The Case for Justin Verlander
The American League Most Valuable Player Award is going to be announced at 2:00 pm today, and there’s a pretty good chance that Justin Verlander is going to become just the 10th player to win both the Cy Young and MVP in the same year. Unfortunately, the reasons that Verlander is probably the favorite to win the award aren’t great – the two best position players both played for teams who failed to make the playoffs, and voters are historically preferential to players who came from teams who played in October. He’s also a stronger candidate to actually win the award because of his 24 victories, and it is too bad that pitcher wins are still hanging on to some influence.
However, you don’t have to buy into the traditional criteria for MVP voting to cast a ballot in favor of Justin Verlander. Even without some tortured logic about rewarding a player for the performance of his teammates, a strong case can be made for Verlander as the most valuable player in the American League in 2011.
Let’s start off with the elephant in the room: with how we calculate WAR here on FanGraphs, Verlander’s +7.0 season is pretty far behind both Jacoby Ellsbury (+9.4) and Jose Bautista (+8.3), so an argument for Verlander seems to require a rejection of FanGraphs WAR, right? I would suggest that it does not. We think WAR is a great tool that has helped push forward the understanding of the relative value of players with very different skills, but we’ve never suggested that it’s perfect or entirely comprehensive.
There are areas of the game that no one has figured out how to quantify with great precision yet, and players who excel in that area are likely undervalued by the current inputs of WAR. For instance, catcher defense isn’t something that we have figured out, and there’s a good chance that Yadier Molina is more valuable than WAR gives him credit for. That doesn’t make WAR useless, just incomplete, and it means that when discussing the value of a player who may very well be an elite defensive catcher, you should acknowledge that context and attempt to adjust for it in the way that seems best.
Likewise, WAR is an incomplete look at a pitcher’s overall value, as no one has figured out how to definitively credit the pitcher for his role in a team’s overall run prevention while he is on the mound. Run prevention is clearly some combination of performance from a pitcher and the defenders behind him, but splitting up precisely who is responsible for what degree of run prevention is something that is just beyond our abilities at the moment.
Because of the problems associated with measuring defense and assuming that those defenders offer identical support for each of the pitchers on the team (for more, see Why Our WAR Uses FIP and the follow-up post), we’ve chosen to a use a FIP-based pitcher WAR so that we’re simply measuring things that we know we can actually measure and not preventing an adjusted metric that relies upon assumptions that we know would have to reflect things that didn’t actually happen on the field.
FIP, of course, is based solely on walks, strikeouts, and home runs, and it does not attempt to deal with the question of how to separate out value between pitcher and defender. It tells an incomplete story, and in exceptional cases where a pitcher’s BABIP differs greatly from the norm, we need to be willing to accept that WAR is an imperfect measure (no matter which “version” of WAR you’re using), just as it is with elite defensive catchers. Accepting that it has limitations isn’t the same thing as rejecting its usefulness – we can believe in the value of WAR while also acknowledging that a FIP-based WAR is almost uncertainly undervaluing Justin Verlander in 2011.
How much is he being underrated? Well, it’s nearly impossible to believe that his .236 batting average on balls in play was entirely the result of great performance from defenders like Miguel Cabrera, Magglio Ordonez, or Delmon Young. The Tigers employed some pretty lousy defenders this year, and while Austin Jackson probably tracked down a couple of Verlander’s fly balls that other center fielders wouldn’t have gotten to, we can assume that Verlander’s BABIP reflects that he simply gave his defenders balls that were easier to field than an average pitcher last year. Whether he can keep doing that is another question, but even the most ardent DIPS adherent should be able to admit that Verlander’s low BABIP is at least partially a result of how he pitched last year.
So, let’s work out how many extra runs he’d get credit for preventing depending on how we distribute the credit for his low BABIP.
Using his 2.99 FIP, we’d have expected Verlander to surrender 83 earned runs over 250 innings pitched. In reality, he surrendered 67 earned runs, so his low BABIP (and the timing of when he gave up his hits) saved him an extra 17 runs, which would translate to an extra +1.8 WAR. If you give him credit for the entirety of his BABIP, his WAR jumps over that of Jose Bautista’s. If you believe that a pitcher should be given credit for all of the outcomes of batted balls, Verlander has just as strong a case for the award as anyone.
