Let’s Make One Thing Absolutely Clear About Aaron Judge
One more thing about this season’s unanimous American League Rookie of the Year, and possible American League Most Valuable Player. At some point, we’re going to fully move on beyond 2017, and all the numbers will just be a part of the record, only occasionally reflected upon. I want to dedicate one more article to what Aaron Judge just did. What he did, that is, and what it means.
We all probably have our favorite Aaron Judge highlight clips. “Favorite” might be strong, if you, say, root for the Red Sox, but we all remember certain moments. I imagine a lot of people out there most greatly appreciate Judge’s hardest-hit home runs. They’re spectacular, the lot of them. Yet, maybe it’s recency bias, but I haven’t been able to stop thinking about a Judge double from early last month. Judge, you’ll remember, was struggling in the ALDS against the Indians, and then, in Game 4, Trevor Bauer threw him something hard, high, and tight.
The chyron down there exaggerated things, but the ball approached the plate at 96 miles per hour, and it left at 98. By the standards that Aaron Judge set, a 98mph batted ball is barely enough to get your attention, but it’s not just about the batted ball itself; it’s just as much about the pitch that Judge turned on.
Everyone struggles against the pitch up and in. Everyone has a difficult time making hard contact against the pitch up and in. Judge didn’t by any means make perfect contact, but the contact was plenty good enough. It was a screaming line drive, after Judge shortened up.
Again, we all remember the balls that Judge hit around 120 miles per hour. He hit this pitch at 113.
He hit this pitch at 112.
Thanks to the wonders of Statcast, we know what Judge can do when he hits the ball just right. You can play around on Baseball Savant for ages. Judge is closely associated with the term “exit velocity,” because Judge has become its powerful king. A statement you’ve probably read dozens of times, or thereabouts: Aaron Judge hits the ball harder than anyone. We would’ve suspected that, without the tracking systems, but now we can know it for sure. Aaron Judge, hard hitter. All right.
What might not be fully appreciated is what that means. What it means is that Judge isn’t exactly playing by the usual rules. To show you what I’m getting at, I’ve prepared a plot. I looked at every individual batter-season from 2015 to 2017 with a minimum of 100 batted balls. For every player, I found the wOBA on contact, and I also used Baseball Savant to find the expected wOBA on contact. So this is wOBA without walks, hit-by-pitches, and strikeouts. Judge is highlighted in yellow. You’ll see him.
Judge just had the highest wOBA on contact in at least three years. He also had the highest expected wOBA on contact in at least three years, and he’s in the lead by 53 points. The name below him, unsurprisingly, is Giancarlo Stanton, but Judge’s data point is extraordinary. By wOBA on contact, Judge just finished a full four standard deviations higher than average. By expected wOBA on contact, meanwhile, he was 4.7 standard deviations higher than average. Stanton, in 2015, was at +3.7 standard deviations. Judge was one standard deviation higher than the next-closest point.
The explanation is simple, and it’s the one you already know: Judge hits the ball the hardest. But let me maybe help drive this home. Overall, last season, Judge posted a .430 wOBA. He did that while striking out almost 31% of the time. There’s concern that Judge might strike out more often. I calculated where Judge would’ve wound up, keeping everything the same but bumping his strikeout rate up to 40%. At 40% strikeouts, 2017 Aaron Judge would’ve still finished with a wOBA of .374, essentially matching the season put up by Edwin Encarnacion. Judge would’ve been great while striking out at a frequency that, for just about anyone else on the planet, would be catastrophic.
In terms of current or recent baseball, the closest thing to Judge is Stanton. That was the talk when Judge was a prospect, and now so much more has come true. Yet Judge might even have an extra gear. He just beat 2015 Stanton in wOBA and expected wOBA on contact, and he also beat 2015 Stanton in his rate of the best-hit batted balls, which Statcast calls “Barrels.” Stanton gives us our closest comp, but Judge is otherwise way the hell out there. He’s incredible by his peak contact, but he’s no less incredible by his below-peak contact. He can squeeze line drives and extra-base hits out of pitches other guys might pop up. I go back to the Bauer double, again. It’s just a sample of one, so, who knows, but that feels like something indicative of Judge’s atypical ability. Bauer even said later he thought he threw a good pitch. He threw a very good pitch. Judge somehow clobbered it.
The point you all knew: Judge hits the ball the hardest. The point that’s less obvious: That means Judge gets to play by his own rules. Strikeouts don’t mean for him what they would for someone else, because Judge doesn’t need a low strikeout rate to be good. He doesn’t need a low strikeout rate to be terrific. He doesn’t even need a low strikeout rate to be a deserving league MVP. Statcast has revealed the whole truth of Aaron Judge, and the truth is that, at least in the American League, he is one of a kind. That might make it harder to see the future, but then, maybe, it doesn’t make it harder at all.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
His zone rates are all pretty solid even for the amount he strikes out, so as long as he continues to develop as a hitter I would venture to guess that K-rate would go down, especially as umpires finally learn how to call balls and strikes properly against him.
