Jacob deGrom and the MVP Precedent
Historically speaking, pitchers don’t win MVP awards — or don’t win them often, at least. There are exceptions to the rule, but the honor historically has been reserved for a league’s top position player. The logic among voters generally follows a couple recognizable lines of reasoning. Pitchers don’t play every day, some voters argue. They have the Cy Young all to themselves, say others. Whatever the justification, the record reveals a preference for position players over pitchers. Consider: since reliever Dennis Eckersley won the American League MVP in 1992, only Clayton Kershaw and Justin Verlander have been recognized as their respective league’s most valuable player.
It’s possible that some voters aren’t using entirely sound logic to arrive at their conclusions. Even if they’re employing the wrong process, however, they’re still usually arriving at the correct result: pitchers simply aren’t the best players in their leagues all that often. Position players make an impact at the plate and on defense. They just have more opportunities to create value. Their roles give them a competitive advantage.
Or, they usually provide a competitive advantage. This season, however, that hasn’t been the case. This season, the best player in the National League is likely a pitcher. While a lot of obstacles stand between Jacob deGrom and an MVP award, he deserves consideration — and there’s a really good argument he deserves to win.
Before we get to the more compelling arguments in favor of deGrom’s MVP candidacy, it makes sense to entertain the less compelling ones, too. First among them is the Mets, who have been poor this year. While voters are explicitly told that the MVP needn’t come from a playoff team, voters have typically evaluated a player’s performance in the context of his team’s performance, the logic presumably being that the player in question has been an asset in the most important situations. While the Mets are heading for a high draft pick now, it would be unfair to say that deGrom hasn’t been pitching in meaningful games. Thanks to their strong start, the Mets’ postseason hopes were remained alive into mid-June. Even if you wanted to assess deGrom some sort of penalty for playing for a bad team — let’s say you discount his second-half WAR by 50% — he would still lead the National League by that measure. It’s also worth noting that deGrom leads all National League pitchers in win probability added. In the games he’s pitched, in other words, he has been incredibly helpful to the cause of potentially winning a game, even if the end result has been disappointing.
Ultimately, there will be voters who are dogmatic in their views on which players are eligible for the MVP award. To those who contend, for example, that pitchers oughtn’t win it or that it should go to a member of a playoff team, I have little to say other than the rules and ballot history suggest otherwise. For those who are prepared to entertain the possibility of such a thing, however, then Jacob deGrom has a really good case.
At the moment, Jacob deGrom’s 8.3 WAR and 8.6 RA9/WAR (that is, WAR calculated with runs allowed and not FIP) lead the National League. If he finishes the season first by both measures, he would become only the second pitcher to do so since 2001. The only other pitcher to lead the league in WAR and RA9/WAR in the last 18 seasons — or 36, really, if you include both leagues — is Kershaw back in 2014 when he won MVP. Going back to 1967, when Cy Youngs were first awarded in both leagues, there have been 29 seasons out of 104 where a pitcher has been the league leader in WAR as we calculate it here at FanGraphs. It’s been slightly more common for a pitcher to lead the league in RA9/WAR, occurring 44 times. Only 18 times has a pitched topped the league in both WAR and RA9/WAR. In 15 of those 18 seasons, the pitcher won the Cy Young.
| Year | Player | League | WAR | RA9/WAR | Cy Young |
|---|---|---|---|---|---|
| 1968 | Bob Gibson | NL | 9.1 | 14.0 | 1 |
| 1969 | Bob Gibson | NL | 9.8 | 12.1 | NA |
| 1970 | Bob Gibson | NL | 11.3 | 9.3 | 1 |
| 1972 | Steve Carlton | NL | 11.6 | 13.5 | 1 |
| 1972 | Gaylord Perry | AL | 8.2 | 12.0 | 1 |
| 1973 | Bert Blyleven | AL | 10.8 | 9.7 | 7 |
| 1978 | Phil Niekro | NL | 9.1 | 8.1 | 6 |
| 1978 | Ron Guidry | AL | 9.1 | 11.5 | 1 |
| 1985 | Dwight Gooden | NL | 9.7 | 13.6 | 1 |
| 1986 | Mike Scott | NL | 8.4 | 9.2 | 1 |
| 1995 | Greg Maddux | NL | 7.9 | 11.0 | 1 |
| 1995 | Randy Johnson | AL | 9.5 | 8.7 | 1 |
| 1997 | Roger Clemens | AL | 10.8 | 12.6 | 1 |
| 1998 | Roger Clemens | AL | 8.2 | 8.5 | 1 |
| 1999 | Randy Johnson | NL | 9.0 | 10.3 | 1 |
| 1999 | Pedro Martinez | AL | 11.6 | 9.9 | 1 |
| 2000 | Randy Johnson | NL | 9.6 | 8.5 | 1 |
| 2014 | Clayton Kershaw | NL | 8.1 | 8.3 | 1 |
| 2018 | Jacob deGrom | NL | 8.3 | 8.6 | ? |
Data comes from our combined leaderboards, which contains pitcher batting value, too.
In 1969, Bob Gibson produced a great follow-up to the 1.12 ERA he’d posted in 1968, recording a lower ERA and 40 more innings than eventual winner Tom Seaver, but the Mets hurler was near-unanimous — Phil Neikro garnered one vote — in an era where voters selected just once candidate and Seaver had 25 pitcher wins. Blyleven and Niekro suffered from mediocre win-loss records, which likely hurt their candidacies. The last 10 pitchers to lead both in WAR and RA9/WAR won the Cy Young*. That bodes well for deGrom this season.
*While the point is obvious, I should still probably point out that voters could not rely on these figures at the time because they didn’t exist.
Here’s how those same pitchers fared in MVP voting.
| Year | Player | League | WAR | RA9/WAR | Cy Young | MVP |
|---|---|---|---|---|---|---|
| 1968 | Bob Gibson | NL | 9.1 | 14.0 | 1 | 1 |
| 2014 | Clayton Kershaw | NL | 8.1 | 8.3 | 1 | 1 |
| 1978 | Ron Guidry | AL | 9.1 | 11.5 | 1 | 2 |
| 1999 | Pedro Martinez | AL | 11.6 | 9.9 | 1 | 2 |
| 1995 | Greg Maddux | NL | 7.9 | 11.0 | 1 | 3 |
| 1970 | Bob Gibson | NL | 11.3 | 9.3 | 1 | 4 |
| 1985 | Dwight Gooden | NL | 9.7 | 13.6 | 1 | 4 |
| 1972 | Steve Carlton | NL | 11.6 | 13.5 | 1 | 5 |
| 1972 | Gaylord Perry | AL | 8.2 | 12.0 | 1 | 6 |
| 1995 | Randy Johnson | AL | 9.5 | 8.7 | 1 | 6 |
| 1986 | Mike Scott | NL | 8.4 | 9.2 | 1 | 10 |
| 1997 | Roger Clemens | AL | 10.8 | 12.6 | 1 | 10 |
| 1998 | Roger Clemens | AL | 8.2 | 8.5 | 1 | 11 |
| 1999 | Randy Johnson | NL | 9.0 | 10.3 | 1 | 15 |
| 1978 | Phil Niekro | NL | 9.1 | 8.1 | 6 | 17 |
| 2000 | Randy Johnson | NL | 9.6 | 8.5 | 1 | 17 |
| 1973 | Bert Blyleven | AL | 10.8 | 9.7 | 7 | 26 |
| 1969 | Bob Gibson | NL | 9.8 | 12.1 | NA | 30 |
| 2018 | Jacob deGrom | NL | 8.3 | 8.6 | ? | ? |
Data comes from our combined leaderboards, which contains pitcher batting value, too.