I’d say that it’s more likely that Verlander’s BABIP was at least partially influenced by his defenders, however. So, let’s assume that half of the hit prevention was Verlander, and half was the result of quality defense played behind him. Giving him credit for 50% of the 17 runs saved beyond what his FIP accounts for, Verlander would get 8.5 runs, or an additional +0.9 WAR. This wouldn’t be enough to catch Bautista, and he’d still be a bit of a ways behind Ellsbury, but it would put him close enough to those two that a vote for Verlander would still be easy enough to justify.
After all, even an RA-based WAR isn’t capturing every facet of the game. Verlander’s quantity of innings did help keep the Tigers bullpen rested, and that allowed Jim Leyland to use them more liberally on days the other four pitchers took the hill. This isn’t the easiest thing to quantify, but it’s another small point in Verlander’s favor, and when trying to split hairs between three guys who all had MVP-caliber seasons, small things are worth considering.
In reality, the votes for Verlander are probably going to come from voters who are influenced by his W-L record and a lack of classic position player candidates on playoff teams, but that doesn’t mean that a vote for Verlander can’t be defended. Ellsbury and Bautista are each great candidates in their own right, but Justin Verlander had an exceptional year as well. You can argue strongly for either of the other two, but a vote for Verlander is not an incorrect vote. He had an MVP-caliber year.
45
Is there a good article on here about why fWAR uses particular positional adjustment factors? Is that borrowed from somewhere else? I only ask both because I think it is relevant here and because I wanted to look into it more.
Nice article Dave, refreshing to see something other than the WAR is the end all be all I see all the time now. (not from writers on this site, mostly from readers and columnists who have done a bit of research, and I do mean a bit)
There was a 6 or 7 part article on this:
Here’s part 1
http://www.fangraphs.com/blogs/index.php/win-values-explained-part-one/
Thanks Filihok
Great article about the value of Verlander, but couldn’t the same thing be said about Sabathia? Sabathia’s own fWAR was higher than Verlander and his FIP was lower too. I know Sabathia had better defense behind him for sure, but I think its interesting that Sabathia isn’t getting much love for his season compared to Verlander. I think Verlander was better for sure, but the fWAR is what it is.
This season, FIP was just 29 % of a pitchers performance. Strike outs were 18.5% , Walks 8%, and HR’s 2. 5%. It’s a damn shame that pitchers at this wonderful website, Fangraphs, can’t get any WAR credit for the other 71% of their performances.
SIERA, tERA, ERA-…
The median FIP in 2011 was 3.97. The median ERA was 3.93. FIP does an excellent job in predicting a pitcher’s performance based on the outcomes he has the most control over. Position players are already given credit for defense in WAR. Why count it twice. The park-adjusted FIP is the best overall way to gauge a pitcher’s value.
That’s actually a good point, if we want to credit some pitchers for part of their good BABIP, we should likewise adjust the defensive WAR for position players based on their pitchers’ BABIP.
So if Verlander gets +0.9 wins for 50% of his BABIP, you have to reduce the value you gave to his defense by an equal amount.
Defense is a two way street. The better the pitcher the easier the plays are for the defense. The better the defense, the easier it will be for the pitcher.
My problem in using WAR, and especially for pitchers, as an MVP selection tool is that it is telling us what should have happened, while I think the MVP award should honor what did happen.
Is Verlander’s BABIP going to be this low next year? I’d bet my house that it isn’t, but that doesn’t change that fact that this year it was. I think its folly to suggest that Jacoby Ellsbury or Jose Bautista was any more/less lucky in their stellar performance.
Verlander was lucky, but anybody who delivers a career season was lucky and so was anybody who ever won an MVP award.
Well, except for all the steroid users.
That is a good point. With WAR, we use DIPS for pitchers, but not DIBS (defense-independent-batting) for batters. Batters frequently have down years or up years based entirely on BABIP.
Exactly. Quickie research:
Here is the last 5 non-A-Rod AL MVPs BABIP:
Hamilton .390
Mauer .373
Pedroia .331
Morneau .328
Vlad .324
Only Morneau’s is not a career best, and the 2009-2010 winners are ridiculous. No one suggested they were unduly lucky.
I suggested Mauer was lucky on both BABIP and HR/FB. He basically had career best in all of the key categories.
But, no one expected him to regress or be injured to the degree that he has.
