The Fascinating Race for the NL MVP
On Monday, I talked a bit about the AL MVP race and how Chris Sale should pretty clearly be the frontrunner at this point. Over in the National League, though, it’s not really possible to write a completely honest piece advocating for just one single candidate, because as of mid-August, there are nearly 10 guys with legitimate cases for the award.
The likely leader in the clubhouse right now is Paul Goldschmidt. The case for him is pretty easy to make: he’s the best player on a team that wouldn’t be in the postseason race without him, and no one is clearly outplaying him. He’s fifth in batting average, third in on-base percentage, and sixth in slugging, and for those who are still into those kinds of things, third in RBIs. He leads the NL in WAR, so he’s not a Juan Gonzalez-type of traditional candidate who gets exposed by looking deeper. The Diamondbacks are pretty likely to make the Wild Card game, so he’ll probably get the playoff-team boost. Overall, he’s got something to offer pretty much every kind of voter and would almost certainly be the favorite to win the award if it were held today.
But while Goldschmidt probably would have the most diverse appeal among voters, and would almost certainly finish in the top two or three on the most ballots, he might not be the top choice for as many voters as you’d think. For one, Giancarlo Stanton is going to hit 50 home runs and might end up at 60, so for those who just want the best slugger, there’s a huge gap between Goldschmidt and the NL’s premier home-run hitter.
(Photo: Corn Farmer)
For voters who prefer RBI as their go-to offensive metric — and there are still some around — Nolan Arenado is an easy sell, as he leads the league in that category and plays a spectacular third base. His team is also a surprise contender who wouldn’t be one without him, and his batting average and home-run totals are nearly the equal of Goldschmidt’s, so for a voter who wants to stick to how MVPs were picked back in the day, Arenado looks every bit Goldschmidt’s equal on offense and offers more value in the field.
But both of those guys are playing for Wild Card teams, and Stanton’s team is out of the race entirely, so if any voters decided to limit their pool for the top spot to players only guaranteed a playoff spot — since the voting has to be turned in before the Wild Card game is played, perhaps such a voter would decide that a Wild Card-losing team didn’t actually make “the playoffs” — they’d still have a cornucopia of deserving candidates from which to pick.
Corey Seager and Justin Turner have been two of the main reasons the Dodgers might challenge the all-time record for wins in a season. With Bryce Harper’s injury, Anthony Rendon — who is basically Nolan Arenado but playing at sea level — might appeal to someone who wanted to vote for a Nationals position player. Of course, you could make a case that Max Scherzer has actually been Washington’s best player this year, and if you just go by runs allowed, it’s Scherzer who leads the NL in WAR this year.
Waters still not muddy enough for you?
Kris Bryant, last year’s winner, hit .292/.385/.554 and got 29 of the 30 first-place votes in 2016. This year, Kris Bryant is hitting .293/.405/.535, and as with Goldschmidt and Arenado, you can make the case that the Cubs wouldn’t make the playoffs without him — unlike last year, that is, when he had plenty of help from his supporting cast. The drivers of his big drop in WAR are fielding and baserunning, which are the variables that voters tend to trust the least. If you voted mostly for Bryant’s bat last year, it doesn’t look that different this year.
Biased towards up the middle hitters but not that impressed with Seager’s 19 home runs? Or want to discount him for playing on a team with too many good players? I’d like to introduce you to Charlie Blackmon, a center fielder hitting .333/.392/.611. Even after adjusting for Coors Field, Blackmon’s been a monster at the plate and ranks up near all the other non-Goldschmidt hitters in WAR. Of all the players playing an up-the-middle position this year, Blackmon has been the best hitter, so anyone who simply wants to adjust for defense and find a guy who hits like a first baseman while playing a premium spot, Blackmon is an appealing option.
And somehow, after naming nine legitimate candidates for the MVP award, we still haven’t talked about the league’s best hitter yet. That would be Joey Votto. Again. Votto is tied with Turner for the league lead with a 166 wRC+, but he’s done it in 135 more plate appearances. Anything you like about Goldschmidt’s bat, Votto can match: his .316/.448/.600 line is a little better than Goldy’s .320/.434/.600 line, and given Votto’s current run and his recent history of second-half ridiculousness, it wouldn’t be surprising if he opened that gap up a bit more by season’s end.