It’s worth noting that most of the players with top-five finishes played on postseason teams. Gibson’s Cardinals in 1970 finished 10 games below .500, Carlton’s Phillies finished in last place, and Gooden’s Mets missed the playoffs (but did win 98 games). Potentially aiding deGrom’s case this year is a lack of a clear frontrunner for MVP, as he’s more than two wins clear of Christian Yelich, the top position player at the moment. Here’s how the competition for the above pitchers looked in terms of a top position player with the gap column showing the difference between the pitcher’s WAR and the top position player.
| Year | Player | League | MVP | Top Position Player | WAR Position Player Gap | RA9WAR Position Player Gap |
|---|---|---|---|---|---|---|
| 1968 | Bob Gibson | NL | 1 | 7.1 | 2.0 | 6.9 |
| 2014 | Clayton Kershaw | NL | 1 | 7.4 | 0.7 | 0.9 |
| 1978 | Ron Guidry | AL | 2 | 7.7 | 1.4 | 3.8 |
| 1999 | Pedro Martinez | AL | 2 | 7.5 | 4.1 | 2.4 |
| 1995 | Greg Maddux | NL | 3 | 7.7 | 0.2 | 3.3 |
| 1970 | Bob Gibson | NL | 4 | 8.2 | 3.1 | 1.1 |
| 1985 | Dwight Gooden | NL | 4 | 7.8 | 1.9 | 5.8 |
| 1972 | Steve Carlton | NL | 5 | 9.2 | 2.4 | 4.3 |
| 1972 | Gaylord Perry | AL | 6 | 8.0 | 0.2 | 4.0 |
| 1995 | Randy Johnson | AL | 6 | 8.2 | 1.3 | 0.5 |
| 1986 | Mike Scott | NL | 10 | 6.2 | 2.2 | 3.0 |
| 1997 | Roger Clemens | AL | 10 | 9.0 | 1.8 | 3.6 |
| 1998 | Roger Clemens | AL | 11 | 7.9 | 0.3 | 0.6 |
| 1999 | Randy Johnson | NL | 15 | 7.8 | 1.2 | 2.5 |
| 1978 | Phil Niekro | NL | 17 | 6.8 | 2.3 | 1.3 |
| 2000 | Randy Johnson | NL | 17 | 8.3 | 1.3 | 0.2 |
| 1973 | Bert Blyleven | AL | 26 | 7.8 | 3.0 | 1.9 |
| 1969 | Bob Gibson | NL | 30 | 7.9 | 1.9 | 3.2 |
| 2018 | Jacob deGrom | NL | ? | 6.0 | 2.3 | 2.6 |
Data comes from our combined leaderboards, which contains pitcher batting value, too.
None of the five pitchers with top-three MVP finishes contended against a position player with an eight-win season. If there’s anything to differentiate the players who finished high in the MVP voting versus those who didn’t, it is the RA9/WAR. The top half of the table above has roughly double the difference of the bottom half. deGrom doesn’t quite have the difference of some of the better finishers, but the only player of recent vintage is Kershaw, and he was barely ahead of Andrew McCutchen. I’d also mention that Pedro Martinez had the most first-place votes in 1999, but Ivan Rodriguez managed to edge him out with a better overall performance among voters. There isn’t a direct analog above, but the closest case is probably Bob Gibson’s — and he had to contend with a great season from Johnny Bench.
While Yelich, Nolan Arenado, Javier Baez, Matt Carpenter, Freddie Freeman, Paul Goldschmidt, and others are having very good seasons, deGrom has been the best performer in the National League this year. If voters are looking for the league’s best player, they need to vote for a pitcher on a losing team with single-digit pitcher wins. It’s unusual, but this season presents an unusual set of circumstances, and those circumstances align with Jacob deGrom deserving MVP.
Craig Edwards can be found on twitter @craigjedwards.
I love deGrom and have him in multiple leagues. But in this man’s eyes, it’s down to Baez & Yelich. I have Baez in 1st and Yelich 2nd
It’s Yelich for me, then Carpenter. Baez is excellent, but I have a hard time with the idea of a player with a .329 OBP being the most valuable player in the league
It should be Yelich in the NL 100%.
He’s getting dinged by notoriously unreliable defensive metrics, when he has exhibited some amazing outfield defense this year. He has a 154 wRC+…if he was even a +0 right fielder on FG, he would be over 7 WAR.
Completely agree. Yelich is a plus defender and is unfairly getting dinged by suspect defensive metrics. He should be MVP this year unless he falls apart the last nine games.
Yelich has +1 UZR, +2 outs above average and +0.2 UZR.
When three different methods basically say the same thing, then I’d buy it.
How do you explain Yelich being dinged defensively this year? I’m not saying you’re wrong, but its an easier argument to follow if there’s an objective explanation for it.
I think the publicly available metrics are largely bunk (the adjustment made to them this year indicates they’re very much a work in progress). Plus, we have that “it takes three years to normalise” thing going on, and are judging someone on a year’s worth of figures.
Downvotes for you, sir!
No, just no. The flaw in your argument comes in the very beginning of the article when you talk about the “best player this year.” Best player should be irrelevant to the MVP discussion.
I’d say being the best player in the league is very relevant.
If you must, then go by most valuable. WAR, a stat that by definition judges value, has him as the most valuable player in the NL by a significant margin. The voting criteria clearly stats that a player need not come from a playoff team
It comes down to how one defines value. To me there is no value in missing the playoffs. Value comes from contributing to winning. You play to win the game. If you have a 5 for 5 performance with 6 rbis, that is nice, but if your team lost, then it ends up being empty stats. Similar to scoring 15 points off the bench in the 4th quarter of a blow out loss. Empty stats. They look good in the box sheet, but in the end, they didn’t accomplish the goal.
Well then Mike Trout should have 0 MVPs, despite being literally the most valuable player of the modern era
How would Trout have 0 MVPs? I don’t see how this line of thinking would impact his 2014 AL MVP when the Angels won 98 games and had the best record in baseball.
By extending that logic, if value comes only from winning, the MVP should only come from the World Series winning team that year, because every other team lost. A 12-WAR player on a team that misses the playoffs by 1 game because the closer blows a save on the last game of the year, you’re saying he should be eliminated from MVP contention because he has a bad teammate? In a sport where individual performance has less of an effect on game outcome than any other? Playing in a terrible division increases your MVP odds because you have a comparatively greater chance of making the playoffs?
Doesn’t MVP voting occur before the playoffs? How could the World Series influence voters?
In that case, MVP should always be on the team the best record in the LDS because playing the wildcard is more valuable than getting a wild card or winning the division but having to play the other division winner.
The MVP is voted on before the playoffs, I believe?
Going 5-5 is definitely *not* like scoring 15 points in the 4th quarter of a blowout. If you define value as how much an action contributes to a win – we can actually look at that! Looking at WPA, deGroom is ahead of the next closest position player, Goldy, 5.81 to 4.51.
Can you really compare position players to pitchers like that?
You can contribute to winning… but without some teammates doing their jobs, you are not going to be on the winning end of the score! Baseball is a team game.
Hypothetical best pitcher ever starts 33 games, pitches 33 perfect games. Each of those games are a win by a score of 1-0. The pitcher also contributes to every single one of those runs, either by scoring the run, driving in the run, or advancing the runner that scored with a sac bunt. His team finishes 33 and 129, missing the playoffs by the widest margin in history.
Is that player still not the MVP?
There’s nobody even remotely like that, so the point is moot. Besides, the pitcher would have to hit and pitch with one hand tied behind his back, at the very least…
And if you want to talk about WAR, I’d like to see a breakdown of player statistics in team wins and player statistics in team losses. Which players are actually performing at a higher level and it mattering for the team. If a team hypothetically went 0-162, no player on that team could say they contributed to a team win. That WAR stat is irrelevant if they do not Literally help the team win any games.