That’s a scary idea for opposing pitchers.
It’s a scary idea, but the bottom of his zone is being called inordinately low. I don’t know how a player can be expected to “develop” around that. Do umps annually review player accuracy?
Well, I meant more that as he develops as a player, his pitch recognition should improve to the point where he lays off more and more of the outside corner junk balls.
As for the umpires, it’s extremely frustrating that he has to deal with that low zone – but eventually you would think that as Judge accrues service time that he would get a little bit more of the benefit of the doubt, and that the umpires would learn from their mistakes. The latter might be wishful thinking but I would hope that he wouldn’t have to deal with an unreasonably large strike zone his entire career.
I know that it affects him considerably less (bottom of the league in pitches taken/only 33% of pitches are high in the zone), but the umps still don’t seem like they’ve adjusted to Altuve’s top end. Looking at last year’s breakdown, they missed a lot of calls (to his detriment) last year, and he has significantly more service time.
That being said, the zone does seem to traditionally tighten or widen for veteran hitters or pitchers, respectively.
The beginning of the year he had better pitch selection but when the umpires continue to call low strikes on you you either just take it and argue or try to put a bat on the ball.
RE: K% affecting wOBA
What are the odds that the umps learn his strike zone? My friend is a former MLB ump, and he’s been very open about the struggles tall players pose. Have we ever seen umps self-correct on players, after significant exposure? Fangraph’s June analysis and subsequent October analysis actually showed that the umps became worse, as the season progressed.
It’s well-documented on Fangraphs that umps call low balls strikes on him ~100% more often than the league average, but I’ve never seen similarly thorough analyses on the likes of ESPN (which is what I expect umps to hear about). If those numbers don’t come more in line with the league average, Judge will likely swing at more balls and make worse contact. What does his wOBA look like, if those calls were league average instead?
Ref: https://www.fangraphs.com/blogs/umpires-are-having-some-trouble-with-aaron-judge/
https://www.fangraphs.com/blogs/the-peculiarity-of-pitching-to-jose-altuve/
You guys wouldn’t all be Yankee fans, would you? Or maybe holders of His Honour in a keeper league?
You seem to be harping on the low strike mistakes as a reason to project upside in Judge…all the while ignoring the even bigger mistakes on the high strike that are going against him. When Judge lays off a high pitch, it’s called a strike just 5% of the time. The difference against the league average is far greater on the calls going in his favor, which when netted out mean there’s actually downside regression.
Not saying anything about Judge – he seems like a good guy on top of a great talent.
Do you have a reference for the high strike data? And nobody’s projecting upside, it’s just a suggestion that as a hitter in Judge’s class develops (and Judge has shown the ability to develop & adjust all the way through the minor leagues and the majors as well), that his pitch recognition will improve allowing him to cut down on the K rate.
On top of that, if you look at Judge’s contact and swing numbers, they mostly indicate solid plate discipline even while he’s running a pretty bad K rate.
If you look at the links I used, you’ll come away with 5 key data points:
67% – MLB pitches thrown low in the zone in 2017
28% – low balls called strikes (league average)
52% – low balls called strikes (Judge)
34% – high balls called strikes (league average)
5% – high balls called strikes (Judge)
Essentially, high in the zone, he’s at +29%, and, low in the zone, at -26%. However, the ball is twice as likely to be thrown low in the zone, so he ultimately derives very little benefit from being tall. Ergo, it’s net negative.
Furthermore, pitches up in the zone tend to be straighter and more hittable. Hence, the saying goes, “if its high, let it fly; if it’s low, let it go.” In essence, breaking pitches that hitters are trained to lay off are falling out of the zone and being called strikes.
I think Jeff makes a good case that Judge’s performance this year was not a “fluke”, and that this is scary.
I also think that people who are looking for him to get better are likely to be disappointed. There is only so high most players can go.
“I think Jeff makes a good case that Judge’s performance this year was not a “fluke”, and that this is scary.
I also think that people who are looking for him to get better are likely to be disappointed. There is only so high most players can go.”
Analysis is supposed to cut through vague disbelief. We need reasons Judge can’t go higher than he is. Because the case Jeff has made can be used to support the notion that he can get better: his expected wOBA on contact was higher than his wOBA on contact and he can cut down on his strike outs. I’m not saying he will cut down on strike outs, but unless someone presents a good argument why he won’t, I am going to guess that there’s a distinct possibility that a 26 year old will cut down on his strike outs going forward after having only 800 plate appearances in the league.
Maybe Judge will never have another season as good as 2017. Maybe for the next few years 2017-like numbers will be what we can expect from Judge. Or maybe Judge will become the standout offensive player in the game over the next few years. To me, the data show that the potential for the last possibility is real.
Analysis shows that Judge will decline because of regression to the mean, because analysis shows that it is highly unlikely anyone is that good. Look at the projections if you want to verify.
The areas where someone might improve are speculative. Many players might improve in those areas, and they do not. It is much more likely that Judge is like them instead of like George Springer, who is kind of unusual.