“My problem in using WAR, and especially for pitchers, as an MVP selection tool is that it is telling us what should have happened”
This isn’t a correct characterization of FIP.
“Is Verlander’s BABIP going to be this low next year? I’d bet my house that it isn’t, but that doesn’t change that fact that this year it was.”
This isn’t a correct characterization of the DIPS issue. We know the amount of a pitcher’s responsibility for BABIP is non-zero, but we also know it’s far from 100%. It’s not just luck we’re trying to sort out, but defensive performance.
Giving him credit for things he didn’t do would be a much bigger mistake than only giving him credit for things we’re certain he did.
Exactly. Anytime FIP is mentioned on here everyone jumps on the ‘what actually happened’ vs. ‘what should have happened’ argument. FIP is what happened. There is nothing theoretical about strikeouts, walks, and homeruns. It just happens to be a better predictor of future ERA than ERA itself, which is what I think people get hung up on.
Thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you for this post.
I 100% agree with this point, but part of Jeff’s point is extremely valid, if perhaps not elaborated precisely in his initial post. It’s not entirely fair to make these claims about luck with regard to pitching and then completely ignore it in regard to hitting. wOBA (the primary basis for batter WAR) is obviously superior to batting average, but it IS NOT luck independent in any sense of the word. It shouldn’t be surprising that many MVPs (particularly ones with high batting averages) posted career high BABIP marks. It also wouldn’t be surprising if guys had outlier years in terms of power in years they won an MVP, that hit tracker would have credited many of their HR as “lucky” (see Joe Mauer in 2009, 5th in the AL in “JE” HRs). Just because hitters tend to regress towards a varied range of BABIP rates (as opposed to simply .300) doesn’t mean the stat isn’t fairly volatile for them as well–it is. Its just quite a bit more difficult to distill from actual performance except in the most extreme cases.
The one type of discussion where hitter luck seems appropriately attended to is fantasy baseball, but in these straight evaluations and single-season award discussions, it is largely glossed over.
,blockquote>This isn’t a correct characterization of the DIPS issue. We know the amount of a pitcher’s responsibility for BABIP is non-zero, but we also know it’s far from 100%. It’s not just luck we’re trying to sort out, but defensive performance.
Giving him credit for things he didn’t do would be a much bigger mistake than only giving him credit for things we’re certain he did.
So, give the pitcher some credit for BABIP. To not give any credit is to ignore 70% of the plate appearances that pitchers experience.
NO ONE is saying to give pitchers credit for things they DID NOT do. They are saying to give the pitcher some credit for things they did affect or do. Justin Verlander didn’t just get lucky on BIP over 900 PA and 250 IP. Nobody’s that damn lucky, I don’t care how many orphaned kittens they pulled out of a river. I don;t care if you regress it significantly toward their career BABIP, but to not give the pitcher some credit for BABIP is not right, IMO.
This can be done easily, without going to the extreme of giving the pitcher 100% credit for BABIP. Is anyone actually suggesting 100% credit for BABIP? No.
The same site, however, DOES give batters 100% credit for BABIP, with the stipulation that “those hits really happened, and WAR measures results not expectations”.
fWAR is not measuring expectations per se, it’s just ignoring 70% of the outcomes with the explanation that pitchers don’t affect BABIP to a significant degree that they deserve any credit for it.
I think we should really listen to the guys that developed the metrics. I strongly agree with Tango that we should average fWAR and brWAR until we have a better % of what credit to give pitchers for BABIP.
I also agree with MGL, that single season UZR should be regressed for WAR calculations.
Apologies to those that have read me saying the same thing many times.
Circle,
As always you come with the best points. Well done.
Verlander had 14 starts where he gave 4 hits or less.
That’s not luck. If that was luck, he just would of just had a few lucky games and bunch of mediocre games. Instead he 1 no-hitter, 1 one-hitter, 3 two-hitters, 5 four-hitters, 4 five-hitters, 4 six-hitters, 7 seven hitters, 4 eight-hitters, and 1 nine-hitter. He never gave up more than 9 hits and he never threw fewer than 6 innings.
He earned the Cy Young and MVP. He just didn’t fall ass-backwards lucky into greatness.
Pointing to those numbers does nothing to quantify them vs what Bautista and Ellsbury did for their teams this season, so how does this help?