The MVP is an individual award. Teams get a World Series trophy
Yes, an individual award based on Value. Value is what you do for your team.
Precisely. A player who gets a 65 win team to 73 wins is just as valuable as a player who gets an 85 win team to 93 wins
Yeah, but I think what bluerum is getting at is those wins themselves aren’t as valuable. Its a bad argument for an individual award like MVP. There is that language in the voting rules about “most valuable to their team” though.
But that player can scarcely control the makeup of his team, so it seams unfair to penalize him for the bad luck of playing with inferior players. If you want to make an argument along these lines, maybe say it’s harder for a player to perform well on a good team than a bad team because of the pressure or stress that is attendant to having to perform at a high level for a contending team. It’s impossible to prove and will lead nowhere, but at least it’s not as easy to refute, so there’s that.
You go from saying it is how one defines value to acting like your definition is right and everyone else’s is wrong.
IMO value is determined void of team context. If every player had the same teammates, played in same parks, vs same level of comp, who is the best. That is my definition of value.
You’re basically saying any positive WAR contributor on a winning team is by default more valuable than a higher WAR contributor on losing team.
My question would be, if you switch Brock Holt for Mike Trout, do Red Sox win more games? If the presumed answer is yes, then Mike Trout is more valuable.
You’ve illustrated beautifully why a purely team-level results-based evaluation of individual performance is incredibly flawed.
Individual awards should not penalize a player for having bad teammates. A player who goes 5 for 5 with 5 home runs in a game that his team loses 6-5 could literally not have done anything more to help his team win, and yet you would seriously argue he provided no value in that game.
I’m attempting a reductio ad absurdum argument here, but I get the feeling that it won’t work on you because you don’t see that there are built-in limits to baseball regarding how much any one player can realistically contribute to their team (especially compared to, say, basketball)
While we’re at it, let’s include player salary, because other things being equal, a $25 million a year player is not as valuable as one in his pre-arb years. Let’s calculate exactly how much money a team gets for winning the WS, how much of that money is due to one particular player, then subtract his salary from that.
This being the internet, I better add that I’m being sarcastic.
I *just* did the math on correlation of fWAR/bWAR to actual wins this year.
R^2 of fWAR = 0.86
R^2 of bWAR = 0.87
Safe to say they are really really fucking relevant
A 0-162 team would possibly be fielded by a team of 0 or negative WAR players, though, like a bunch of Alcides Escobars, or whatever.
Whoo buddy, you are at the wrong place for this take
Trust me I know that. Every time I mention this type of argument the down votes come a calling. It is sad. To me value only comes from winning. You play to win, not compile major stats. That can happen along the way and help your team win yes. But if you don’t make the playoffs, then you are simply practicing to get better for next season. Every sport is all about making the playoffs. Yes, individual award, but a very different award than Most Outstanding Player. 1 for 4 with a solo home run that wins a game is much more important than 1 for 4 with a solo home run in a loss.
“I mention this type of argument the down votes come a calling. It is sad.”
I think what’s sad isn’t the downvotes but a) you knowing this isn’t the site to spew this nonsense yet continue to for whatever troll reasons and b) you don’t understand how wrong you are, not just in who should win MVP but in the basic concept that not making the playoffs has no value.
There are three types of teams who miss the playoffs:
1) tankers, who missing the playoffs has IMMENSE value to
2) teams on their way up, who care more about player development than the playoffs (so they’re OK with missing the playoffs)
3) teams trying to contend, where every win matters even if they end up missing out
All three find crazy value in their seasons despite not winning enough to make the playoffs. There’s no other way to put it than you’re completely wrong.
So I don’t know if this is you, but I find your attitude reminiscent of a really gross type of ultra-competitive, do-or-die machismo. That in a given season, if your favorite team doesn’t win the championship, then every game was wasted and worthless. That every effort, no matter how great, how driven, how passionate, has no value if the aim is not achieved. That the destination is everything, and the journey is nothing.
I don’t know if you subscribe to that exactly, but your argument points in that direction, and it makes me sad. What an ugly, Machiavellian outlook. At least when I go to a ballgame, I know that I’ll get some enjoyment out of it regardless of who wins the game.
That and from just a purely mathematical, logical standpoint, you still can’t coherently argue that a player who does plenty to help his team win provides no value just because an entirely different player makes a mistake. So you are wrong either way.
You are right about not taking any enjoyment from a game I go to and the team I root for loses. When that happens I walk away thinking I wasted my time and money. Growing up I learned to love baseball watching TBS and the Braves. From the first year I started following baseball, they made the playoffs every year. I learned anything less than that was not good enough. Missing the playoffs by 1 game is more painful than 40 games out and no chance. Missing the playoffs is a wasted year as a fan. There is value in building towards the future, but that is about talent development, the stats don’t really matter. Stats matter when they contribute to wins or when evaluating a player over the course of their career or for fantasy sports.
The down votes come a calling because your very fuzzy definition of value seems out of place in the context of team sports.
I’m confused why you think making the playoffs is what every sport is about rather than winning the title itself.
There is also the elephant in the room, which is that there just aren’t any super-valuable position players in the NL this year. They’re all in the American League. There is not a single player who (a) earned most of his value in the NL and (b) is in the Top 6 of the WAR leaderboards. Cain, Rendon, Baez, Goldschmidt, and Freeman are (at a minimum) 2.6 fWAR lower than deGrom. That’s an enormous distance. Yelich has a stronger case (he’s “only” 1.9 fWAR behind) but he’s still a long way back.
If, say, Francisco Lindor or even Matt Chapman played for an NL contender we might have a real argument about whether a guy on a terrible team outweighs their contributions. But I think that deGrom’s case is very strong empirically as it stands right now, if not with the voters who will punish him for being on a terrible team.
I think it is worth recognizing that WAR isn’t value – its one narrow-minded approximation. Lets look at those players at the top of the heap and think about if one doesn’t fit the WAR algorithm very well… that player is Javy Baez. He doesn’t walk, which crushes his WAR – but as we see regularly doesn’t actually make him worthless. He also plays all over the place defensively, which certainly doesn’t feed his WAR like it would if he had a cushier defensive assignment. His base-running also certainly is impossible to quantify. WAR overvalues some players and undervalues others – that is worth thinking about in any MVP discussion. deGrom’s WAR is basically pulled out of a hat – everyone understands that it is problematic to compare an every day player to a player that plays every 5 days.
Thanks for the rich discussion all. Currently 8 downvotes and no comments.
That’s because you seem to be confused about how linear combinations work.
Go ahead and explain.
I’m an art school drop out for whom baseball has been a life-long obsession, but let me take a stab at this, as advanced sports metrics are responsible for spawning my curiosity of mathematics and analytics.
In short, linear combinations take a series or set of terms , multiply them by some constant, then add the results. In the case of WAR in baseball, one arrives at WAR by summing a player’s defensive value, offensive value, and base running value , and those values are themselves the results of rather long linear equations.
Adjustments are made based on position and the individual formulas themselves may also include adjustments based on park, etc.
I may be a bit off here, so I would not be opposed to someone coming in and correcting my faux pas, but this is my basic lay understanding. In short, WAR, and other holistic metrics, take a bunch of factors, weigh them, adjust them, then combine them to try and assess an individual player’s contribution to a team.
This is why people love WAR, because it adds context and allows us to compare apples to apples, where we were not able to before (back ind the day, we would not necessarily see defense in box scores (errors, aside), or add the context of offensive production coming from players who play difficult defensive positions, etc.
