Jacob deGrom for NL MVP
Before we get going, allow me to say four things:
- This is not the official FanGraphs position. FanGraphs doesn’t have an official position on any awards. This is a company of a bunch of different writers, and any bunch of writers will possess a bunch of different opinions. These are my thoughts, and my thoughts alone.
- My vote this year is for the AL Cy Young. I do not have a vote for the NL MVP. If I did have a vote for the NL MVP, I wouldn’t be allowed to write this right now! As far as this race is concerned, I’m an outside observer.
- Reasonable people can conclude that Jacob deGrom shouldn’t be the NL MVP. In such an event, I imagine the support would go to Christian Yelich. Yelich has been amazing, especially of late. Every number has error bars, and Yelich has an argument. This case isn’t open and shut.
- You’ve probably read much of my argument before, written by different people in different places. This is the “best player” argument. It’s the Mike Trout argument. I’m just going to make the argument with different words.
So we can get into it, then. Last week, I wasn’t sure who I supported. I’ve never voted for the league MVP, so I’ve never given it all that much thought. But now I’ve come around, and I can say that, if I had a vote for the NL MVP, my first-place choice would be Jacob deGrom. My second-place choice would be Christian Yelich. deGrom, of course, is done for the year, because the Mets were bad. Yelich’s Brewers are playing literally right now, and for all I know, he’ll provide the winning hit that sends the Brewers straight to the NLDS. For so many voters, that’s likely to be a factor. Perhaps that’s likely to be the factor. I don’t believe that it should be. I believe that deGrom made a winning case.
No one’s MVP ballot should ever be based just on a WAR leaderboard. But WAR provides a pretty good starting point. According to our regular version of WAR, deGrom was the best player in the National League. According to our other version of WAR, which holds pitchers responsible for their actual runs allowed, deGrom was the best player in the National League. According to the version of WAR over at Baseball Prospectus, deGrom was the best player in the National League. And you can examine their context-based performances, too — by Win Probability Added, deGrom was the best player in the National League. That stat can be weird and imperfect, but when you’re pitching well in low-scoring games, the WPA can add up pretty quick.
As you know, the Mets didn’t ultimately win all that many of deGrom’s brilliant starts. That’s not at all deGrom’s fault, and it doesn’t make any sense to fault deGrom for the poor performance of his teammates. The MVP is an individual award, given to an individual player. As deGrom’s WPA is able to demonstrate, he consistently gave the Mets a good chance to win. That’s the whole point. deGrom’s teammates couldn’t pull their own weight. So it goes, when you play for a non-contender.
One of the neat things about WAR is that it’s a counting stat. It accounts for performance and playing time. But still, I know this is always a conversation whenever we compare pitchers and hitters. Hitters play almost every day. A starting pitcher goes once every five games. To put this in necessary terms: deGrom, as a pitcher, faced 835 batters. Yelich, as a hitter, batted 647 times. Now, Yelich as a defender in the outfield had something like 270 opportunities. But according to Statcast measurements, Yelich had something like 93 defensive opportunities that weren’t just routine cans of corn. Yelich also gets some credit for all the time he spent running the bases, but I think that, overall, the playing-time argument is unconvincing. deGrom played a lot, too. He played enough to push his WAR into first.
Now feels like the right time to revisit those familiar MVP voting criteria. You can read them here, straight from the BBWAA. I know there’s always a lot of emphasis put on the beginning, the part about how there’s no clear-cut definition of “most valuable.” That’s true! But I want to call your attention to three excerpts. One, the first listed rule:
1. Actual value of a player to his team, that is, strength of offense and defense.
Two, from the first paragraph:
The MVP need not come from a division winner or other playoff qualifier.
Three, the last line:
Keep in mind that all players are eligible for MVP, including pitchers and designated hitters.
You see articles and tweets every year saying the fun of the MVP is how it’s so open to interpretation. I get a kick out of it, too, because awards are basically thought experiments. But, so often, my sense is that these criteria are interpreted incorrectly. To be clear, it’s not like I have access to hidden documents assembled by the original MVP award creators. I didn’t get to be a fly on the wall. I very much wasn’t present, because I very much wasn’t alive. Yet I think what’s there is pretty explicit, in what gets mailed out to voters every fall. The very first listed rule seems to suggest this is just a performance award. There is, also, a general character aspect, but it’s mostly performance. Which teams’ players are eligible? Every team’s players. Which kinds of players are eligible? All kinds of players. If you played on a major-league baseball field, you’re an MVP candidate.
Too often, voters ignore that last point. I understand why: Many people are of the opinion that the Cy Young is for pitchers, and therefore the MVP is for position players. By the actual rules, this is not the case. A voter might want it to be the case, but it is not, according to the criteria itself. And, of course, pitchers have won the MVP in the past. Clayton Kershaw won in 2014. Justin Verlander won in 2011. But pitchers are up against an obvious bias. In last year’s AL MVP voting, Corey Kluber was left off seven of 30 ballots, and Chris Sale was left off 15. In the NL MVP voting, Max Scherzer was left off 18. The year before, Scherzer was left off 18, and Madison Bumgarner was left off 28. The year before that, Kershaw was left off 18. Zack Greinke was left off six, and Jake Arrieta was left off five. Some of the best pitcher seasons in recent history haven’t gotten across-the-board MVP support. I don’t mean for first place. I mean for even just tenth place. MVP voters have ten slots to fill. A chunk of voters still leave pitchers out of the picture. I don’t think there’s a good argument for why they should.
Now let’s go back a point, to the point about the teams. As is said right there, the MVP doesn’t have to come from a playoff team. And we can see this in practice, because last year’s Marlins were terrible, but the MVP went to Giancarlo Stanton. Every year, we have this conversation about Trout. Still, many, many voters place a priority on players playing meaningful games. The closer the race, all the better. Many voters like to try to look for the player who made the difference between a playoff spot and hitting the links. One could extend this to the absurd.
Imagine, if you will, two teams tie for the second wild card. They have to play a tiebreaker, to determine who advances. The game goes into extra innings, and then the road team slips in front. In the bottom half, the home team is down to its final out, with a thin bench. A bench player or September call-up pinch-hits, and then that player hits a dramatic, walk-off home run, turning a deficit into a playoff berth. Technically, that player would have made virtually all the difference. Is that player therefore the league MVP? Of course not. No one would claim so. And yet.
This is all about the interpretation of “value,” with the popular belief that a player provides minimal value when his team is out of the hunt. But if you want to look even closer, why not fold in the player’s salary, too? Is that not a part of the player’s presence on the roster? You can make a good argument that salary ought to be included, as well, but nobody wants that. What if a player is a particularly good box-office draw? There’s no appetite for this. Nobody wants to have to think about these other factors. So keeping it just on the field, it’s true that every team is trying to make the playoffs, but it’s also true that every team is trying to win every game. Every game is of import. Even the Mets’ games in September mattered.
Yelich’s Brewers were trying to win every game, because they’ve been aiming for the playoffs. deGrom’s Mets were trying to win every game, because they’ve been playing for jobs. They’ve been playing for respect. They’ve been playing for the organization; they’ve been playing for the fans. They’ve been playing for themselves, and they’ve been competitive, just as their opponents have been competitive. The goal is always to win as many games as possible. Even in the last week of the season, when all that’s left to decide is draft positioning. All players should be recognized for their performances in competitive baseball games. All regular-season baseball games are competitive baseball games.