In Verlander’s case, his 2011 BABIP performance seems more likely to be primarily related to luck/defence based than the result of superior pitching. His FIP was better in 2009-10 and he struck out more batters, but had a BABIP of about .300. His line-drive rate was down, but key factor was that he allowed a .061 BABIP on fly balls (!) and a .671 BABIP on line-drives (league averages- .137 and .720). It is likely that fewer loopers were hit than normal, and that the outfield defence helped him out quite a bit.
50-50 is kind to Verlander.
Verlander threw his fastball significantly (10%) less in 2011 than he did in 2009, relying more often on his changeup, as he continues to refine his approach to pitching…this off-speed approach may have helped to drop his BABIP by 30 points from 2009 to 2011 – hitters are off-balance when facing him, all the time. This further reduction in BABIP (another 50 points) may indicate that he’s just continuing to improve on his ability to keep hitters guessing wrong, as he further utilizes his full arsenal.
And re: the outfield defence helping him out quite a bit, I’d suggest that the range indicators for the corner outfielders would suggest otherwise. Jackson is tremendous going side-to-side and back on the ball, but Young, Ordonez & Boesch are sedentary.
Normally I don’t let post season results influence my decision on who I want to win the MVP award, but in this case I cannot help but want to take a sledge hammer to Ellsbury’s candidacy. His was a team that was destined for the World Series even before spring training, he easily had one of the most impressive and unexpected seasons this year, but it still wasn’t enough to vault his team to their preordained playoff spot.
The Red Sox didn’t play like champions this year and it showed in game 162 when the Rays won and the Sox lost. No one on the Red Sox played horribly below expectations, they all had solid individual efforts (minus John Lackey). Considering that Boston ranks in the top two (BA was topped by the Rangers) in the triple slash, lead wOBA, lead wRC+, third in the majors for FLD (second behind the Rays for the AL) and lead the league in position player WAR, not to mention 4th in pitching WAR, they should have had the playoffs easily.
Ellsbury’s contributions just were not that valuable.
Advanced stats aren’t perfect (and can be manipulated). WAR, FIP, etc. are nice, but each has weaknesses. Thank you Dave Cameron for pointing out systematic flaws and a large margin of error.
For example, fWAR and bWAR try to measure the same thing (value in Wins).
fWAR says Ellsbury was better than Verlander in 2011 (9.4 to 7.0). bWAR has Verlander over Ellsbury (8.5 to 7.2).
From part two of FIP based WAR:
FIP blames BABIP entirely on defense. That’s definitely wrong. Defense-adjusted RA [bWAR] assumes that each pitcher got the same support from their teammates. That is also definitely wrong.
So, we’re left with two imperfect options. Which should you prefer?
Defense-adjusted RA sounds much more reasonable to me. Definitely not perfect but better than ignoring all balls in play.
Verlander Won, WOW!! Broke my six year streak of picking the winners correctly, bummer.
Done deal. Well deserved.
Wow…did not expect that. Congrats Justin!
Okay, maybe not the exact right moment, but does WAR include PB and WP for pitchers and catchers?
How is Verlander’s WAR just +7.0 when his record was 22-5?
I would assume a replacement-level pitcher would have like a .333 Win%, or a 9-18 record in 27 starts. Even if Verlander was a bit lucky, and his record should’ve been 19-8 given an average bullpen, isn’t he still well above +7?
did you figure in a replacement level offense, defense and bullpen?
I don’t see why that should be factored in.
I would assume a replacement-level pitcher would have like a .333 Win%, or a 9-18 record in 27 starts.
How wouldn’t that record change with a stronger defense and more run support? Or more games with no decisions if the bullpen blows the lead a few times?
It is 38%. The relevant paragraph is, “If we assume that a team has a league average offense, a league average defense, and a league average bullpen, and that they are playing a league average opponent, we would expect them to win any single game started by a replacement level starter 38% of the time.”
http://www.fangraphs.com/blogs/index.php/pitcher-win-values-explained-part-three
And you are assuming that he played with a league average supporting cast vs league average opponents (that latter one is probably a good assumption, the former is not).
So can we stop pretending that the BBWAA is embracing advanced statistics now?
This should have happened last year when Felix won the Cy and David Price finished 2nd. (yet most confused an ERA vote with a “sabermetrics making progress vote”)
You were deluding yourself if you ever thought they were. However, JV getting elected MVP shouldn’t have been your first clue, since it was pretty well deserved if you look at more stats than just fWAR.