Those weights and adjustments are arbitrary. Look no further than the ways that different organizations apply different weights and calculate their own WAR. There is also the supposed reality that teams have their own propriety adjustments and data. WAR attempts to make a science out of comparing apples to oranges.
This is something I think about a lot. WAR is great, I love it, but people here take it as divine truth and as if it didn’t have flaws. Pitching WAR has flaws, defense in yearly WAR has flaws, catcher framing, base running, etc.
One shouldn’t be looking at it as a divine truth, but as our best available approximations (because their are in fact multiple WARs).
A better WAR can be built. But the end of season is approaching fast and a decision needs to be made.
How do you make a call that isn’t a more flawed method?
Evaluate the players as individuals. I know it is not convenient and it is a lot to ask.
Call me a pessimist here, but I also think it’s worth recognizing that the stat being cited here most frequently (fWAR) is also the product of this site. So of course its writers will seek to convey that it is the singularly greatest statistic there is in displaying player value.
WAR includes contributions on defense and baserunning. https://www.fangraphs.com/library/misc/war/ I do agree that multi-positional value may not be fully incorporated.
Multiple positional value is certainly not accounted for. The more nuanced discussion is the negative effect of having to prepare for multiple positions or more accurately the inability to exclusively prepare for one.
“deGrom’s WAR is basically pulled out of a hat ”
You want serious responses to a post containing this statement?
Ok, but you can account for the difference in how an “everyday player” compares to a pitcher by simply calculating how much each of those players has the opportunity to affect the outcome of a game. A position player who starts gets like 4 to 5 plate appearances, per day and, depending on the position, touches the ball on defense maybe several times a day at varying degrees of difficulty and leverage (a catcher is obviously more important than a left fielder on defense, for example).
A starting pitcher may start only every 5 days, but may also face over 20 batters and throw over 90 pitches, if not 100, so they may actually have more of an opportunity to affect the outcome of a season by season’s end than a corner outfielder, for example.
How many ABs is an IP worth? 3.14? Do you think this number is set in stone or does it change? Does it vary by situation? Is it park adjusted? When did we arrive at this conclusion? Are we going to change it next year? I would say it is problematic.
Please explain how WAR isn’t value. Please explain how it’s a narrow-minded approximation. Obviously you completely understand how the stat is computed — tell us explicitly what it weighs improperly and tell us your deserving MVP candidates and what they have contributed to be the most valuable player in baseball.
I mean, I’m asking you honestly and seriously. Even though I am literally 100% positive you have no idea what you’re talking about.
WAR is absolutely an approximation. The equation will be different next year.
How many starters have won the MVP with 8 wins? That would be zero.
How many players have won MVPs on a losing team? There have been a couple but none since ARod in 2003 I believe. Before him you have to go back to Dawson in 1987. (Just going off the top of my head – correct me if I missed someone) Both were position players having monster years with no other strong candidates.
Then there is the real elephant in the room – the fallacy that WAR is anything remotely like an agreed upon stat. B-ref has Nola ahead of deGrom. Baseball Prospectus has deGrom BARELY ahead of Scherzer and he was trailing until Scherzer’s last start.
deGrom has had a fantastic year and I think he will finish either 1st or 2nd in the CY voting, but MVP? Don’t see it myself.
This is spot-on. How can WAR be the ultimate tiebreaker and arbiter if we can’t even decide what WAR is?
To me – and as a Cubs fan, I’m biased – the choice is clear: Baez. The Cubs are headed for the best record in the league despite down years from Bryant and Contreras and the disasters that were Darvish and Chatwood. Why? Baez, Baez, Baez. If that doesn’t check the boxes for MVP…
“I’m biased–the choice is clear!” The whole point is that it’s not clear. Being “biased” towards the Cubs, however, is a weaker methodology than being biased towards results.
You miss my point. I concede my bias as a Cubs fan. And with that said, Baez is, in my opinion, more valuable to his team than any other NL player in 2018. I understand that WAR suggests otherwise, and I reject the conclusion that the MVP should be the player with the highest WAR.
If not Baez, I see a much better case for Yelich, Goldschmidt, Carpenter, or Arenado than deGrom. And if you agree with THAT, then you, too, reject the automation connection between WAR and MVP.
I’m fine with deciding what the choice is. The choice is fWAR. bWAR is an overfitted model that uses some seriously flawed fielding metrics (DRS) to contaminate everything else, especially pitching.
Now you have no excuses.
I think its okay to take fWAR in context though. Yelich is probably a better defensive outfielder than he’s getting credit for (I’m not prepared to say that Schwarber is a better LF). Baez probably doesn’t get enough credit for both his tagging ability and defensive versatility at multiple premium positions. (though I’m not ready to give Baez credit for the bizarre errors he seems to be inducing).
The problem isn’t just that DRS is worse than UZR. I mean, it is; it is supposedly more precise, but instead just produces totally unrealistic estimates because it models the error. The problem is that then all that error gets poured into the pitching bWAR when they do defensive adjustments on runs allowed. It’s usually better than RA9, but not always.
I think there are some weak spots with fWAR, and Baez’s positional versatility probably is one of the things it underrates. But not enough to make that big of a difference.
Isn’t your argument the same argument sabermetric types have been making against traditional stats? “Your stats are bad, my stats are good”
Oh, and the problems with WAR go way beyond just whose defensive stats you use. I would also note that now that Statcast is around, we have a new set of defensive metrics to consider. Are those better than UZR or DRS? What does that say about fWAR or bWAR?
There is no consensus on how to calculate WAR and the over-reliance I see on it is ridiculous.
But those stats ARE bad and our stats ARE good.
Our methods are always going to be handicapped by the data available to make them. Statcast’s ability to collect superior data should allow for better metrics on the defensive side of the ball, where there is pretty much agreement that the numbers aren’t great.
Also, there seems to be a persistent problem in some of these comments that things like team success and Baez’s ability to play multiple positions, while relevant, somehow make up for the numbers we can quantify. Those are little things that could put Baez over the top in a close race, but it’s not a close race. Baez has had a really good season, but even if you use bWAR’s wack defensive adjustments instead of FIP, deGrom still holds the same edge in WAR over Baez (2.7 instead of 2.6).
Could you elaborate on the flaws in DRS that cause you to be so dismissive of rWAR? According to Fangraphs’ own glossary, DRS and UZR are very similar systems with few significant disagreements. rWAR, like fWAR, is the product of years upon years of the entire sabermetric community’s research. They are both defensible implementations of the WAR framework according to the man who came up with much of that framework, Tom Tango. So what are Tom Tango, Sean Smith, Sean Forman, John Dewan, and the rest missing here that sadtrombone is seeing?
I too am curious about this
It was actually Mitchel Lichtman who convinced me. It is possible that the methods on DRS have been changed so that these are no longer issues, but I was very convinced by MGL’s argument here:
http://www.insidethebook.com/ee/index.php/site/comments/john_dewan_and_research_assistant_speak/#3\
The key points are two things:
1) The “buckets”–MGL is abolutely right. You can’t break up the buckets into stuff that small and not have enormous error problems, for reasons that MGL explains quite cogently. And Tango concurs below.
2) Assigning debits/credits–DRS is aggressive at assigning credits to someone or against them.
The result is that you get some truly wack values with DRS. In some cases, it is more accurate. In other cases, you get values so extreme they can’t be real. Both Harrison Bader and Kolten Wong have saved 21 runs in 800+ innings. Nick Ahmed has a wRC+ of 86 at Fangraphs, but has put up 3.8 bWAR this year. No doubt, these are good players, but that’s insane.
But that’s not the real problem–the real problem is then they plow it right back into the defensive adjustments used for pitchers. Neither UZR nor DRS is accurate enough for that. So you wind up contaminating the pitcher WAR with a measurement that just isn’t very accurate.