That’s 10 names with which you could easily fill a ballot, and really, all of them have at least some kind of argument for the top spot. Some arguments are stronger than others, but still, this is a remarkably crowded field. The list will undoubtedly narrow itself further — Harper and Clayton Kershaw already fell out of contention due to injuries, or else this would be even more ridiculous — as we get closer to when the votes are due, but I don’t envy voters who got handed an NL MVP ballot this year. My NL Rookie of the Year pick is going to be quite a bit easier.
Dave is the Managing Editor of FanGraphs.
What IS up with Votto’s 1st/2nd half splits? I wanted to say it’s a crime against humanity that he didn’t make the All-Star team each of the past two years, but then you look at his 1st-half numbers and it’s suddenly not that absurd.
Every winter NL pitchers spend furious energy figuring out new ways to get him out. Every Spring it works, a bit, for a little while. But the more Votto sees of them, the less they are able to fool him. If the season was 500 games long you might see him OBPing 1.000 by the end.
If memory serves he ran like a sub .200 BABIP in April.
I’ve had this theory on Votto for a few years now. I think he is intentionally manipulating his heat map and scouting report in the first half, which allows him to really benefit in the second half. Hitting a few weak grounders in his hot zones, swinging at pitches in areas he doesn’t normally swing at, all those things to fool pitchers into thinking he has weaknesses that aren’t actually there.
He’s Canadian. He’s distracted until June by the Stanley Cup.
As a reds fan, I think, like myself, he’s bored by the all-star break as the Reds are now out of contention. By a lot. So he decides to entertain himself by playing at a ridiculous level.
Always include the previous year’s 2nd half stats in ASG voting. 2016’s trash(no credit for good 2nd half) is 2017’s treasure(hate voting on SSS).
Blackmon, Jose Ramirez, Mookie Betts, and Kluber are all going to exceed 5.5 War this year. Should I go to Vegas and bet the house on Max Schrock 2019 MVP?
I was wandering through Josh Vitters’s page and wound up across an old Fringe Five post where he was included in the Next Five (though he should’ve probably been ineligible as a top-five draft pick). The actual Fringe Five–Chad Bettis, Mookie Betts, Danny Salazar, Marcus Semien, and Danny Winkler–have accumulated more career WAR than Baseball America’s actual top five prospects from that year.
(This would be a more fun fact if two of the BA top five hadn’t died in senseless drunk-driving/boating accidents.)
Speaking of that NL Rookie of the Year…I wouldn’t be surprised if he got more MVP consideration from the voters than the two teammates you mentioned.
Was thinking the same thing. I know the voters are more sabermetric than they used to be, but RBI still sell well and Bellinger is likely going to be the only player on a 115 win team to exceed 100 RBI. Throw in the 40-45 HR and he’ll get some MVP love.
The Dodgers are also like almost an .750 team when he plays.
They were 9-11 when he was promoted. They’re 75-23 with him.
Throw in the fact that he’s a 20 year old rookie phenom who has been breaking records and has generally been the flashiest, most-talked about player on potentially the best regular season team in MLB history.
Still hard for me to see an argument for anyone but Goldschmidt. He’s has a clear lead in WAR (5.5 to Rendon’s 5.2), and would still be first even if you tossed out or regressed the UZR portion. He has the clear lead in offensive-runs-above-average (44.0 to Votto’s 36.7). He’s second in both RBI’s and runs scored, behind only two Coors-inflated Rockies who don’t crack the top-5 in any park-adjusted categories.
Of course, everyone’s numbers will change over the next month and a half, and all these arguments could be pointless by then.
0.3 WAR is not close to a clear lead.
That’s kind of what I’m thinking too. With Harper out and not likely to post the counting totals that voters love, Goldschmidt’s only competition is Rendon and Rendon is precisely the kind of player that the BBWAA has never given much love – the all-around good player who’s not elite at anything.
As much as this site tends to ignore it, MVPs tend to go to sluggers from good teams who post big RBI totals with the occasional leadoff hitter (Pedroia, Rollins) and starting pitcher thrown in.
“the all-around good player who’s not elite at anything”…yet he is ELITE.