“we can assume that Verlander’s BABIP reflects that he simply gave his defenders balls that were easier to field than an average pitcher last year. ”
When you say he gave his defenders balls that were easier to field, you probably have in mind pop-ups, softly hit grounders, and can o’ corn fly balls. But part of his low BABIP- perhaps a large portion- was due to neither his skill in yielding easily fieldable balls nor the quality of his defenders. If we went back and watched every play, we’d probably see that Verlander had more of his hard hit balls happen to be hit right at or very near to his defenders than the average pitcher. We probably want to count as nearly pure luck the fact that a higher than average proportion of his liners and hard hit grounders were hit directly to defenders and not a couple feet more to the left or right.
One might also want to account for the fact that some portion of his low line drive rate (and thus more easily fielded balls) was itself luck.
I think it’s a lot more likely that his low LD rate was because of hitters just not being able to square up his pitches as well as, say, a league average pitcher like Rick Porcello.
As offense decreases for whatever reason, don’t we see the elite pitchers further separate from the pack. Such as, didn’t the elite pitchers of 60s have a larger difference in WAR from the average than elite pitchers in offensive eras?
Verlander’s control also improved.
I’m fine with splitting the difference between his 2011 BABIP and his career BABIP in terms of deciding how BABIP to give him credit for. Give him credit for a .260 BABIP.
Dave wins. Roberts and Klaassen can suck it!
Dave, were you forced to write this article as devil’s advocate because you lost a bet, or something? Because if not, I’m pretty upset (humorously, not actually) to read this on Fangraphs, by YOU of all people:
“I’d say that it’s more likely that Verlander’s BABIP was at least partially influenced by his defenders, however. So, let’s assume that half of the hit prevention was Verlander, and half was the result of quality defense played behind him.”
No hax please! So suddenly BABIP is now 100% (or close to it) attributable to pitcher skill and defense?
It’s possible that it is primarily skill, only that the range of skill at the MLB level is small.
Think about it, the MLB pitchers that “stick” in MLB are the top pitchers of the top league or top professionals.
How much range should we expect in a certain skill at that level?
We chalk up BABIP to luck because it varies and because there’s a small range between the best and the worst.
Rather than say it’s not a skill, we should perhaps view it as just a lesser skill, than say strikeouts.
The idea that if a BABIP isn’t .290 or .300 or whatever the exact number one wants to use is due to good/bad luck is just lazy IMO.
We’re basically saying Verlander got lucky over 250 IP on BIP, rather than saying he just displayed unusual stuff (command, control, etc) this year, and batters struggled to make consistent, good contact.
I still prefer the 50% credit thing until we have a better measure.
At what point does a study get done or some ACTUAL analysis? (this is not directed at you but if someone else threw 50% completely out of the air toward Dave to make an argument like this, Dave would likely ridicule him for doing that)
How about looking at his xBABIP?
How about looking at the UZR behind him based on batted profile? (not just a simple aggregate)
How about looking at a park effect based on his (and opponent hitter) handedness?
Do other pitchers on the staff show similar BABIP deviations? I recall looking at this earlier in Sept and only seeing Fister (who had a small sample size) as having a significantly below normal BABIP – this would suggest it’s less of a defense related thing and more luck and/or skill
How about looking at amount of inherited runners of his the bullpen allowed to score? (if looking at an RA based metric)
But but but then how will we ever be able to detract from the achievements of someone we feel is getting too much credit from the MSM if we can’t just look at BABIP???
It seems like the backtracking on what I’ll very generally refer to as DIPS theory started around the release of Moneyball. I know there has been a softening for a while to reflect a more appropriately nuanced stance.
But the reality is that just like Einstein was wrong about a lot of things, very significant components of DIPS has been proven wrong because we now have better ways to measure them. In the physical sciences they say you always make sure to state what you used to measure something because a value is only as precise as the instrument used to measure it.
Somehow DIPS became the saber equivalent of Relativity Theory. Really, it wasn’t. It was just what we had. And now we have better was of measuring balls in play and otherwise. Time to move on.
Thanks to Dave for repeated attempts to move the herd.
Saying Verlander got lucky on his BABIP isn’t saying he got lucky on 250IP of his BIP. What it’s saying is that on average, Verlander had more fortune then misfortune.