I’m certainly nowhere near as competent as Tom Tango or any of the other guys at developing ways of measuring baseball stuff, but I also don’t think that there’s as much consensus as you suggest. I’m of the mind that if you really want to do this “right” you need a factor analysis of a few different systems to extract a latent variable. Hopefully, the errors in the systems aren’t correlated with each other and you can pull out something approximating the “true” value.
I’m sorry guys. I posted something, it said it’s awaiting moderation, and it must not have made it through.
Short version of what I wrote: The problem is mostly with using defensive adjustments to account for pitcher performance instead of using FIP. Neither UZR nor DRS is accurate enough for this purpose, which is bWAR is not predictive. Instead, the model is overfitted to compensate for DRS’s problems (take a look at the formula for it sometime)–trained for maximum explanation of prior years than for out of sample prediction.
That said, I also think DRS has some problems, which was discussed on The Book Blog. I think the link is what got it flagged for moderation, but it’s the link at the bottom of the UZR vs. DRS page. And the consensus on UZR vs. DRS is much more split.
Thanks for the reply. As I understand it DRS and UZR have limited use for a single season for a single player. However once you have three years of data for a single player then it starts to stabilize. So I don’t see why using advanced defensive metrics at the _team_ level would be “polluting” rWAR’s pitching measurements. There should be more than enough signal there to overcome the noise. Especially since issues at the individual level like positioning and the range of a neighboring fielder would seem to come out in the wash once you add up the entire team. At the very least it’s no bigger a problem than fWAR completely ignoring contact management skills.
When you say it’s not predictive, do you basically just mean that FIP is predictive whereas RA9 is not? That is true enough, but I’m not sure how it’s an indictment of rWAR since WAR itself is not really meant to be a predictive stat. It’s meant to measure past true talent. Really the major difference between the two WARs is rWAR assigns all sequencing luck/value to the pitcher whereas fWAR assigns none of it to the pitcher. Manifestly, neither extreme approach is correct, which is why I’m glad that both models exist. I much prefer a blend of the two to either one individually.
No, it’s that bWAR is not predictive of out-of-sample events (e.g., years the data wasn’t trained on). And that lack of prediction is because it isn’t capturing the phenomena very well…predictive validity is the only validity we have in this case.
RA9 isn’t predictive, but we know that already.
I personally think that if you took whatever BP is doing these days, fWAR and bWAR and extracted a latent variable out of it, you’d probably do better than just one. It usually does in those situations. I just think bWAR is probably the weakest individually of the three; the cFIP method seems to work better than bWAR.
Dude. No. While there are several WAR formulas, we can definitely decide on what WAR is. Look into it. It’s pretty interesting.
Errata: I apparently have a short memory and forgot that Stanton won the MVP last year from a losing team. But that fits the model – even though a few other guys had great years, he had a mammoth year and probably more importantly, nobody from the top 3 teams had a big year.
I actually don’t think DeGrom is the MVP, and would not vote for him, but these conversations are important because they have the potential for us to re-contextualize traditional counting stats, like player wins, and to (once again) debate on a player’s worthiness based on team success. In other words, just because stuff like this is historically rare, it does not mean that we should not change the way we think about these things, given the fact that we have far more data now.
And speaking of 1987 and Dawson, if I had a time machine and could steal a vote from a haggard old sports writer, I would vote for Dale Murphy, Tony Gwynn or Jack Clark over Dawson in a heartbeat. I think Eric Davis also had a really good year, too.
When DeGrom doesn’t win (he won’t), we will be left to wonder how much of his perhaps unfair defeat came from each of these three categories:
1. being a pitcher
2. losing team
3. single digit wins
Without doing the research, I’m going to assume that the last one, which wasn’t really covered in this post, is also a significant factor. I’d assume the pitchers that were top 5 MVP were at or near the league lead in wins. I wonder how much voters have moved away from wins? I think some voters may have moved further away from looking at wins for Cy Young consideration than MVP, as silly as that is.
He doesn’t even have a clear case for the Cy Young. Go ahead and pull up two tabs and look at Max Scherzer. He hasn’t even been head and shoulders above the rest of the SP field unless you are blind to just looking at select metrics. Baseball-reference has Nola as #1 in pitcher WAR. deGrom only has one clear claim to being the Cy and that is his run prevention, which is questionable in the first place.
The problem I’ve always had with giving a pitcher MVP is that it means that position players get nothing. Either create a new position player award that has the status of a Cy Young, so the MVP can go to either pitcher or position player, or make a rule that pitchers aren’t eligible for MVP.
The Hank Aaron award exists… and clearly should be more publicized. Or perhaps more succinctly, it should be adopted by the BBWAA and made a more official award. That may go a long way to stopping this position player vs. pitcher argument for MVP.
Isn’t the Hank Aaron award technically for “best hitter” rather than “best position player?” I think it would be a good idea to change it to best position player and do what you suggest, but I think there is that one extra step to it.
The Hank Aaron award is given to the best offensive player in his respective league. The MVP award is supposed to recognize the player whose total game is most valuable to his team.
It’s for this same reason I dislike the MVP award going to a pitcher. You may look at a pitcher’s WAR as being the highest in his league, but you have to remember this WAR has an entirely different formula than for a position player. Even if we grant that pitcher WAR accurately provides the win value from pitching, we calculate WAR for pitchers without accounting for their offensive production (or lack thereof), while we consider offense AND defense in WAR for position players. They’re two different measures and comparing them is erroneous.
The Aaron award has never had status remotely close to MVP. I’m not sure how one increases the status of an award. Plus, it’s not for all-around excellence, it’s for offense.
A possible alternative is to make pitchers ineligible for MVP, but create a new award that could go to either pitcher or position player. The novelty of such an award–the first in which pitchers and position players would compete for on more or less equal terms–might help in enhancing its status, so that winning it would have real meaning. One could also stipulate that the winner had to come from the top 3 vote-getters in the Cy Young and MVP., to emphasize that the winner was a CY or MVP level candidate.
Usually, the winner would also be the CY or MVP winner, but there would be some distinction. A pitcher might get credit for his defense, and in the NL, his bat, as well as his arm. A position player might get more credit for his defense than is currently the case. E.g., if Martinez were to win the triple crown this year–he probably won’t–he might well win the MVP. I definitely think he shouldn’t, but if he did, this new award could help redress that problem, by giving the new award to Betts or Trout. It could also be given to a player who was arguably more valuable, but lost MVP votes because he didn’t play for a contender.
Not sure what one would call it. Best all-around player?
Why does a position player need an award that constitutes both offense and defense? There is already defensive awards (gold glove by position and platinum glove for overall best defender), and the Cy Young doesn’t worry about pitcher defense. The argument is that there is no high-profile offensive award for all around offensive value (battle titles and silver slugger awards don’t really count)… thus, the Hank Aaron award would be good to be given an elevated status. Then, we are free to use the MVP as a single award for most valuable player at any position, as all facets of the game would have their own award(s).
In general, I think MLB needs to standardize its awards so that all facets of the game and all awards have similar visibility and weight. Right now the BBWAA stuff gets all the publicity but we already have other awards to address the deficiencies.
Because great offensive value can be worth little if the player can’t play defense. Khris Davis has a 134 wRC+ in 616 plate appearances but only 2.4 fWAR because his defense is simply that bad.
For context, Kyle Schwarber has only a 115 wRC+ in 478 plate appearances, but 3.0 fWAR, because his defense has been simply that much better than Davis’ at the same position.
A position player award that constitutes both offense and defense is better because it actually captures how good the player is overall instead of just half of the game.
The Hank Aaron Award is more reason, which I bring up each year why the word “value” is so important to the MVP. The best player is not what the MVP is for.