Rendon hits for average and power and is an on base machine, but doesn’t play at Coors Field – Lite like Godschmidt. Not to mention Rendon plays a tougher defensive position and does so at an ELITE level. I guess a guy who does a bunch of things at 60-70 on the 20-80 scale isn’t elite to you though.
Not sure of the reason for the down-votes; I’m just offering my opinion. And I wasn’t trying to say the race isn’t close. I’m saying that regardless of what measurements you want to use, it’s hard to come up with an overall measurement system where someone is *ahead* of Goldschmidt (unless you ignore park factors or only look at rate states).
The down-votes, I think, are for the reason stated by the guy with all the up-votes.
I didn’t downvote you, but I think the reason is that you are going a little too far. In the article, Dave says that Goldschmidt is the leader in the clubhouse (which is a terrible analogy if I’m going to be pedant). Then you write a comment as if Dave made a mistake in his analysis. However, you don’t address Dave’s many points where he talks about who voters might vote for depending on what they are weighing heavily. Your assertion that it will always come back to Goldschmidt is wrong for the many reasons Dave talks about in the article.
If Stanton hits 62 homers, he’ll win.
Otherwise, it should be Goldie.
If the season were to finish today I would vote for Anthony Rendon. The ultimate all rounder.
I think Goldschmidt is kind of an obvious choice for a lot of reasons, since he (a) is on a team that would be totally dead without him, (b) he ranks well in basically every traditional metric, (c) he leads in WAR.
The other guys…
Votto and Stanton should get considered if they keep up their torrid pace but they won’t get any love because their teams are bad.
It’s weird that Dave mentions Rendon stepping up in Harper’s absence, since Harper actually has more PAs than Rendon. Rendon leads in WAR, but if we’re considering the likes of Nolan Arenado I think Harper is going to get some support. Maybe because Harper’s injury is more recent people will forget about him, but he’s also a brand.
Arenado and Blackmon are good players, but no one believes anything in Colorado.
Scherzer has only won 12 games this year, and he’s not Chris Sale. No way.
And then, let’s not forget Tommy Pham, who the Cardinals kept in the minors to start the year but has been putting up about 1 WAR per every 100 PAs. Selfishly, I want him to win the award to help burnish his future hall of fame candidacy (or at least the Hall of Awesome candidacy, where we can plunk his bust next to Jim Abbott).
“no one believes anything in Colorado”…you’re right, those stoners out there couldn’t possibly be playing actual baseball.
If you go by bpWAR(or pWAR), then Stanton comes in 1st with 6.6 and Scherzer comes in at 2nd with 6.0.
http://www.baseballprospectus.com/sortable/index.php?cid=1918875
Interestingly, Scherzer leads in rWAR with 6.6 while Goldschmidt is at 5.8.
One of my favorite parts of WAR is that different sites have different methods. I know that traditional baseball fans hate this, but I think it promotes competition aka fun.
While it should not necessarily be a consideration in this year’s MVP voting, I think it would cement Goldschmidt’s legacy as one of the great players of this decade.
NL MVPs this decade:
2010: Votto
2011: Braun
2012: Posey
2013: McCutchen
2014: Kershaw
2015: Harper
2016: Bryant
As a Diamondbacks fan, I think it would be nice to see Goldschmidt recognized along with these players; particularly considering that the team just isn’t very good often, so it could be the last chance for him to have a great season on a playoff team.
This is going to be the most interesting award race for sure. But there’s a lot more discussion here of the good arguments than there is of the voters, and their usual prejudices. What if we try to model those?
My sense is the Dodgers are going to be one of those teams that suffers worst from a split vote — there are just too many guys to choose and no obvious headliner — so the (likely) best team for at least a decade or so is going to get almost no MVP consideration. Turner and maybe Bellinger will get a scattering of votes but I’d bet Seager will be overlooked almost entirely.
The voters like big-name star players. But Harper has missed too much time to rack up superior counting stats.
The star who’s going to end up with some of the glitziest counting stats is Stanton, but he suffers from the non-contending-team bias. Hitting the most dingers by itself isn’t usually enough to make up for that.