If a hitter has a career HR/FB of 3% and then suddenly shoots up to 30% wouldn’t this be chalked up to something?
It’s equally lazy to suggest that Verlander’s BABIP had nothing to do with luck unless there is something to point to a significant shift in his pitching. He dropped his BABIP by over 50 points of his previous career average and over 40 points in his previous career low. Luck played a role.
I’m trying to see how you justify giving it to Verlander over his teammate, Miggy Cabs.
That’s what I’m sayin. That guy has such a huge ripple effect on the rest of his lineup, which is why Leyland looked like a genius making the playoffs with idiots like Dirks, Maggs, and Guillen getting significant ABs. Miggy is such a difference-maker…I don’t care what the numbers say, he produced more wins for Detroit than Verlander did.
hey tigers might move miggy to third base. would be amazing for my fantasy team and just ghastly for the tigers. he would be a butcher.
The numbers say he produced more wins than Verlander, actually.
No really, I didn’t deserve this award.
[quote]Let’s start off with the elephant in the room: with how we calculate WAR here on FanGraphs, Verlander’s +7.0 season is pretty far behind both Jacoby Ellsbury (+9.4) and Jose Bautista (+8.3), so an argument for Verlander seems to require a rejection of FanGraphs WAR, right?[/quote]
brWAR
Jose Bautista = 8.5
Justin Verlander = 8.5
JV as MVP could actually be a very sound sabermetric decision.
I appreciate one site utilizing different methods than another and both striving to fine tune their metric for the highest degree of accuracy. But we also should not be limiting ourselves to a single metric when there’s so much grey area around some of the inputs.
I’m not so much referring to Dave, but the people that keep mentioning Sabathia’s slight lead over Verlander in fWAR.
If fWAR is all you ever have, then you’re simply the guy who has a hammer and views everything as a nail.
My preference is that pitchers not win the MVP, but if we use at least one (BR) WAR version, he deserves it.
In lighter news, David Robertson received a vote. http://bbwaa.com/
@keithlaw
Not as bad as Victor Martinez appearing on 4 different ballots.
Michael Young received at least 1 vote in each ballot spot. Last time that happened was 2003.
he is 28 times more valuable than sabathia of course he won
In regards to Verlander’s BABIP … both his walks and strikeouts were down per 9 rates).
Now this likely does not mean that JV had lesser stuff than previous seasons, but maybe batters took a different approach, such as hitting hard to square up pitches to avoid 2-strike counts like the plague.
I’ll check BR later to see how his 2-strike counts compared to previous seasons. If that be the case should Verlander be credited with batter contact on tough to hit pitches with high velocity/movement? Seems to me you credit pitchers for quality pitches with the understanding that over 900 BF that batters will eventually hit at normal levels unless there’s an increase in pitch quality.
With improved control there’s also the possibility of JV pitching in pitcher’s counts more often than previous years, which helps the pitcher and is something they could be credited for.
His walks and hits allowed were down despite more BIP. Either batters kept hitting the ball right at the D or the D played better when he was on the mound and/or JV was doing something to supress hits.
I’m always interested in things like batted ball velocity, % of flies that fall for hits and things of that nature and how they affect BABIP. If he were allowing, on average, lesser mph off the bat we should expect a lower BABIP. I think we should look at other things when trying to determine what % of BABIP is luck and what % is due to something the pitcher is doing.
Personally, I would like to do a study on BABIP for pitchers as it relates to count. I have a personal suspicion after watching JV all year that 1. he got full or behind less often than he has in the past and 2. hits resulted in more of those counts than would be expected (vs K’s).
Edit: Sub “balls in play” for “hits”
One thing that I haven’t seen anywhere is an argument based on the workload that Verlander takes on in relationship to the bullpen and the rest of the rotation. The fact that he went so far into games so often creates enormous value insofar as allowing Jim Leyland to rest his bullpen guys every fifth day. It seems like all of the stats that are parsed and debated take into account an aggregate of value based on games played, but may not account for value added by not taxing middle relief.
On a separate but related note, I think that the ‘everyday’ argument for not including pitchers in MVP consideration should be countered by the above point as well as the fact that if you count each batter faced by a workhorse SP, you often end up with a greater number of at bats than an everyday player who stays healthy for the entirety of the season.