Or not… there are lots of awards that nobody cares about. I am good with awarding it to a position player unless a SP does something really special and this isn’t that.
You could also just rebrand the Cy Young Award as the “Cy Young Most Valuable Pitcher Award”, giving it equal footing with a rebranded “Mike Trout Most Valuable Position Player Award.”
Of course, then you are possibly screwing Ohtani (not this year, but possibly in future years).
I would be interested in seeing the gaps, in addition to those between the pitcher’s WAR and the best position player’s, between the pitcher and the second-best hurler. Kershaw and Verlander were both unanimous Cy Young winners in their MVP years, but you could make an entirely legitimate case for voting for Scherzer over DeGrom this year, or Nola, if you prefer bWAR. If it is debatable whether you’re even the best pitcher in your league, you’re not going to get much traction for an MVP bid with the slant towards position players already baked in.
This.
I’ve felt when the discussion has come up for a pitcher MVP (Verlander and Kershaw) it was because they so lapped the cy young field, it seemed appropriate to include them in the MVP discussion.
I think you might be right, but this is a pretty silly rationale for including or not including a pitcher in MVP discussion.
Was it one? I assumed it wasn’t meant as an argument for or against, but rather as an additional factor that might explain voter behavior (irrational as it probably is). I think it’s probably true that “waaaay better than the second-best guy” explains a significant chunk of voter psychology.
I was not claiming that the commenter was making this argument. It did not seem as though they were, to me. I was just saying that those that do make this argument are making a silly argument.
Ah. In that case I think we all agree!
The below opinion hinges on the assumption that there are 2 prestigious awards in baseball, MVP and Cy Young.
My opinion on Pitchers winning the MVP is “no” until the Hank Aaron award attains the same level of prestige that the MVP and Cy Young get. I don’t think it’s fair to consider pitchers for MVP when position players can’t get Cy Young. If the Hank Aaron award had the same “ring to it” that Cy Young and MVP had, then you could have “best hitter’, “best pitcher” and the MVP would float more evenly between the two. I’d feel the same about Platinum Glove if I had more faith in the Gold Gloves in general.
i think about the losing team angle- have to go back to just last year where Stanton/Votto were 1-2 in MVP voting despite being on losing teams….
I respect this piece very much both as an act of contrarian trolling and as a commentary on the NL MVP race this year — which is going to be a truly chaotic, wide-open field. Of course we’re all pretty sure this will never happen (and I’d be interested in modeling voter behavior to predict what will), but it’d be very interesting to live in a world where the voters thought this way.
Every year we come across the argument of what “most valuable player” entails… is it value created, value replaced, value translated to results?
Looking at the replaced and results criteria, DeGrom fares poorly. First, the Mets could have saved his arm and some valuable innings this year, as they were out of contention, putting replacement level pitching in for 2 innings per start, with minimal damage to their bottom line. Second, those extra high WAR innings he pitched garnered little impact in the W/L columns for the Mets as they flat out refused to score.
The argument for DeGrom is strictly value created. Given that he leads WPA as a statistical outlier, and is currently 3rd in WPA/LI is evidence that in the games he pitched, he was the primary reason they were even competitive. The higher leverage factor means that he was leaving close games (or not getting much offensive help). He has far and away been the best pitcher, with Scherzer and Nola regularly dueling for 2nd place votes and Patrick Corbin a darkhorse. It just isn’t enough to get him the MVP on a depressing Mets team. The study into the park effects of Citi Field don’t help his case either (not saying it should hurt it, but having this kind of result in a neutral park would be more easily explained.
He should be walking with the Cy Young hardware, but barring odd circumstances, will be 4th-6th behind Goldschmidt, Yelich, Baez, Cain and possibly Carpenter and Freeman. My guess is Yelich/Cain split enough votes to fall behind Baez/Goldschmidt, with DeGrom falling between Yelich/Cain.
In 1972 Steve Carlton may have been the most valuable player ever. With him starting, the Phillies were 29-12 (.707). With anybody else starting, they were 30-85 (.261).
Is there an error in this, or did they really only play 156 games?
It’s correct. THere was a strike in spring training in 1972 and the season started late so NL teams played between 153 and 156 games. In the AL it actually decided the playoffs since the Red Sox played one less game than the Tigers and lost out by half a game (imagine that today).
Was actually a pretty significant strike since the owners agreed to add salary arbitration
At Baseball-reference, deGrom doesn’t even lead pitchers in WAR – that would be Aaron Nola. Don’t you think that it is pretty problematic that much or deGrom’s value is coming from runs allowed? Personally, I would award the Cy Young to Max Scherzer, who has the better WHIP, but the worse ERA. Does anyone really think ERA is a great metric? By looking at the lines, it is not clear to me why deGrom is better. Scherzer has the better K rate, and they are are really close in BB rate – I thought we were trending toward things pitchers can control like Ks and walks? To me it looks like Scherzer is getting penalized for having a lower BABIP and the inflated HR rate and that is the source of the WAR differential. So you have one guy looking better in K/BB and the other benefiting from some arbitrary conclusions drawn from batted balls. I think they are close, but there is no way that a debatable Cy candidate is an MVP.
Future HOFer and 3X Cy Young winner, Max Scherzer, is on pace for a new career high in Ks, K rate, the second lowest WHIP of his career, and the lowest ERA and here we are talking about deGrom. This has potential to be one of those years where we look back and ask, why in the hell did we not hand the award to Mad Max? It gets boring to talk about the same guy but he doesn’t seem to be part of this conversation and he certainly should be. Rather than get uber-excited for a changing of the guard, why not celebrate the future Hall of Famer? This is exactly the type of season that puts Max in the conversation for one of the all time greats… or not…
1 problem is that you can also control homers and Max has given up double per 9 innings vs deGrom…..
Also, wouldn’t say we’ve exactly been trending towards things like K’s. Had Kershaw in 2015 and Sale in 2017 strike out 300+ batters with neither one winning the Cy Young.
deGrom has almost as big of a lead in FIP as he does in ERA. and he has pretty big lead in xFIP as well. 0.41.
Even if deGrom goes out and is an awful pitcher from 2019 on, and Scherzer dominates going forward- it just doesn’t matter. This is only about 2018. Period, the end. Your prior career should have NOTHING to do with it.
I hear you. I disagree with premise and I always have, but I understand that I am in a minority. Pitchers don’t control HR absolutely – any pitch can be hit for a HR. Some have problems and some are great at it, but most are pretty average – isn’t that how we view BABIP? Strange that we haven’t figure that out yet. Any FB down the middle can be taken for strike 3. It seems really dumb to me that the smallest batted ball outcome has so much weight. One bad night or a few trips to Coors can go a long way towards inflating a HR total. Look at the historic numbers – with deGrom there is reason to think that he doesn’t possess the ability to keep the ball in the park much better than Scherzer – last year he was worse. In all likelihood, it is mostly luck – when you are talking about 21 HR over 200+ innings a few outings go a long ways. I think everything you point out is just derivative of batted ball distribution – you cite multiple metrics, but they all have the same inputs.
I think you are wrong about prior career. If this was Max’ first dominant season, he would be getting more arguments for winning. There is absolutely incentive to push for a guy that has never won it (bold headlines, I said is first or loudest, yada yada, yada) and great players get snubbed in the middle of historic runs all the time – that is across sports. See Johan Santana for a good example. You misunderstand my point regarding careers completely. I am pointing out historic precedents supported by facts, you are making a valid point that has nothing to do with my argument. Its OK if we disagree – I just see Max getting shafted and want to try to raise some awareness. deGrom is going to be just fine without my support.
actually, Scherzer for career has given up 1 hr/9ip. deGrom is at 0.77 per 9. That’s a pretty fair difference….