To take a step back, to look at the broader picture: What are the MVP results actually telling us? In theory, I know the argument is that the results tell us the collective opinion of 30 professional baseball writers. That’s similar to how it’s supposed to work for the Hall of Fame. But out of the whole pool of BBWAA voters, some writers vote for the best players, period. Some writers place a heavy emphasis on the best players on playoff teams. Some writers ignore pitchers, believing them either ineligible or dramatically inferior. We have all these writers voting, but as things are today, they’re essentially not even all voting for the same award. If everyone has a different opinion of what the MVP is, and if some people leave pitchers out altogether, what question is being answered?
That’s really at the core of all of this, for me. Forget all the familiar arguments about who played for which teams and where they are in the standings, because I want you to forget about the familiar premises. We know what the MVP has turned into. It’s almost hopelessly complicated and inconsistent. Why is there even an MVP award at all? What question is baseball trying to ask of its voters? Does it not make the most sense that baseball would be looking to recognize the single best player in each of its leagues?
Imagine that baseball had no awards. Imagine that you were in charge of creating them. What’s the very first award you create? What’s the most important award you create? It’s the best-player award. Of course it’s the best-player award. It’s not the best-player-on-a-team-that-makes-the-playoffs award. It’s not the best-player-on-a-team-that-makes-the-playoffs-but-only-just-barely award. It’s not the best-player-on-a-team-that-makes-the-playoffs-and-there’s-not-another-really-great-player-on-the-same-roster award. You create an award that looks to recognize the best player. Baseball has such an award. It’s the highest-profile award given by the BBWAA. You can definitely overthink it, and over the course of several decades it’s turned into something extremely overthought. It’s turned into something other than what I have to imagine was the original intent. Sometimes I don’t even know what voters are arguing about anymore. I understand how we got to this place, but we don’t need to stay here.
They say it’s the most-valuable-player award, not simply the best-player award. The most valuable player is the best player, according to the only intent that makes good sense. Even by these simplified criteria, there’s still infinite room to argue. There are statistical error bars everywhere, and it’s not like the conversation becomes less fun. Rather, I think it becomes more fun and engaging, because at least we’d all be trying to answer the same core question. In my opinion, this season, Jacob deGrom was the best overall player in the National League. Therefore, he’d get my vote for the NL MVP. It’s a vote that I don’t have, but it’s a vote I would give him with my full and total confidence.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
Compelling argument, but my also completely non-existent vote goes to Yelich, and DeGrom gets my hypothetical Cy Young vote
I know the MVP is clearly written to include pitchers, and states no penalty, but if you were designing the awards from scratch you would not do one pitchers and one everybody. I can see why people *want* to have the MVP favor position players.
The Hank Aaron Award could fill that void, but gets zero publicity…
That award is just for offense though, right?
Yes. There is also a MLB Defensive Player of the Year award at each position. Also Silver Sluggers and Golden Gloves at each position and per league.
All the components, basically, get their individual awards. The MVP is the best damn player overall.
Most Valuable, not Best player. That would need to be a different award.
Cy Young isn’t for the best player whose main position is pitching. It’s for the guy that pitches the best. The Hank Aaron award isn’t for the best position player, it’s for the guy with the best offense. The Gold Gloves aren’t for the best players who aren’t DH, they are for the guys who defend the best.
So, no, the Cy Young isn’t exactly the pitcher version of the MVP award. The MVP is the only all encompassing award and it does make sense to have such an award. In practice, the pitcher who pitches the best is usually the same as the guy who is the best player whose position is pitcher because pitching is the central role of their job, but, for example, if you were to use only bWAR to chose your awards, Nola would get the Cy Young, while deGrom would get the MVP because his offense was good for a pitcher so his overall WAR is higher while his pitching WAR is lower.
Why do you feel this way? You already got the down votes, might as well just commit to it.
I strongly agree with Jeff and think his argument is 100% correct, but part of the reason why Stanton won it last year was because he fit the traditional mold (lots of dingers) in other ways, in addition to leading all hitters in fWAR.
I know what you’re saying. If you can vote for a guy on a losing team, and you can vote for a pitcher, than there’s no good reason you can’t vote for a pitcher on a losing team. But this isn’t rational. The Mets, despite actually finishing with a non-embarrassing record, are perceived as an embarassment; Yelich finished the year by going nova at the plate, so there’s recency bias. And ultimately, I suspect that most voters just vote for whoever they “feel” should win it, and then make up reasons to justify it later.
i think Yelich’s case is built on a lot more than just recency bias. You don’t lead the league in BA, stoke 36 HR’s and finish one back of the RBI lead , all the while playing sound baseball, without being tremendous throughout the entire season. All seasons are segmented to some degree and his hottest stretch came when it counted most. The fact that he carried his team to the NL Central title, down the stretch, should count and count a lot. While I think defensive metrics are virtually useless, Yelich is not Miggy and he is a good base runner so we have a player who almost won the Triple Crown, carried his club to the top while contributing in all areas. What more does a player have to do?
Had he homered today, he WOULD HAVE won the Triple Crown.
wrong.
Had he homered twice today, he WOULD HAVE won the Triple Crown.
Well, had he homered twice today, he WOULD HAVE been playing for a different team, which would be weird.
I’m a huge Yelich fan, and I don’t think he would have a case with the voters without being a really good ballplayer on some level. And this isn’t like the Trout vs. Cabrera case because in that case, you were comparing two position players to each other (and Trout was clearly way ahead). In this case, Yelich was, hands down, the best position player in the National League this year. He’s the only position player that should probably even be considered in the National League right now for MVP. The gap between him and deGrom is actually less now than the gap between him and Rendon.
That said, I do want to note that he didn’t start off the season all that well, and ended it with a hot streak that was just spectacular. That will help him a lot. And it’s not that he didn’t do “enough” to win it–he led the NL in position player WAR while leading his team to a dramatic division championship! It’s just that deGrom was a better player this year.
I always read your comments and hold your opinion in high regard but i will admit to having a HUGE bias against pitchers for MVP. Your appreciation of Yelich cannot come close to my admiration for deGrom. I have a special spot in my love of baseball for the players who were never highly regarded and became stars. The Rule 5 success stories like Johan Santana and Odubel Herrera, the non-prospects like Goldschmidt and Donaldson and then there is deGrom. For no good reason I watched deGrom’s first game in 2014 and came away with this disbelief in what I saw, A 6’4″ pitcher with a beautiful delivery putting 94+ heaters on the corners and adding in a sharp slider. I have followed his career closely and still wonder if the scouts who were at his games were blind or if he just bent over and found THE MAGIC on the mound at Citi Field that day and never lost it. Even with all that background however, Jacob deGrom does not enter into my MVP thoughts. I can never bring myself around to the the idea that a man who only brings his game to the field 32 times year can compare favorably with the best position player of that year.