The voters like star players on wild-card teams the best of all, because it’s easiest to tell the story where “valuable” means “reason the team is in the playoffs.” This makes Goldschmidt and Arenado the probable frontrunners in my opinion. I’m guessing the huge majority of first-place votes go to the two of them. And it’s just a hunch, but I do think it’s going to be Goldschmidt’s year — every broadcaster and pundit I pay attention to has made some little speech about how unfortunate it is that he’s “overlooked” at some point this season already.
Such a bummer Freeman got hurt.
I agree that Goldschmidt is the current frontrunner.
However, there is one legitimate argument against him.
He only ranks 9th in WPA.
Why is that? His Clutch score is -1.33 which is fifth lowest.
(Incidentally, we have Bryant at #1 and Stanton at #4)
If we look deeper, Goldschmidt has wrc+ of 168 and 170 for low/med
leverage but wrc+ of 57(!) in high leverage situations.
#NoClutchGene
I don’t know. Do you really think the MVP voters are digging down to WPA to decide things? Really? I mean, half the voters just discovered WAR in the last 3-4 years and probably have no idea what WPA is (which is ironic because voters of that ilk would LOVE WPA).
I doubt they’re digging down to WPA, but the idea of “clutch” is a fairly oldschool, meatball thing that could certainly influence voters. You don’t need to know which metric best applies to have a general sense that a guy seems to wilt in big spots.
What kills me about this is that this is the BEST case — it’s literally the one place in arguing about baseball where it’s the most defensible to base an argument on WPA instead of other metrics! Because it’s an award given retrospectively for performance over the season, it’s completely reasonable not to care about the predictive or skill-based attributes of the other metrics at all. That they don’t do it is how you know that the guys who should be basing their arguments on WPA mostly don’t really believe in empirical argument or quantification in the first place.
I am really skeptical of the specific “clutch” scores cited here. It just seems like the least clutch players are always ones near the top of the WAR leaderboards, which kind of makes me wonder how it is calculated and if there isn’t something weird going on.
On the flip side, there are some excellent players who appear to be “clutch” this year (Mookie Betts, Nolan Arenado, George Springer) but you also see some of the worst hitters this year (Dansby Swanson, Albert Pujols, Alcides Escobar).
I think the idea of a clutch score is really good, but it’s not clear why it’s correlated with wRC+ in such an unexpected way. It suggests there’s a correlated error in the formula used to calculate it.
How is it calculated?
If it’s how much better/worse you do than your average performance in “clutch” situations, then elite hitters have a lot more to live up to than a guy like Escobar.
Yes, that’s correct. It’s just a record of how each hitter did in high-leverage situations compared to how he did in all situations. It ought to be expected to fluctuate randomly, there’s essentially no predictive skill involved. But that of course doesn’t mean that it’s invalid as a retrospective description of what already occurred, if you happen to want one.
http://www.fangraphs.com/library/misc/clutch/
Yeah, see, if that’s the case then it’s not all that useful as a statistic. If you’re terrible at the plate in low leverage situations and slightly less terrible in high leverage situations and benefiting from a low denominator, I don’t know how that helps us. You’re still terrible.
You can BE “terrible” in the sense of not having a special “clutch” skill that doesn’t exist, and at the same time you can still HAVE BEEN “good in the clutch” in a certain set of high-leverage situations that already happened. (See, for example, Pujols this year.) That it’s not a skill doesn’t mean it’s not useful as a retrospective description of things that have already happened. People are totally within their rights to think that performance in high-leverage situations deserves to be weighted heavily in something like MVP voting.
…but “Clutch” doesn’t actually TELL you anything about “performance in high-leverage situations.”
So…if a .300/.400/.500 hitter hits .280/.400/.480 in high leverage situations, “Clutch” will tell us he’s worse than a .200/.250/.300 hitter who hits .250/.300/.350 in high-leverage situations?
I don’t want to say that’s a useless joke of a stat and anyone who uses it to evaluate a player should be roundly mocked…so I hope that my characterization is incorrect.
WPA is silly. A homer in the first counts just as much in a 2-1 game as a walkoff.
He’s 2nd in RE24, and is actually a run closer to Votto than he is in wRAA.
Feels like baserunning puts (or ought to put) Goldschmidt over the top. He’s as good a hitter as Stanton, plus the 5th best base runner in the NL.