Another difference with the 2 is that deGrom gives up fewer fly balls- right now it’s deGrom at 32.6% and Scherzer at 46.8%. So even if they had the same HR supression skills with fly balls- the fact that Scherzer gives up 14.2% more fly balls would mean he’d be giving up more homers even with that.
I think you are wrong about prior career. I mean just look at last year with Sale/Kluber. I think the voters do a pretty good job just looking at the season without looking at prior career.
That is heavily weighted by this season. I am not sure where that rate sits if you throw out this season… Go ahead and look at last year, where it was 1.25. The rates are anything but steady… if you actually look at year over year you can see the volatility. Scherzer’s worst HR rate was well before deGrom’s MLB debut which figures into his career average. I would bet you that over time his HR rate will regress upwards. I’m not sure that we aren’t celebrating fluky outcomes. deGrom really hasn’t pitched that many innings in his career. Probably also worth pointing out that Scherzer is a FB guy and gives up a few more HR (literally a few), but that actually makes his low BABIP more sustainable. As a GB guy, deGrom gives up more hits but less HR – I think it all adds up to a wash or at least close to it. Funny to me that hits are not attributed to the pitcher but HR are. That is how deGrom is ahead of Scherzer and its debatable. I am cool with you and other readers thinking that deGrom is better, but I don’t think the overarching analysis is very complete – it is based on very select stats.
What do you mean about Sale and Kluber? Sometimes the established guys win, but they also get passed over more than they should. You don’t have to look further than the AL to see Sale getting passed by Snell in all the headlines. Nobody is writing about the guys to win a second, third or fourth – all the pushes are for the new guys. Just look around you and you will see it if you want to. I don’t get it obviously! Then again, I am not writing for clicks.
Sale is getting passed by Snell in all the headlines(both of them haven’t won before) because Snell has 19 innings more than Sale and that number will likely increase….. That’s got nothing to do with Sale’s career, but rather that he won’t qualify for the ERA title, and a lot of voters see that as a complete DQ.
Sometimes the newer guys get more press because the newer guys just have been better. .
I would say that Kershaw’s reputation hasn’t gotten hurt even though he “only” has 3 Cy Youngs.
As far as HR rate last year was the only year in deGrom’s career that he had a hr/9 over 1. 6 of the 9 full years for Scherzer he’s had a hr/9 over 1.
I’d also say that extra base hits- which I feel a pitcher controls far more than 0- Scherzer 34 2b/ 5 3b while deGrom has only 27 2b/ 3 3b.
So, let me get this straight… deGrom is MVP because of his WAR lead, but Snell’s WAR is irrelevant to the discussion. That’s kind of my point here. This crowd likes to attempt to deal in absolutes, but the application of the metrics is as biased as any fan from any other generation. Isn’t it at least a little funny that deGrom’s case revolves around ERA and Snell’s is about Wins lol.
Snell’s case is as much about the ERA as it is the wins. Oh and it’s also the innings, something that Sale does not have at all.
As long as they call it mvp i think winning matters. You don’t give the world series mvp to a guy on the losing team. But there’s always been an inconsistency with the mvp and cy young in that the latter is clearly perceived as strictly a best pitcher award while the mvp has those vague connotations of “value.” And as long as there is a CY I’m in the keep pitchers out camp. At this point there isnt even consensus on Degrom winning cy young (i think he should as of today but its still too close). Its a fun argument for his mvp worth but one no writers who have votes will be having.
I’m the most statistically minded fan you’ll find, but every year I still have a very hard time convincing myself that a player from a non-playoff team (or worse, an awful team) can win the MVP unless there are no other viable candidates remotely close to him.
I know it’s a very unpopular opinion on Fangraphs, and I’d love to have a real conversation about it. I’m 100% open to having my mind changed.
I believe you can look at value in (at least) three ways:
1) Context-neutral, statistical performance (e.g. WAR)
2) Context-sensitive performance (i.e. direct contribution to victories) – e.g. WPA
3) “Where would the player’s team be without him, and how does that differ from where they are now?” – basically #2, but over a full season instead of split up and totaled by game.
In the MVP discussion, I keep going back to #3, and I can’t vote for DeGrom because of it (where are they with DeGrom? horrible. Where would they be without him? horrible.)
A few questions which I think can be instructive that I hope people will answer honestly. I’m sure you can guess my answers, but I am genuinely interested in hearing other viewpoints:
1) Should the player with the highest WAR win the World Series MVP, even if he’s on the losing team?
2) If the winning World Series team had the following two players, who would you vote for MVP?
A – starting pitcher who had 3 starts, pitched 3 complete games, gave up one run in each, and lost each game 1-0.
B – hitter who went 4 for 30 with no walks, 26 strikeouts, and 4 walk-off hits (say, 2 homers, a single, and a double).
3) A team that went 50-112 had a player who hit .300 with 80 HR and led the league in WAR, but every home run was a solo shot in the 9th inning with his team down 10-0. MVP or no?
When it comes to 3) would you use the same logic to not vote for players like ’16 Bryant or ’17 Altuve since their respective teams would have comfortably made it into the playoffs anyways?
Great question, and I’ll be honest, it’s not as cut-and-dry as that. It’s more like a player gets a bonus for his team making the playoffs, and another bonus for him being the “difference-maker” in his team making the playoffs or not. In other words, if the objectively better player (e.g. Bryant) was significantly better, his team being “too good” wouldn’t disqualify him.
In these specific examples, yes – I probably would have voted for Seager in 2016 because his WAR was close enough to Bryant’s and without him, the Dodgers are sitting home.
In 2017, from a strict “which player made the difference between his team making the playoffs or not” standpoint, Dozier is the only possible choice. But Altuve’s dominance over him is way too much to overcome.
Thank you. Someone else with some common sense on here.
You’re welcome, but I wouldn’t go as far as saying it’s common sense, but just a different reasonable way at looking at it. I once heard this argument:
You and your friend both need to buy something really important (say, medicine for your sick kids). It costs $25.
You have a $10 bill and 4 $5 bills and buy the medicine.
Your friend has a $20 bill and can’t buy the medicine.
Which bill was the most valuable?
It’s tough to argue that objectively, in a vacuum, the $20 is not the most valuable. It’s $20. All the other bills are worth half of it or less.
But I’d argue that value is much more meaningful when it’s given context. Here, the context is getting that medicine, and the argument that the $10 bill is thus the most valuable is completely reasonable as well.
Further, I’d argue that an analogous context exists in a given baseball season – specifically that there is a goal of making it to the playoffs.
So it really comes down to how you define value, and if you look at it in a vacuum or with context. That $20 bill is clearly the most valuable in a vacuum, and if you were starting from scratch the next day, trying to find enough money for the medicine, of course you’d want to start with the $20 bill over any of the others. But given the context of what has already happened, saying that the bill (or player) that made the difference towards the goal is the most valuable is perfectly reasonable as well (and, I’d argue, *more* correct).
Instead, the medicine only costs $20, so you both buy the medicine, which is a liquid. At home, when pouring the dosage, your friend drops the bottle, spilling the medicine everywhere, preventing his kid from getting the benefit of the medicine.
Is the $10 bill more valuable simply because the $20 bill had a clumsy teammate?
I”ll be honest too, I think a team making the playoffs can be used as a contributor factor to break a near tie, but I think it is given to much weight, and further, that many people apply an analysis of “best player on a playoff team” to the MVP award.
This year, Trout and Mookie are neck and neck, Trout has gotten on base a little more and hit the ball a little harder, but Mookie is credited with being a better defender. In that situation, I agree Mookie is the MVP.