But deGrom, when he is on the field those 32 times, is far more important to his team than any other player. So for 1/5th of the games, a starting pitcher impacts the team’s performance in a way no position player possibly can. (I believe Jeff covers this)
That is understood but is that 1/5th of the games enough, and very importantly, he doesn’t even impact those games for more than 75% of each game in most cases? Your argument would carry more weight if starters still pitched complete games and actually influenced the final outcome a lot more than they do today. One of the major reasons deGrom only won 10 games is that he was pulled before the 8th in all but 11 of his 32 starts. The game is 9 innings! Is it good baseball to remove deGrom automatically at a certain point and replace him with one of the myriad selection of stiffs that loiter in the Mets bullpen? I think not but that is another debate. Until proven otherwise I have yet to see evidence that limiting the innings of starting pitchers has protected any arms. If deGrom was on the bump for even 250 innings, much less than Feller or Gibson or Carlton or Ryan threw, at max effort, every year for many years, then he would have a stronger argument. I have followed this game closely for 70 years and understand the value starting pitching but it appears that the game itself is screaming at us that they no longer see that value. Yelich has an opportunity to influence every game and does in every one, whether he fans twice and hits into two double plays or hits for the cycle, steals two bases and makes a great catch. The math says that deGrom only influenced 75% of the 32 games he started, 217 innings out of 288, plus or minus a couple. This is the reason I cannot support any pitcher in today’s game for MVP
“No one’s MVP ballot should ever be based just on a WAR leaderboard.”
When was the last time that the fWAR Leaderboard had someone number 1 that wasn’t worthy of being the MVP? Not as familiar with rWAR, but I assume if fWAR and rWAR were different for a position player, a coin toss or gut feeling would be sufficient to guide who should get vote as both are likely deserving. Pitchers are a pain in the neck and really depends where on the spectrum a voter falls on FIP-style value versus run prevention value.
I went back and looked and I would say 2009 AL hitters. Zobrist had 8.6 WAR on the backs of a 22+ defense that is way out of character with the rest of his career (he only once had even a 10 other than this year).
Mauer wound up with 7.6 despite a wRC+ that was 18 points better than Zobrist in more PA, entirely because fWAR doesn’t do catchers well and penalized him a lot for DHing 25 games. Somehow Mauer wound up with only 4.6 defensive runs.
I think it’s entirely possible that Zobrist’s defense wasn’t really quite as good as he got credit for and Mauer was shortchanged by fWAR because we don’t really know how to calculate catcher defense.
Mauer won MVP anyway, which I think was justified.
I can buy an argument that Mauer may be a better selection as MVP. I doubt Zobrist would have been a bad choice. Voters put Zobrist as 8th. A 1B that hit worse than Zobrist finished second.
I’ve seen little evidence that makes me think the average voter for the MVPs does a better job at assessing value than the fWAR leaderboards…even with the problems with assessing catchers and pitchers. If all the voters had knowledge of WAR and its inadequacies, voters would be better, but at this time, I don’t think that is the case.
Not that I necessarily disagree with you, but if your yardstick is the WAR leaderboards, then of course the WAR leaderboards are going to have the best ballots.
I would say my yardstick is wins. fWAR has been proven to have a good degree of correlation to wins (R squared of about 0.70- 0.75 if memory serves). I’ve seen nothing that suggests writers on average could outperform that correlation.
Just looking at 2009, how does Mark Teixeira finish second (225 points) and Zobrist 8th (34 points). Mauer got 387 vote points. Zobrist and Teixeira hit basically the same not accounting for Zobrist playing in the much harder park to hit. Is Mark Teixeira that much of a better defender at first than Zobrist is everywhere at tougher positions? Is this a sign that maybe a good portion of non-analytical writers are not as good at assessing value as a spreadsheet?
What evidence is there that the writers on average are better at assessing value than the fWAR leaderboard?
I think that pretty much every year, the #1 guy on the fWAR leaderboards in that league is a strong candidate, if not the best guy. You can sometimes make a case for someone else at the top, like Mauer in the example below, but generally speaking you’re looking at guys who are close enough to the leader that you can talk yourself into it. Jose Altuve last year was a fine choice…he wasn’t that far off of Judge. Josh Donaldson wasn’t that far off of Trout in 2015. There are some really egregious cases, but it’s hard to find a single year where someone topped the WAR leaderboard in their league and wasn’t at least a strong candidate.
Supporting Yelich over deGrom I think is understandable. But the idea i hear that J.D. Martinez and not Mookie Betts is the Red Sox/AL MVP is truly egregious. in that case fWAR alone is enough to shut that one down.
My rule of thumb is that if you’re within 1 fWAR of the leader, there’s potentially other mitigating factors at play. If you’re more than 1fWAR away, there probably isn’t, but let’s hear you out anyway…it’s certainly possible (especially if the lower player is a catcher with good framing). If it’s more than 2 WAR, I’m going to pretend to listen to you, but I already know you’re wrong. If it’s more than 3 WAR, then I’m not even letting you finish your sentence.
So I’m willing to hear out the reasons why Yelich’s team success and deGrom’s lack of it make a difference. It’s a potential case, although right now I don’t think that’s enough. But favoring Martinez over Betts isn’t that far away from “you should stop having an opinion on this” territory.
It was certainly laughable during the summer seeing Martinez being spoken of the same sentence with Betts for MVP. What I have found interesting, and quite different, this year is that the very large tribe of MIke Trout supporters, all of whom think he should have 6 straight MVP awards already and might even be right, are resigned to the fact that he was not the best player in the AL this year.
So I suspect part of the reason Betts grades out so well is because he’s more or less a center fielder playing in Fenway’s weird and huge right field. I suspect the defensive metrics don’t know what to make of him.
If you believe that the defensive metrics are overrating him, then you could make a case for Mike Trout. But I don’t really know that, or think that. And I do believe that Mike Trout probably should have had the MVP from 2012-2016.
I don’t consider myself a Trout supporter, but he should have won 5 straight MVPs. He was likely the best player those years. He was likely the best player a 6th year, but didn’t play enough to provide enough value to be the MVP. This year, Betts has likely been the best player. I would say there is a large tribe that believes the best player should win MVP, and that tribe believes, not resigned to the fact, that Betts is the best player this year.
Trout would undoubtedly have won it more if it were called Most Outstanding Player of the year.
Another really good take here written by someone who most certainly isn’t me. Sure is handsome though.
https://goodfundies.com/jacob-degrom-faces-an-uphill-battle-but-heres-how-he-can-win-the-mvp-7f7a9cc7b224
I wonder if that author might change his conclusion in light of the fact that Yelich raised his fWAR by almost a win and a half in the last 2 weeks of the season, leading his team to the top seed in the N.L.
“You create an award that looks to recognize the best player. Baseball has such an award. It’s the highest-profile award given by the BBWAA.”
Well, kind of. They have two. Unlike all of the other major sports leagues. Who cares who the NFC MVP was, or the Eastern Conference MVP? Do they even vote for those? I know they have separate all-star teams but it’s a weird historical quirk that even now when the teams play so many games across league borders that the awards are still split.
Mookie Betts for NL MVP.
I could go for a bracket system: NL best position player, NL best pitcher, AL best of each, these four face off for two league MVPs, and then the MLB MVP.
(Or would it make more sense to rotate 90° and have MLB best pitcher and MLB best position player. Think it would. But you know what, we can do both as long as we’re launching pie castles into the sky.)
“Unlike all of the other major sports leagues. Who cares who the NFC MVP was, or the Eastern Conference MVP? Do they even vote for those?”
Actually, the NFL doesn’t have a single recognized MVP. There are 3 separate awards by three separate institutions: The Associated Press, The Pro Football Writers Association, and The Sporting News.
The AP then further complicates things by awarding an Offensive Player of the Year and a Defensive Player of the Year. The MVP almost always goes to an offensive player; but about 50% of the time the AP gives the Offensive Player of the Year to someone who didn’t win their MVP.
Every MVP article includes the sentence to the effect of, “No one’s MVP ballot should ever be based just on a WAR leaderboard” and I feel like that should be amended to “No one’s MVP ballot should ever be based just on a WAR leaderboard, unless that WAR leaderboard shows one player that’s more than a full win above all his competition. In that case just give him the damn award regardless of context.”
