic season, posting the highest on-base percentage by a player 22 or younger in a full post-integration season; the only youngsters with higher OBPs are Ted Williams (.553 in 1941, in case you thought his .406 batting average wasn’t impressive enough), Joe Kelley (.502 in 1894), and Soto himself in the short season last year (.471).

The knock on Soto’s candidacy is that he wasn’t all that valuable in the first half, and his early shortcomings were part of the reason the Nationals wound up having a fire sale in late July and becoming even more irrelevant over the season’s final two months. His Championship WPA — the extent to which his offense helped his team towards a World Series win — is meager compared to the guys whose teams were in the hunt for longer, not that he had much control over that. He’s hit .369/.542/.682 since the All-Star break, and some might view that as padding his stats while very little was at stake.

Bryce Harper

For those who believe that an MVP should at least be doing the heavy lifting on a contending team, Harper’s the most obvious pick. Since the trade deadline, he’s hit .337/.474/.751 (207 wRC+) with 18 homers in 54 games while the Phillies have gone 30-24 to make a run at the NL East flag, though Tuesday night’s loss may have been the mortal blow, knocking the team’s NL East odds from 13.0% to 4.1% while their minuscule Wild Card hopes were extinguished. He’s second in fWAR by a hair, and tied for fourth in bWAR, but he’s first in WPA, WPA/LI, and cWPA. He’s the guy coming up big at the right time, and despite Soto’s strong season, it’s Harper with the overall lead in both SLG and wRC+.

Harper’s biggest negative is his defense. By UZR, he’s in the black, which has helped him stay close to Soto but by both DRS and OAA, he’s more than half a win in the red, and via the former, substantially below Soto in the WAR rankings.

Fernando Tatis Jr.

Despite the Padres’ ugly fade, Tatis has a few things going for him. First, he’s been the most productive player on a per-game or per-plate appearance basis; prorated to 600 PA, he’s produced 7.1 fWAR, ahead of Harper’s 6.9 and Soto’s 6.4. Second, he’s spent most of his season at a premium position, playing 97 games at shortstop, another 20 in right field and six in center; while his fielding metrics have been in the red, his overall defensive value including UZR plus the positional adjustment is -4.2 runs, still ahead of Harper (-4.3) and Soto (-6.4). Third, he’s done all of this with a left shoulder issue for which he quite reasonably could have undergone season-ending surgery months ago. Instead he’s played through it, and at an elite level.

For as good as Tatis has been, his absences have carried a cost; Padres shortstops besides Tatis (Cronenworth and Ha-Seong Kim) have produced just 1.3 WAR while making 59 starts, for example. Despite that, it does look a little odd that the Padres are 20-12 in games that Tatis didn’t play compared to 58-66 in games that he did.

Trea Turner

No player has ever won an MVP in the middle of a season in which he was traded, and things are probably going to stay that way, but Turner deserves some appreciation for his seamless transition from Washington (.322/.369/.521, 136 wRC+) to Los Angeles (.322/.373/.519, 140 wRC+), that while mainly playing at second base, a position he hadn’t played at all in five seasons. He’s the MVP race’s true five-tool player, a speed demon who leads the NL with 31 steals while only being caught four times. Among the candidates here, he’s tops in overall defensive value (7.5 runs).

For as talented as he is, Turner feels a bit like a cipher in this race, as he’s rarely been the focus in either Washington or Los Angeles due to the presence of more famous (if not necessarily higher-performing) teammates. He’s got good power, and strong on-base skills, but his offense isn’t in the class of the big three — and it’s not uncommon for voters to over-weight offensive stats in this context.

Brandon Crawford

For those who believe that the NL’s best team should be represented in the MVP discussion, Crawford is the man. His combination of improved plate discipline and a new swing featuring a more open stance has led to his best offensive season at age 34, with substantial contributions in the WPA categories along the way. His defense isn’t what it used to be, though his glove is still an asset; his OAA is a personal best for the 2016-21 period, and even with a more modest UZR, his 7.1 defensive runs (including the positional adjustment) is second among these candidates.

The problem for Crawford is that based on WAR he’s merely the third-best shortstop in the discussion, and one has to apply a significant bonus to his Giants teammates to elevate him past Tatis and Turner. In the long history of MVP voting, that’s been a common paradigm, though in this context it feels a bit retrograde.

Paul Goldschmidt

What’s elevating his season beyond the typically solid Paul Goldschmidt season is timing. He’s hit .333/.405/.631 (174 wRC+) in the second half, including .341/.435/.714 in September, and he’s been a major part of the run that has turned the Cardinals from Wild Card race afterthoughts to champagne-soaked clinchers riding a 17-game winning streak, the NL’s longest since 1935. His 5.17 WPA/LI is good for third in the league behind Harper and Soto. He’s been a difference-maker.

That ought to earn Goldschmidt placement on the ballots but again, unless one is placing a whole lot of emphasis on what his teammates have done — compile the NL’s fourth-best record, not its best — it’s difficult to see what should propel him past the top candidates here.

Corbin Burnes and Zack Wheeler

I’ll confess that I don’t know what to do with pitchers in an MVP discussion except treat them as afterthoughts in an already-crowded field, but Burnes (7.6 fWAR, 5.8 bWAR, 2.29 ERA, 1.56 FIP) and Wheeler (6.9 fWAR, 7.1 bWAR, 2.79 ERA, 2.62 FIP) at least deserve mention here, and probably within the top 10 of any voter’s ballot. Voters have turned to pitchers for MVP awards just twice in this millennium (Verlander in 2011, Kershaw in ’14), so I’m not going to sweat my lack of full consideration here. Somebody else can dig in on the NL Cy Young debate.

All of this is a lot to consider, but having dug through the numbers I still come back to my original trio of Harper, Soto, and Tatis. Going completely context-neutral, my arrow would point to Soto, but I don’t think it’s unreasonable to consider context, particularly in a close race. The fact that Harper’s offense both before and after the All-Star break has been a bit better than Soto’s, in a much more competitive context — even if the Phillies do come up just short — is worth enough that to my mind it tilts the balance. Bryce Harper is my NL MVP pick, but I wouldn’t blame anyone for looking at the same set of numbers and concluding it should be Soto, and I’ll be fascinated to see where the voters land.