My argument for Anthony Rendon as MVP is essentially the same as it was in 2014. He deserved it then and he deserves it now. Give the man his due!
I don’t believe we have enough data to accurately account for a player’s home park in a given season. That being said, most would agree that Goldschmidt plays in one of the best hitter’s parks in baseball. Stanton plays in one of the worst. What’s more is that the NL East doesn’t offer much help for road games. Only Philly is what someone would consider a hitter’s park so that’s 104 games in neutral or pitcher’s parks for Stanton. Goldschmidt plays 82 games in his home bandbox and another 8 a year in Coors Field, so that’s 90 games in arguably the two best hitter’s parks in baseball. I’m not sure it’s enough to give it to Stanton over Goldschmidt, but I believe it’s an overlooked factor and if Rendon or Stanton have a hot last month should be considered.
…so use park-adjusted stats?
“For voters who prefer RBI as their go-to offensive metric — and there are still some around…”
its never a “go-to”, however people with brains/playing experience know it takes onions to drive a man home when it really counts. To not hit into the double play, to not pop out. To be able to hit the sacrifice fly just about every time, not many guys can do this.
Thank God I didn’t read Yasmani Grandal’s name, I totally expected to.
Re: “and for those who are still into those kinds of things, third in RBIs.” Can baseball be utterly reduced to mathematics and science? Maybe not always. Some peons like me still believe that “clutch hitters” exist–players who don’t freak out or scare or re-adjust too much when things get dramatic and dicey, and the crowd gets involved. Some players seem to THRIVE in those scenarios and some don’t. Is that possible? Is it possible that some of these players are different from one another even though the numbers can begin to look alike? For example: 1. With runners on, don’t some hitters make adjustments so as to just try to make contact vs. trying to hit homers? If so, wouldn’t it be possible that these different approaches cause different results? 2. If all scenarios for driving home teammates can be called equal, why not call all home runs equal? All groundballs equal? The Stathead’s disdain for RBIs is starting to sound like mere wine-snobbery.
“Is that possible?”
Yes, in a counterfactual sense, it’s certainly imaginable that such a thing could be true. It’s just, you know, NOT true, in the sense that it’s ever been rigorously shown that this purported skill was distinguishable from chance. And that’s not from lack of trying. You’re asking questions that have long been answered, based on empirical evidence, as if they were deep. It’s like saying it’s “wine-snobbery” to keep saying the earth is round. That is, dumb.
Now if you want a retrospective account of who did more “clutch” things in a given season to be part of the argument over the MVP award, that’s perfectly fine — and if you actually want an empirically grounded quantification of them, that’s exactly what WPA and WPA/LI are for. But they’re not on baseball cards, so they’re probably not going to satisfy your real, unstated motivation here.
Wow! You have proof that there is no difference, no impact at all, when a player approaches an at-bat with the desire to make contact vs. a desire to hit a home-run? Wow! Please do share it. After reading your proof–your evidence that this question has “long been answered,” I will apologize to you for being dumb and praise you for being so super-duper smart. But, to be fair, if you fail to provide such proof, I will…not be surprised. LOL.
“Some players seem to THRIVE in those scenarios and some don’t.”
The key word in your sentence is ‘seem’. It’s all in your head.
I’m still waiting for someone to explain to me why a “clutch hitter” isn’t someone who can’t be bothered to give full effort unless it’s a showboat situation…
Ha. Do I think players approach at-bats differently when a game is very close vs. when a game is a blow-out? Yes, I do. But I can’t prove it–that is correct. But you can’t prove why “showboating” isn’t just a theory in your head either. Nice try!
Re: “and for those who are still into those kinds of things, third in RBIs.” I think even very smart people who are very confident in their views should try to be open-minded to other reasonable theories. Some of the best scientists have been open-minded rather than online know-it-alls. The theory that clutch hitting is a myth is a theory, not a universal, incontestable truth. Here are 2 reasonable arguments that suggest that the theory of clutch hitting (which can perhaps be connected to RBIs) is not necessarily all bunk: http://www.foxsports.com/mlb/just-a-bit-outside/story/clutch-hitting-swing-difference-081414 [ and: http://sabr.org/research/clutch-hitting-major-leagues-psychological-perspective ]