The margin has to be bigger than that. 1 WAR is well within the range of reasonable disagreement about issues like defensive metrics, defense-independent pitching, etc. But if you’d said “three full wins” I think I might agree. (I guess with the further proviso that anyone can make a “WAR leaderboard” since WAR is a scale of measurement, rather than a metric, and at some point there might be a new WAR that I thought was totally nuts.)
Agreed, it’s an empty and obligatory homage.
There is actually extremely little variety in MVP ‘debates,’ and the very few topics for disagreement (leverage, position, team performance) are typically just content fodder.
You make an argument, but No.
Christian Yelich is the National League MVP.
First, I’m going to admit my bias. I’m a Brewers fan who grew up in Milwaukee, and I’m quite overjoyed to see the Brewers make the postseason for only the 6th time in their 50 year history. One cannot reasonably argue that Yelich’s play down the stretch didn’t have a lot to do with that, as he collected big hit after big hit to propel the Crew to the N.L. Central Division championship and the best record in the National League.
That said, it’s definitely not just a mere “homer” argument that Yelich has rightfully earned the N.L. MVP award.
First, there is the WAR argument (for the purposes of this comment, I will refer to fWAR). Final numbers have yet to be posted, but Yelich’s lead in WAR among all position players will almost certainly extend when the final numbers are tabulated, as he went 3-for-4 today with an RBI and a stolen base. Of course, his WAR total will *not* account for the atmosphere in which he did this: on the road at Wrigley in a game that clinched the division crown *and* best record in the National League. Coming into the day, his WAR total was 7.4, which is more than a full win above his closest competition–Anthony Rendon, who is done for the season with 6.1–and almost 2 full wins above the two guys who were tied for 3rd place at 5.6 WAR–Nolan Arenado and Lorenzo Cain.
The bottom line is, in terms of WAR, Yelich has been the best position player in the N.L. this season, and it really isn’t close. One could also add here that his wRC+ of 165 coming into today’s game (sure to rise a little more after today’s performance) is also tops in the N.L. by 25 points over the aforementioned Anthony Rendon. So if it comes down to position players, there’s no argument for anyone else besides Yelich. But what about DeGrom, with his 8.8 WAR? Why not DeGrom?
The answer is that WAR isn’t the *only* thing that matters. Position (Pitcher vs. Everyday Player) *does* matter. And more importantly, team context (Playing for something versus Playing out the string) *does* matter.
The fact that the pitchers have their own award shouldn’t mean they *can’t* be named MVP too, but the standard for sweeping both awards should be very, very high. After all, if it were all the same to name pitchers MVP versus position players, why would we need a separate award for pitchers? We need it because even the most dominant pitcher only affects one game out of every 5, and so he’s not going to be relevant to more than about 32 to 34 team wins in a season. You can parse the numbers in terms of number of plays, but the fact remains that any full-time position player affects far more *games* than any starting pitcher. Why would that not matter? I think most folks would agree that it matters a lot, and that’s why position players are rightfully favored in MVP voting, and that’s why we need a separate award for pitchers.
But what about a situation where a pitcher greatly outperforms even the best hitter, as we seem to see here, with the gap in WAR between DeGrom and Yelich being more than a full win, (probably) even after today’s game?
An even larger factor is the team context. I don’t agree at all that wins for the Mets in September–when the season is basically over and they’re just playing for their statistics–are the same as wins for the Brewers in September when they’re in a playoff seeding race and have a shot at best record in the league. As you say, this concept can be taken to an extreme, but a reductio ad adsurdum argument is not a valid argument that this factor has no weight at all. It is (at best) an argument that this factor should not be *over-weighted*.
Sure, it matters what a guy does in the beginning of the year, and it matters what a guy does even if he’s on a bad team. All of that is counted in WAR. But if we’re talking about value, we’re talking about wins and playoffs. Nothing is more valuable to a team than making the playoffs. So there has to be some additional value attributed to what a guy does when his team is in the battle down the stretch and playing for something. Showing up in the clutch when it really matters means something, and Yelich has been doing that for the past week, the past month, the entire second half, and even just today. Meanwhile, DeGrom pitched in 5 games in September that had no playoff implications for his team. His team lost 2 of those games, and few people cared.
So with all that said, the argument for Yelich goes like this: Yelich is clearly the best position player in the N.L. this year. The real question is, was DeGrom so dominant as a pitcher that he deserves the MVP award as a pitcher, despite affecting fewer games than an everyday position player? The answer is easily no. The smaller number of games he affected were not totally meaningless, but they weren’t as high-leverage as the games Yelich contributed to in the past week, month, and more.
And all of this ignores the elephant in the room: for all of DeGrom’s dominance, his team only won 14 of his 32 starts. That’s a team win (if results matter) in fewer than half of the small amount of games he had the ability to affect at all. So if there’s a season where a pitcher has *actually* affected his team’s bottom line so greatly that he deserves to be MVP over any position player, DeGrom’s 2018 season is not it.
You literally just regurgitated all the points Jeff dealt with in the article.
Reading comprehension is not your strong suit.
“his team only won 14 of his 32 starts.”
How many of those games do you think they would’ve won without him? With, say, a replacement-level pitcher substituted in his place? There’s a metric someplace around here that actually answers that question!
Cute response, but you miss the point. The point is… who cares how many they win without him? Either way, they miss the playoffs. We’re talking about value. As in… Most Valuable Player. Teams derive value from wins. Wins = ticket sales. Wins = playing extremely valuable extra games in the postseason. Teams don’t derive nearly as much value from losing by less while still losing, as compared to actual winning.
There are many starting pitchers this season whose marginal contributions in 32 or 33 starts led to more team wins. So by that metric, DeGrom is not even the most valuable pitcher in the N.L. He’s certainly not the most valuable player.
Arguing the MVP should be determined solely by WAR is an absurd argument that puts statistics above all else, with no regard at all for the value those statistics actually added to the team’s bottom line. A position player on a losing team (i.e., Trout) can, over the course of 150+ games, drive ticket sales at home and on the road due to his greatness, and thus can offset his team’s failure to compete. A pitcher gets 32 or 33 turns, and thus cannot offset his teams lack of wins in the same way.
“The point is… who cares how many they win without him? Either way, they miss the playoffs.”
Except most of his value has zero impact on whether or not they made the playoffs. If Yelich had had a 1 WAR season, the Brewers would have still made the playoffs.
Another point. Let’s say Jose Ramirez had stayed hot and led the AL in WAR.
Would he have been eligible for the MVP? After all, the Indians (just like the Mets) didn’t play a meaningful game for the last 60 or so games of the season.
This nonsensical argument leads to kind of an interesting thought experiment, actually. If the MVP voting is supposed to take into account the nonlinearity of added wins, then you have to be choosing among the best guys on, what, the 82-to-87-win teams that either just missed or just squeaked into the playoffs? Would that make this year’s NL MVP… Max Scherzer?
I doubt that.
The Brewers would very likely not have made the playoffs if you replaced Christian Yelich with say Marwin Gonzalez or Billy Hamilton. I understand the value in WAR, but to interpret it so literally and imply that it directly correlates to how many games a team actually wins is stretching its application. This site even admits that the statistic is imperfect, but it’s referenced as though it’s the Pythagorean Theorem. The various sources don’t even agree on how to calculate it!