But imagine by some fluke, the Brewers collapse and lose the wild card to the Cards and Rockies. Does that disqualify Yelich and force us to choose between Baez, Freeman and Carpenter? There are far too many people that would answer that question yes.
Yeah – I feel the same about the AL, just that I give more weight to the player’s team making the playoffs. I agree that Betts should be the MVP, but I think it’s a no-brainer over Trout. If you wanted to take my views to the extreme though, you’d say that Chapman should be the MVP because he’s the only AL player whose team goes from making the playoffs to not if he is removed (though depending on how things shake out out, it’s possible Bregman could be the difference between winning the division and not making the playoffs for the Astros).
Good question about the Brewers. Again, this is more of a gut feel type of answer, but I think in that case, maybe Yelich gets some “credit” for Milwaukee being in position for a playoff spot this late in the season (assuming he’s not the one who causes them to collapse). But in terms of WAR, Baez isn’t very far behind. Honestly I still could go with Baez in that case, but I wouldn’t argue too much with Yelich.
Also it’s possible the Baez could be the difference between the Cubs winning the division and not making the playoffs. If that ends up being the case, I’d have to go with him in all likelihood.
And yes, it does feel wrong that the Rockies winning or losing one or two games after they are eliminated (i.e. putting them in or out of the range of Baez’s WAR) would decide the MVP. So I don’t know how I feel about that. The analytical part of me says “well, that’s what the numbers say”, while the logical part of me says “that’s ridiculous.”
I agree on the comments about Bregman. As of now I have him at the top of the AL list because he seems to be the most consistent reason that the Astros will be able to win the division and hold on to the lead over the A’s. While the Red Sox and Indians have players who have had more dominant seasons, they also have much larger playoff cushions.
I think it’s reasonable to include team success as part of it, as per #3. Depending on how much you weight that, you could put Yelich above deGrom. But even then, it’s hard to make a case when deGrom is about 2 wins above him. That’s a big difference! You’d have to weight team performance fairly heavily.
You’re conflating position player wins with pitcher wins. They aren’t the same.
Do you mean that position player WAR is different from pitcher WAR? Because that’s what I assumed he was comparing. Or did you assume he was comparing position player WAR with literal pitcher wins (as in wins and losses)?
“it’s hard to make a case (for Yelich) when deGrom is about 2 wins above him. “
I know the words he used when he compared Yelich and deGrom. My claim is that he is comparing Yelich’s WAR with deGrom’s WAR, and thus it’s a valid comparison. Do you think otherwise?
From everything I’ve read, I don’t think it’s valid to compare pitcher WAR with position player WAR, and that’s just based on reading how others feel about them. Pitcher WAR sounds much sketchier.
Ok, that was my question – I wasn’t sure if you were comparing hitter WAR with pitcher WAR, or comparing hitter WAR with pitcher *wins* (which of course would be ridiculous).
1. No, and also no if he is on the winning team. I wouldn’t look at WAR for World Series MVP. So really my answer is “not necessarily.”
2. I’ll answer the second question. I would vote for player A. Player B will be remembered forever, anyway, despite the fact that he sucked. No reason to throw the MVP on top of that.
3. This is the most ridiculous of your questions, but I would say no, However, this is just a preposterous hypothetical.
Of course #3 is a ridiculous question. But clearly you get the idea.
For #1, is there something different about selecting the MVP of a short series vs the full season (in terms of evaulating players using WAR)?
What the pitcher I described in #2 played for the losing team? Would you vote for him then also?
Yes, there is a difference in evaluating players for a short series. I’d feel much more comfortable selecting a “Playoff MVP” from a team that lost the World Series than I would a World Series MVP.
First off, I watched every second of the World Series that I am voting on, and I watched it within the past 6-10 days. If WAR were even calculated, it would probably be very close between the top few players. If you are looking at one guy who has 0.9 WAR and another has 0.7 WAR, what does that tell you? Not enough. All the players are playing on the same field in the same environment. I think I can judge who was the best, who deserves the vote, without falling back on the WAR model.
And I would vote for that pitcher if he were on the losing team, if it is between him and Mr. 26 strikeouts. But I’d also consider other players. But Mr. 26 strikeouts isn’t getting my vote, no matter what.
That’s fair, it’s great to hear others’ opinions.
Maybe World Series WAR wasn’t the best description. What I was getting at was “would you vote for the guy who was clearly the best, statistically, regardless of context?” Seems like the answer is yes, in your opinion. So is it correct to say that you would also treat the World Series MVP award as the “best performance in the World Series” award?
Personally, I would give weight to playing on a bad team. Not having any support makes the job harder. Good luck quantifying that though! There is also the part where players on good teams already get to participate in a lot of exclusive contests like playoff series MVP and rings. I am not saying to penalize players on competing teams, but I like the idea of giving credit to a player that does it without any protection… like Trout. One thing is for sure, nobody is ever going to agree on winning/MVPs – this discussion has been had a million times and will be had a billion more.
1) – No. WAR is an approximation – go ahead use some judgement. Maybe maybe not.
2) A wins. Walkoffs are silly. So a go ahead hit in the 9th is more important if you are the home team? I am one of those folks that thinks a run in the 5th counts for as many runs as one scored in the 9th. I think blowout runs can be discounted, but in a close game runs are runs. MVPs of playoff series are different than the regular season as an SP can very well appear in half the games or close to it. A lot more than 1 out of 5.
3)Of course he wins. I don’t see how you could actually dispute that. If that guy plays for a contender, he has 20 walkoff HR lol. How funny is it that Bonds was actually better than that fictional season? Go ahead and trade 7 HR for about .30 pts of AVG and add 177BB – I’ll take it! So I guess, no that player loses to peak Barry Bonds, but in a regular season he wins unanimously.
No one knows what the MVP is, what the criteria for winning it is. How can we hope to evaluate deGrom’s qualifications? I love debate, but this seems like an exercise in futility. Me, I’m still hung up on Ernie Banks beating out Hank Aaron in 1959.
“It’s possible that some voters aren’t using entirely sound logic to arrive at their conclusions”
Ya don’t say?!
Considering in 2016 Bumgarner finished 1 point ahead of Jeurys Familia!! in voting (16th and 17th respectively)
Bumgarner – 226.2 IP, 15-9 2.74 ERA
Familia – 77.2 IP, 51 saves 2.55 ERA
Great season for Familia but come on.
In a year in which there was a truly standout position player in the National League, I would likely oppose giving the NL MVP to a pitcher, even if he was leading the league in fWAR. And while I do not wish to diminish the quality of the seasons put in by Yelich, Cain, and Baez, it just feels that Degrom has lapped the National League this year.
I doubt his astonishing performance this year will be able to outweigh the fact that Degrom’s team was a full blown dumpster fire in the minds of most of the voting BWAA members. But, If I had a ballot, Degrom would get my vote for Cy Young and MVP this year.
It always ticks me off to see a pitcher take the MVP.
Why? If a starter went 35-0 with 35 shutouts and won the MVP, would that tick you off?
I was talking about in reality. Give the pitcher a Cy. That’s it.
It might be somewhat deserving, but it has ZERO chance of even coming CLOSE to happening. Pitchers on losing teams NEVER win MVPs (since the Cy Young award was established). I’m not sure one has ever come close.
“Even if they’re employing the wrong process, however, they’re still usually arriving at the correct result: pitchers simply aren’t the best players in their leagues all that often. Position players make an impact at the plate and on defense. They just have more opportunities to create value. Their roles give them a competitive advantage.”
Isn’t this exactly the logic of old-school voters? You took their reasoning of “pitchers don’t play everyday”, and added the obvious, implicit specifics of what that means.
Maybe your process needs to be updated with the insulting and stating the obvious aspects removed.