This argument is not rare at all and is overdone…. he averaged ~7.4 innings per start, that’s ~1 inning more than the average starter – so certainly not insignificant but it’s not like the bullpen takes the day off like is often romanticized with some of the workhorse pitchers. Verlander completed 4 of his 34 starts.
Giving your HIGH leverage relief off is significant, not sure giving middle/low leverage relief time off is as significant… after all they are in middle relief for a reason. I’d think complete games would be more significant in terms of truly giving your bullpen the day off (and resting the high leverage guys) if you wanted to look at that impact.
Pretty sure in games where you are winning and by a low margin, any relief pitcher thereafter appearing is in a high leverage situation. This includes 7th and 8th inning guys.
It’s pretty obvious that Verlander accounted for more wins above replacement than any other pitcher last season. It’s also pretty obvious that Smith’s method does a better job of gauging player value. Your WAR calculations are broken. Either fix them or omit them. The present system benefits no one.
So, it’s “obvious” that fWAR is wrong b/c it didn’t come up with the conclusion that was so “obvious” to you? It’s broken b/c it didn’t fit your preconceived notion of what it should have said.
Or it’s “obvious” that it’s broken b/c Halladay finished more than a win better than Verlander did this year?
This is a very facile conclusion. “I don’t like the result, therefore, it’s wrong!”
I should qualify my statement. It’s obvious to people with eyes and working brain.
“It’s pretty obvious that Verlander accounted for more wins above replacement than any other pitcher last season.”
CC Sabathia had a higher WAR.
WOOOOSH!
That was the sound of the point whizzing over your head.
Isn’t there some way to normalize BABIP to see if it is a reapeatable skill? Maybe by comparing an individuals stats to his team mates? They pretty much play in front of the same defenders.
EG: if Verlander has a BABIP of .230 and Scherzer (or anybody else) is at .290 over something like 800 PA playing for the same team doesn’t that suggest that Verlander is doing something to create that difference and not just lucky? If so, he deserves credit for that talent.
It’s more like 600 PAs where there’s a ball put into play. A sample sufficiently small enough to suggest that luck is more than likely the primary driving force, and not some super-special skill that no one else possesses.
Slightly OT, but do Fangraphs readers tend to be for or against the BCS in college football?
What’s college football?
My problem with this line of reasoning is that it essentially gives Verlander the benefit of all doubt — he, alone, is the reason for his low BABIP — and he still just reaches Bautista and still falls short of Ellsbury in WAR. If we give Verlander half the credit he’s still short of Bautista.
Therefore, in order for Verlander to warrant the MVP, not only do we have to give him the benefit of all doubt but we also have to detract from BOTH Bautista and Ellsbury’s performances this year. In other words, we have to suggest that both of them weren’t really as good as their results indicate due to BABIPs that were too high (and thus, unduly influenced by luck) or their HR/FB indicated they were too lucky or their UZR was too high and that they really weren’t that good in the field as they seem.
So the argument for Verlander, aside from playoff performance and wins, is that all of the unmeasurables benefit him and hurt the case for both Bautista and Ellsbury. That’s a possible scenario but hardly the most likely, in my opinion. I’m inclined to acknowledge that he does probably deserve a boost for the BABIP but he’s 2.5 wins short of Ellsbury and a full win short of Bautista. Should we really believe that every bit of those 2.5 wins and 1 win is absolutely accounted for by the unmeasurables? I don’t think so.
“…we can assume that Verlander’s BABIP reflects that he simply gave his defenders balls that were easier to field than an average pitcher last year.”
So we can “assume” that his BABIP will normalize and sit under .240 for the remainder of his career?
BABIP is mostly a luck-derived statistic that fluctuates greatly from season to season. Just ask Verlander who has a 40 point difference in his previous career highs and lows
I’ll rarely, if ever, state that a pitcher get full credit for BABIP. But I have a science type of thinking in that I need to know why and conclusions like “luck” leave me wanting.
In my comments I stated some improvements by Verlander, namely better command that could affect BABIP by placing him in counts favorable to pitchers.
This year compared to previous seasons hits his BABIP and K/9 are down. So what I am asking (perhaps suggesting) is that batters may have been making weaker contact rather than striking out at times.
Certainly I don’t feel Verlander’s stuff is lesser than it was in his 10 K/9 seasons.
What I am doing is exploring all options before relying on luck as my answer in trying to figure out what % of BABIP (between 1 and 99) may be due to the pitcher.