Let’s never forget to watch the games and to reward players for their real life results. Baseball is not played in a context neutral vacuum.
I don’t even mean to say that I think Yelich should win over deGrom. I just think we all need to recognize that, just because WAR is the best metric we have, it isn’t perfect, and it most certainly should not be applied literally.
Where do those 6.6 wins go if he’s a 1 WAR player rather than 7.6? They float off into space? Or does one of his opponents win those games instead? Maybe even the Cardinals?
First off, I would think it’s obvious, but I’ll say it because it seems as though some are under the impression that baseball is played on a spreadsheet and not in the real world, but you can’t win or lose 0.6 games. Wins and Losses are just binary like that.
Beside that, if he’s a worse player, it changes the average variables that WAR is based on, so these types of hypotheticals must assume that another player or group of players are increasing to the degree that he is decreasing.
Can’t we agree that WAR is the best metric for displaying player value without needing to apply it directly to the amount of actual wins a team accumulates?
So you think the Mets would’ve gone 6-26 in the games he started with a replacement level pitcher? That’s real hard to argue. In 8 of the14 games they won, they scored at least 5 runs. In 5 of them they scored more than 8. The lowest scoring games they won with deGrom on the mound, the Mets scored 3 runs, this happened 3 times all year. Even Josh Tomlin had a couple starts this year where he gave up 1 or 0 runs, so it’s not like a replacement level pitcher is going to lose every low scoring game either. If you put any average starter in those games, the Mets likely win … 12? 14 still? More, if the pitcher happened to pitcher better in the games where his team scored a couple and/or the bullpen didn’t get blitzed?
A replacement level pitcher would have given up 5 runs per 9 innings fairly often – so yes they probably would have lost some of those games.
deGrom did his best to help the Mets win baseball games. He did this the best out of anyone in baseball – which is why his WPA is the best out of any hitter or pitcher. The MVP is an individual award – the fact that deGrom had worse teammates than Yelich, should have no impact.
“any full-time position player affects far more *games* than any starting pitcher. ”
Except most of the time the impact of a single position player in a single game is negligible. Whereas the impact of a starting pitcher, assuming they’re in the game for a decent amount of time (as deGrom was), is quite large.
“And all of this ignores the elephant in the room: for all of DeGrom’s dominance, his team only won 14 of his 32 starts.”
So it’s deGrom’s fault that the Mets bullpen and offense sucked? He should have demanded a trade before the beginning of the season to the other New York team and then he’d be worthy of winning an MVP?
Also, if we’re going to go for accuracy, we should include deGrom’s hitting. And also average between FanGraphs and Baseball Reference, since no WAR metric is 100% accurate. Doing that, we have 9.6 WAR for deGrom and 7.4 for Yelich.
Jeff answered most of these misunderstandings in his article.
Your main misunderstanding is pitchers effect their team less because they appear in fewer games. That is a logical fallacy, relief pitchers appear in more games than SPs but are not thereby more valuable. As Jeff wrote, Degrom faced 835 batters, Yelich batted 647 times: he also analyzed the volume of defensive plays. WAR, of course, also does all this with linear weights: and also finds that Degrom impacted games moreso than Yelich.
The real debate is between the “any reason can be legit” camp and the “consistent, comprehensive, method-based approach’ camp. This comment is a honest fan’s passionate case for Yelich, but it is clearly a pick-a-reason grab bag based on recency and the feel good euphoria of the Brewers’ playoff chase (i.e. the common factor in nearly all award voting that is retrospectively acknowledged as misguided), whereas the Degrom case will stand the test of time (though clearly Yelich will win) as it is based on fact-rich consistent method.
Sure, but position players account for a far smaller portion of each game than a pitcher does for their starts. That’s why Jeff points out that deGrom actually had more PA against than Yelich had at the plate.
But the starts he actually pitched in, his performance didn’t really matter much due to circumstances mostly outside of his control–his offense not scoring much and/or the bullpen getting bombed and coughing it up anyway. That can happen when you’re only appearing in 1/5 of your team’s games and you’re not going all 9 innings in those games. But when you’re in 147 of 162 games as a position player, chances are a bunch of the games where you performed well, your performance actually mattered to the outcome.
1. Measuring what “mattered to the outcome” in the individual game situation is what WPA is for, which is in fact addressed in the article.
2. “Chances are”? I don’t get why people think this argument favors position players — presumably it’s neutral. Each plate appearance has a pitcher and a batter; so each plate appearance has exactly as much “mattering,” given the game situation, to distribute to one side’s run prevention and the other side’s run creation. “Chances are” that Yelich did a lot of his hitting in non-decisive game situations, too.
for #2, there’s a very clear, obvious difference. If a position player only appeared in 1/5 of the team’s games, it’s very possible due to randomness that you the player could overall perform extremely well, but have limited impact on the outcome or any of them. E.g. in those 30 some games, they won 15 of them by 4-5 runs, and in 10 others his team’s pitching was awful and they lost badly. Our good player contributed runs in those games, but they would’ve won or lost them anyway had he gone 0-4 or 4-4 with a home run. Then our player has some bad games in some of the remaining games that were actually close. But over the course of 150 games, instead of 30, those leverage opportunities are going to naturally even out.
So you can make the argument that Jeff does that deGrom influenced more innings … well, that’s fine, except the innings he influenced didn’t end up meaning much for the most part because of how those overall games played out. To say you can ignore that for an MVP award seems quite a stretch.
So on the one hand, we have WPA, an actual quantification of the individual “impact on the outcome” and/or “meaning” and/or “mattering” of each play in every game — by which the starting pitcher here demonstrably DID do just as well as the position player. And on the other hand, we have a vague intuition about how that is or isn’t going to “naturally even out” across games and situations. I know which of those I believe, but I’d invite you to justify this with anything at all besides handwaving.
So, position player, yadda, yadda, yadda, clutch, Brewers fan. Got it.
I tend to like this thought experiment:
Taking the season in a vacuum, which one player would I most want to take into the playoffs and/or build a (hypothetical) 2018 season around?
To me, that gets down to the heart of what the MVP should be. Yelich is a great pick. I think deGrom is a better pick.
Playoffs vs regular season is a big difference.
The ‘no one should ever vote strictly on WAR’ line is overdone imo. The key is that WAR is a best fit model, it’s the best guess of player value. There isn’t a better guess. It’s just fashionable to pretend that some ‘folksy good ole’ fashioned common sense’ is of course needed, rather than a consistent, evidence-based, best estimate.
Additionally, it’s clear that the author actually is simply equating most WAR with MVP, which he’s right to do imo. After all, his argument here is ‘three different types of WAR say Degrom is more than a win better than Yelich, therefore Degrom is MVP.”
That’s a perfectly fine argument- but with WAR being a bit unpopular, no journalist is likely to say ‘most WAR = MVP,’ it’s simply become an unpopular stance. Many journlists now adopt a seemingly obligatory ‘of course no one would ever be so crazy as to just vote on WAR’ line before saying AL MVP should be Betts, over Trout, with Ramirez in 3rd (as it should be imo).
The author isn’t arguing “we need to find the correct MVP formula, because WAR has unacceptable low R-squared: once we find a formula with predictive values of .95, then we can award all MVPs with the new formula.”
Instead, he is tipping a cap to the traditionalist, romantic, sentiment-based marketing meme that the MVP race should be about opinions, so that everyone can have their own view that is legitimate in its own way.
Fair enough, Jeff’s work is top stuff: but Jeff’s work is great exactly (and to be honest only) because he embraces WAR and best-fit models/projections, not because his folksy common sense insights are valuable (like nearly everyone’s, they are not). That might sound a bit harsh, but there are lots of lovable authors with great hiking tidbits and baseball passion: what makes Jeff’s work stand out as great is that he has a serious empirical basis for his ideas, based off of a studious reliance on best fit models. Thus a smart article arguing Degrom for MVP because three types of WAR indicate he’s been a win better than his peers. Journalists would never say it because it’s unpopular and anti-marketing friendly, but the smart ones ought to proudly just vote WAR, and if the WAR formula is missing something and therefore not accurate enough, improve it.
.
Agreed. Everyone has their own “WAR.” That is, every traditionalist has an opinion of each candidate and how valuable they are on an overall level. It’s just that WAR is the only systematic way to measure these things. Traditionalists think their ad hoc appraisals are more reliable than a mathematical system, but they manifestly are not. Start with WAR, explore the facets and margins of the model that we know are not bulletproof, like defensive metrics and how to allocate responsibility between pitcher and defense. That should give you an idea of the error bars for each player. Then make your best guess. Any other method is nonsense.
He only mentioned 3, but deGrom also leads in bWAR.
No
Counterpoint: Yes
Rebuttal:
“Good signing, especially at this price, I’d say. In addition to the January discount we could speculate about the likelihood this was a hometown discount for New Jersey kid Frazier.”
https://www.fangraphs.com/blogs/mets-add-todd-frazier-at-bargain-price/
Your biases are showing. The pounding the boards with votes makes it so much sadder.
Oy vey.
“The most valuable player is the best player, according to the only intent that makes good sense.” I think you could view deGrom’s w-l record to make a pretty strong counter-argument, and the reason why it’s pretty hard to give the award to a pitcher, who only impacts about 20% of their team’s games in any way. deGrom went 10-9, and the Mets went 14-18 in his starts. If the Mets had started a perfectly average starter with an ERA in the mid-4s, how many fewer games would they have actually won? Being that he won games where the Mets won 6-2, 8-2, 5-1, 12-2, 8-0, 8-5, and the Mets won a few of the others he pitched but didn’t get a decision by scoring over 5 runs, it’s feasible that if you’d given his starts to, say, Cole Hamels, they literally might not have won any fewer games. They almost certainly wouldn’t have lost more than 3-4 more of those.
If you remove Yelich from the Brewers on the other hand and replace him with an average outfielder, I have no problem thinking they easily win 5-6 fewer games, if not more. Because he’s in almost every game, how his performance impacts the team is a lot less variable than it is with a pitcher.
Again deGrom is NOT responsible for how many runs the Mets score when he’s on the mound or hows those runs are distributed.
But let’s try something here. The Mets scored 113 runs in deGrom’s starts or 3.53 runs per game. If you’re going to replace him with an average starting pitcher, then you should replace him with an average offense as well. The Nl league average was 4.37 runs per game so we’ll go with that for simplicity’s sake.
Now an average pitcher with an average offense should go 16-16 across 32 starts. (or rather, his team should, assuming the bullpen is also average).
On the other hand, if the Mets had scored 4.37 runs per game for deGroom, they would have gone 23-9 in those games. (note: I’m giving them 4.37 run every game; we should probably distribute those runs in some reasonable fashion but I don’t have the time or energy to do that and I’m not sure it really matters).
So that makes deGrom 7 Wins Above Average. Now Baseball Reference actually calculates Win Above Average. They have deGrom at 7.8. So my little back of the envelope calculation actually worked out pretty well. The other 0.8 wins are probably because the Met did NOT have an average bullpen when deGrom pitched. Just like the offense, it was decidedly below average.
Yelich, by the way, had 5.7 Wins Above Average. So you were right about that. Just wrong about deGrom.
But how many runs the Mets score can affect how valuable deGrom is to them. The voting instructions quoted in the article specifically speak of the player’s value to his team, not his value to an average team.
“Again deGrom is NOT responsible for how many runs the Mets score when he’s on the mound or hows those runs are distributed.” Of course he’s not, but this is exactly why it doesn’t usually make any sense for a pitcher to win an MVP award. His ability to influence games is determined greatly by semi-random performance of the rest of the team, when any starting pitcher is only even capable of influencing 15% of the team’s innings. When you’re only able to influence that small % then you’re susceptible to things like deGrom this year, where he didn’t really end up adding much of anything to his team’s win total with his individual performance.
“So that makes deGrom 7 Wins Above Average. Now Baseball Reference actually calculates Win Above Average.” In the real world of what actually happened, which is what you should be looking at for any award, deGrom clearly did not contribute 7 more wins compared to an average pitcher. It’s not like the games they won, the won 1-0. The games they won, they mostly actually scored a bunch of runs. They won 0 of the games where the team scored 2 or fewer. If you put an average starter in his games the overall outcome is almost certainly about identical–they win the games where they scored a bunch of runs, they win a handful of the games where they scored 3-4 runs, and the bullpen has to try a little less hard to blow a bunch of games because the starter already gave up some runs
Do you know the average margin of victory in Max Scherzer starts when the Nationals win? It’s 4.5 runs per game. There are very few MLB pitchers who are 4.5 runs/9 worse than Scherzer, so it’s reasonable to assume a league average pitcher would have won the same number of games, right?
Great stuff Jeff. I like how you reasoned through the argument instead of regurgitating the stats we already know
Here’s why I will always go with the position player, versus a contemporary starting pitcher- except in the most extreme cases. Today’s SPs don’t pitch into the most crucial part of the game, the 8th inning onwards. DeGrom is not an exception.
DeGrom pitched all of 11 innings (39 ABs) past the 7th, in total – all year. Yelich had 130 such plate appearances. Was DeGrom in with the game on the the line? He faced 3 batters in the 9th. Yelich had 58 plate appearance in the 9th or extras.
Position players are just more likely to be in there when it’s all on the line.
By this reasoning then the MVP should usually go to the best closer. You’re penalizing deGrom because he’s good enough to start rather than close. You’re also hand-waving an entire class of players because of something that can actually be measured by WPA, by which deGrom grades out very well. It’s the same as Hall of Fame voters refusing to vote for a DH, or someone who played in Coors. They correctly identify something that makes a player less valuable, but instead of simply adjusting for it, they just refuse to vote for any such players.
In the future, when every pitcher is pitching three innings at a time as God intended, then it may very well be that no pitcher is affecting enough plate appearances to be considered for MVP. But for now, by the best measurements we have, starting pitchers absolutely do accumulate enough value to rival position players and contend for MVPs, and the leverage issue is in no way big enough to settle the debate over total value.
“You’re also hand-waving an entire class of players because of something that can actually be measured by WPA, by which deGrom grades out very well. ”
You’re being a bit modest there. He actually led the NL in WPA. Okay, he was tied with some guy named Yelich. But yeah, a great pitcher can accumulate just as much WPA as a great hitter.
No, it wouldn’t go to a closer, because SPs face so many more batters. Though it’s obviously true that an inning pitched in the 9th is more important that one in the first – and WPA takes that into account.
Yes, SPs can rival position players – that’s what WAR comparisons are for. However, take this case. Player 1 and Player 2 have the same WPA. Player 1 gets his from dominating the first part of the game – Player 2 from being a little less dominant in the last three innings. Who is the more valuable player? I think Player 2 – because it’s easier to be dominant, when the game isn’t on the line.
Look, I’m talking tie-breaker stuff here. But if SPs do get to the point where they’re only pitching 3 innings – then the closer with the same WPA gets my vote. Because he’s not out their dominating when people are still finding their seats. He’s a little less dominant, but when they’re all screaming their lungs out on the road.
“You’re also hand-waving an entire class of players because of something that can actually be measured by WPA, by which deGrom grades out very well. ” This goes to show how every stat is imperfect. deGrom grades out well in WPA because he pitched in a bunch of innings where his team didn’t score much and he kept his team in it by not giving up runs either, i.e. the games were often close when he was pitching … and his team ended up losing almost all of those games anyway. Not his “fault” … but I don’t see how you can just ignore that for a debate about who is the most VALUABLE player. He literally got outpitched by the opposing starter in the majority? half? of his starts. Again, all it shows is how the maximum value of a starting pitcher is always limited, and can be limited even more by the randomness of what happens on the days he pitches.
“He literally got outpitched by the opposing starter in the majority? half? of his starts”
LOL wut?
god no his team stunk and he really made no difference, whether they had him or not it made no difference to their final standing.
Those reading words are hard to brain think aren’t they? Just so many pain. SO SO MANY PAIN FROM THE READ WORDS!
(srsly bro, did you read literally any of the article at all?)
I was all for deGrom MVP, hands down, until the few weeks. Yelich did the MVP thing of putting a 237 wRC+ up for the last month of the season, including going 3 for 4 with a stolen base in the division-deciding game 163.
He was already an excellent player, but finding a new level and using that to propel his team, not only into the playoffs, but as a division-winner, is MVP to me. If Yelich wasn’t in the picture, I would vote deGrom and Scherzer as 1 and 2 hands down.
If we’re not going to look at context, I don’t see why we have a vote for the award instead of taking some sort of weighted average of different WAR calculations. We already know deGrom had the most valuable season based on any measure of WAR, and anyone looking back at this era will see that deGrom had one of the absolute best pitcher seasons, hands down.
I agree with every argument in this article. But it assumed, and didn’t really bother to argue, that deGrom was the league’s best pitcher. I’d give the N.L. Cy Young, and maaybe M.V.P., to Max Scherzer.
Just so you know you are not alone. I would also vote for Scherzer for the Cy Young. I would give the MVP to Baez. I don’t think anyone meant more to their team than Baez this year. WAR doesn’t appreciate much of what Baez is/does. I don’t think the Cubs are anywhere near the playoffs without Baez. It is easy to follow why others would chose Yelich. He had an amazing year.
When will MLB end this madness and create a third award for best position player?
Better yet, participation trophies for all!
I agree with DeGrom for MVP and in general think that the best player should be evaluated based on his performance and not his team’s performance. That said, if ever there was a time when the “playoff race context” thing comes into play it would be this year with what Yelich did down the stretch. 218 2nd half wRC+ and 237 in September, a month in which the Brewers started 5 games out and ended up winning the division. Hit after hit coming in high leverage situations. Ending the season on an 8-game win streak during which Yelich had 11 hits, hit 4 HRs, and took 12 walks in 37 PAs. In 3 consecutive 1-run games before the last game of the season he took 5 walks in the first and then hit a combined 3 HRs in the next two, and then went 3/4 in the tiebreaker.
So normally, yeah, I would agree, it doesn’t matter how good the player’s team is or how close they were to making it/not making it in the standings, but this is a bit of a special case. I would still go with Degrom but the context argument is not entirely un-compelling here. At the very least I would make it a tie-breaker if the argument came down to Yelich and another position player with the exact same WAR.
Degrom had arguably the best pitching season of anybody in the last 15 years, since Pedro and the Big Unit were on the mound. In the absence of a clear Mike Trout type performance I think you have to reward that with the MVP.
Good arguments. Ones I would agree with if it was DeGrom v Rendon or Baez for example. But I’m a Brewers fan so my head is too busy being stomped on by my heart for it to follow such a rational point of view.
My only minor quibble would be that to pivot straight from using WPA to him not being punished for his team being bad is a little disingenuous.
If he shouldn’t get dinged for his team being garbage, I’m not sure he should get credit for it either.
I’m pretty sure that he ended up leading the NL in all 4 main WAR systems (fWAR, RA9, WARP, and bWAR), not just the three that you mentioned for whatever that is worth.
Nola had more bWAR than deGrom. They were tied until Nola’s last start of the year, where he took the lead.
Nola went ahead in pitching WAR, but tied deGrom for overall WAR (and is actually behind if you go more than a few decimal places.
deGrom doesn’t lead WAR at baseball-reference. What is the community position on that? The only smart people are here? Are they a bunch of incompetent idiots or does what they do have value? Does anyone even want to acknowledge that WAR is computed differently in different places?
Yelich was one RBI and two HR from the first NL Triple Crown in more than 80 years on the team with the best record in the league. By any standard he had a MVP season. WAR skews toward pitchers. I don’t see how anyone without an extreme East Coast bias could even put DeGrom in the discussion.
Because RBIs are a meaningless stat? And “Almost winning the triple crown” isn’t really impressive to most people (never mind the fangraphs crowd).
“WAR skews toward pitchers”
Please explain, as I have seen no evidence that supports this claim.
If we are to measure the value of a player’s contributions during the season it seems like WPA is what is trying to do the right job. But that would be heavily influenced by luck, as opposed to true playing quality. WAR is nice in that it comes closer to a measurement of talent, but there are some odd compromises in its definition. Hits gets decontextualized but no attempt is made correct in the defensive metrics for the number of opportunities to make a play different players have faced, and player gets a chance to accumulate higher WAR by having more plate appearances. So it is not measuring either what the actual value of a players contribution was, nor really what his skill is. Plus there are the well-known measurement errors.
Still, WAR is probably what comes closest to giving us a measure of the “best” player and I can accept that anyone who is suggesting a player with clearly lower WAR as MVP winner should explain what WAR is getting wrong.
As someone who played a (different) sport at a semi-pro level, I’m convinced that “luck” is a skill. If you’re a good athlete, have good fundamentals and you hustle, then you can turn some bad plays into ugly-but-good plays that look like “luck”.
We shouldn’t penalize a ballplayer for what is perceived as luck.
Just so I have this straight.
Zack Greinke has a career .295 BABIP, so it isn’t that he’s had neutral luck, but that he’s neutrally skilled in BABIP.
When in five consecutive years he put up .306, .276, .311, .229, .294 was that changes in skill or did he have three roughly normal years with one where he was rather lucky and one where he was very lucky?
I’m saying he earned each of those BABIP figures. Nothing complicated. You earn your luck.
Just because he has a career average of .295 doesn’t mean he performs at that level every year or even every day. I’m sure everyone here will agree that there’s plenty of room for statistical variance, but any athlete will also tell you that there’s plenty of personal performance variance.
One thing’s for sure, Sullivan (again) gets my nonexistent FG MVP award. This piece actually persuaded me on the general issue (not on NL MVP; I’m an AL fan and have no opinion).
Maybe we need to have discussion about what it means when the comment section is (significantly) longer than the original piece…? Oh, wait. That could be a future feature!
In this case it seems to mean first and foremost that a lot of people didn’t read the article. And secondarily that it’s a topic that a lot of people feel like arguing about right now.
The only real question that a voter (or anyone) needs to answer is:
Does “value” mean “added the highest number of wins to his team’s total”, or does it mean something else?
It’s a *very* subjective thing. Those who agree with the former definition cannot be faulted – and thus DeGrom should be the MVP.
Those who believe that value is more tied to team success (in, for example, “where would the team be without this player, as opposed to where they were with him?”) can’t be faulted either, as it’s reasonable to argue to “value” is meaningless without context (e.g. a hundred dollar bill has no value if you can’t buy anything with it). And thus Yelich should be the MVP.
(The one argument that *can* be thrown out is that a pitcher shouldn’t win – that can be shown to be objectively false.)
There are passionate opinions on both sides of this; we should not be criticizing or downvoting anyone just for being on one side of it.
… and it can get trickier. Where were the Mets without Cespedes? Well, it meant that Nimmo got PAs and he put up a better year by fWAR than all but Cespedes’ best years and even comparing to that year, YC had 129% of Nimmo’s fWAR in 126% of the PA, so had he just gotten hurt earlier, the two would likely have been even closer.
Does that mean if YC was healthy and got 676 PA again and the extra PA came in at the same fWAR rate as his 157 PA this year [3.9 fWAR], that he’d have negative value because his replacement was actually even better?
“The most valuable player is the best player, according to the only intent that makes good sense.”
Overall this is excellently written, but with all due respect I don’t believe you’ve made much of an argument for why this is the “only intent that makes good sense.”
If anything, you’ve convinced me more that the MVP award should be abolished, and replaced by the Player of the Year award, with clear voting guidelines that indicate it is to be given to the best player in the league.
This article is extremely self-fulfilling.
Baseball writer works for website > Website creates single metric to display individual player value across positions which accounts for their performance in all facets of the game > Writer says Player who leads in that value metric is the MVP
I mean, duh? If someone on this site wrote an article that someone other than that player should win the MVP, that would pique more of my interest.
Now beyond the implicitly obvious, these arguments are all extremely valid. Players from losing teams or non-contenders should be included equally in voting. Pitchers should be included equally.
But there is 100% merit to the argument that higher leverage games are more difficult to perform in. Not saying that losing teams aren’t trying to win, or that their opponents aren’t trying to win, but there just isn’t as much at stake for the individuals. Not saying there is nothing at stake, but just not as much.
And the difference and the variance is difficult if not impossible to quantify. But let’s not pretend like it doesn’t exist. Baseball is not played in a vacuum, no matter how badly we want it to be.
In closing, who cares who wins the MVP? We all acknowledge the evident biases in voters. This is an article crusading for the Fangraphs metric, and that’s fine. But even if deGrom doesn’t win, it could be for any number of reasons. If it went to the best player every year, Mike Trout would have won it more than twice already. It’s about PR as much as it’s about merit, frankly.
If I could amend my comment above, I would say that the commenters, more so that the article itself, are crusading for WAR as a metric. But we must admit its imperfect. The various sites cannot even agree on exactly how to calculate!
I also totally do care about who wins the MVP and appreciate the debate surrounding it every year. Just saying, don’t let the winner deter your opinion of who was most valuable.
Like Coach Boone told Coach Yoast, “You’re a Hall of Famer in my book.” Isn’t that what’s most important?
This whole “WAR isn’t perfect” argument is so tired. No one is saying it is perfect. No one believes it to be perfect. It doesn’t have to be perfect. It just has to be the best thing we’ve got currently, which it is. If there is a better way to capture the totality of a player’s individual contribution to winning, someone tell us what it is.
I agree that it’s the best metric we have for this purpose. Never said it wasn’t.
People absolutely do reference it as though it is infallible though. Every time I hear the argument that if you replace a 6 WAR player with a 3 WAR player, that team would literally win three less games, I cringe.
I just think it is often referenced too literally.
You’re 100% correct, but of course the reason is not because WAR is missing the mark. It’s because of the random things, the statistical variance in the sequencing of events, which are the very things we should be stripping away when trying to determine the value of the actual individual player. And we can’t actually know how this type of “if player A were actually on that team instead of this team” experiment would work out. Shouldn’t we then instead focus on the things we actually can measure?
This gets back to the point sadtrombone has been making here and in other threads. If you add up deGrom’s pitching and batting fWAR, it comes to 9.1. I understand the argument that says that WAR doesn’t capture certain things, but Yelich’s fWAR is 7.6. A full 1.5 WAR gap is sizable. What are the things that are not measured that are LIKELY to account for this large of a gap. Focusing only on pitchers, 1.5 fWAR is the difference between Blake Snell and Charlie Morton.
Totally agreed, and thank you for articulating. It is certainly a sizable gap and I wouldn’t begrudge anyone for thinking that deGrom should win.
My argument above wasn’t for Yelich > deGrom. More so I wanted to voice that there is a difference in the level of pressure on the individual in a pennant race versus one who is playing only for pride, which we can’t quantify, and that we shouldn’t allow ourselves to be bothered by the writers’ selection knowing their evident, dated biases.
That said, when a player’s performance so clearly drives his team’s push for a pennant, and that performance isn’t what we view as “anecdotal”, it’s backed up by advanced statistics, I’m happy using the circumstances of that situation as a tiebreaker of sorts over the player who was on a team playing out the string. I acknowledge that this is outside of their individual control, but that’s just life.
So I guess my argument here is for Yelich > deGrom – LOL
Also, I think it’s time to place a much higher emphasis on winning a division in the context of these MVP discussions. Not saying we should exclude players who don’t play for division winners, but we can reward those that do specifically in circumstances similar to Yelich’s. We continue to group division winners and wild card teams together as “making the playoffs”, but there is obviously a huge difference. Yelich’s performance won his team a division. Any deviation from his regular season excellence would have likely relegated his team to the WC game and could have cost them a shot at the title.
I was surprised to see deGrom and Yelich tied in WPA.
That’s nice, but not concerned with the “best player” argument when discussing MVP.
While I fully believe in the use of statistics and other metrics as supporting tools to find out who the best players are and what is the best formula for the team to win games, I find some of the conclusions that are presented difficult to comprehend. For example, in this thought provoking article that I just read, Jeff said that Jacob deGrom earned a lot of WPA credit for pitching well in close, low scoring contests which is obviously true but that brings up an interesting question. It goes without saying that he did not win many of those games but in this strange new world that was good for his metrics. If, instead, the Mets had not actually played like the Rumble Ponies when he pitched and scored enough runs so that the games were not close he might have won 18 games, but, but, but he wouldn’t have amassed nearly as many of those wonderful WPA credits because he wasn’t pitching in those tight situations. Sounds like the horse is well ahead of the cart
Wow. Well put.
It is my opinion that this is an indication that pitching WAR needs some major adjustment. You can weight the value of an inning or a batter faced however you want. I would imagine that a SP roughly faces a comparable amount of PAs as a hitter has in a season. You could make WAR tell a story where SP and position players are of equal value with the proper weights or you could make it come out where its half as valuable, or twice as much… etc. Its arbitrary. deGrom has not had a historic season. His WHIP doesn’t even lead the league (tie), nor does his K/BB ratio which are pretty good metrics. He doesn’t even lead the NL in pitching WARA at baseball-reference. deGrom doesn’t even have a clear cut argument for the Cy Young award – it should be close, but people seem to be campaigning really hard for deGrom, so I don’t think it will be. I don’t think that is fair to Max Scherzer who has had a highly comparable season.
The only time a pitcher should even be in the MVP discussion is under extraordinary circumstances and this doesn’t qualify. This looks like proof to me that pitching WAR needs some major adjustments . Bad equations produce bad results… or as they say in computing – garbage in, garbage out.