On the other hand, regression from a season in which you club 50+ homers and run a .357 BABIP or whatever it is is something that tends to happen, with much more frequency and much more dramatically than the chances of improvement. Also, injuries.
I think that Jeff Sullivan has given us reasons to think that the projections might undershoot his value, since his batted ball data looks very good. I do not think it gives us hope that he will cut down on strikeouts.
“Analysis shows that Judge will decline because of regression to the mean, because analysis shows that it is highly unlikely anyone is that good. Look at the projections if you want to verify.”
Will analysis show that? Are you sure? I think you are painting with too broad a brush here. I’m sure the projections are too. Statistics are tools. They must be applied with judgment.
“The areas where someone might improve are speculative.”
The areas may be speculative but the idea that a rookie can be expected to improve is not. I agree that a rookie that starts hot can be expected to be more likely to perform worse the next year than a rookie that starts cold (if both are given an equal chance to do either). But Jeff’s observations throw up a red flag to consider before just plugging and chugging.
“I think that Jeff Sullivan has given us reasons to think that the projections might undershoot his value, since his batted ball data looks very good. I do not think it gives us hope that he will cut down on strikeouts.”
I agree. The hope that he cuts down on his strike outs comes from his AAA and MLB records of strike outs. I am also making an educated guess that Judge is actively trying to cut down on his strike outs, something I doubt that someone like Adam Dunn, Chris Davis, or Joey Gallo is really trying to do. Obviously Judge will run up against a wall eventually, but perhaps he has few enough at bats in the league and is young enough to suggest there is a good possibility he will be able to improve his strike out rate.
In the end, we need to be right, not to be safe. I’m not saying you are wrong. I just think you are being nonchalant. “There is only so high most players can go” is sort of a thought-terminating sentence. I don’t know what it means. There is only so high every player can go, in fact.
The projections often win. When they are wrong, our a priori understanding of why the projections are incorrect are almost always wrong. Regression to the mean is a thing. It is real.
While the regression probably will not be as much as the projections say, it’s pretty optimistic to say that not only is the projection wrong in point estimate, but that he’ll actually get better. There is just not much room for improvement except cutting strikeouts, which is notoriously difficult to do. He is not going to hit the ball harder. He is not going to get faster, or become a better fielder. His strike zone is not going to magically shrink (sorry to everyone who is counting on this).
So when I say “there is only so high most players can go”, what I’m saying is that while he has a shot to replicate what he did last year he’s probably reached his ceiling. Is that really so controversial that a guy who wasn’t a Top 25 prospect almost certainly hit his ceiling in a year where he put up over 8 fWAR? It shouldn’t be.
“Regression to the mean is a thing. It is real.”
I never said it wasn’t real. In fact I said it was: “I agree that a rookie that starts hot can be expected to be more likely to perform worse the next year than a rookie that starts cold (if both are given an equal chance to do either).”
“Is that really so controversial that a guy who wasn’t a Top 25 prospect almost certainly hit his ceiling in a year where he put up over 8 fWAR? It shouldn’t be.”
Yes, I think that’s controversial. I don’t understand how you can claim to know his ceiling is at 8fWAR. He just happened to hit his ceiling his rookie year? Why wouldn’t he perform significantly worse or significantly better next year? Jeff’s information provides exactly the right amount of positive insight to say that Judge neither over-performed nor has any room for improvement? That sounds like a bit of a knife’s edge.
As far as my opinion of what can be expected from Judge, there should be a distribution for his predicted output next year that, without any other considerations than his prospect status and his minor league wOBA, would have some results that exceed his 2017 but overall would be heavily skewed towards results that underperform his 2017. However, I think consideration of the facts that Jeff presented result in a more favorable distribution than that naive distribution.
I think statistically judge benefits more from not called high pitches than low bad calls since the difference from bottom to knee is lower than the difference from knee to letters.
however pitchers might throw low more often.
Well, I unfortunately read the comments and thought the same thing thinking maybe all just Yankee fans but I remember him getting some “low” strikes called on him. Also however, I remember him swinging and whiffing on what seemed like endless hittable belt high fastballs (or even breaking pitches). It didn’t make sense, and “slump” couldn’t explain it. The author mentioned RedSox fans, and for them all I could think about was “wait wouldn’t they just remember Judge bailing out a depleted Drew Pomeranz who couldn’t throw strikes and had lost velocity?”
Either way the swing and miss is one of the things I like about him. If he didn’t whiff as often, assuming everything else stayed in line, Yankee stadium would just start looking like a softball field because he’d be OPSing 1.200 based on the 800 slg% alone.
I think the Problem is that a pitch moves very fast. you don’t actually see the pitch crossing the knees but you basically have to visualize the pitch path and then judge it.
basically the ump developes an internal database what a ball and what a strike Looks like over time.
and since he does that mostly for Players from 5″10 to 6″3 they will struggle without out of the norm height hitters (altuve, judge).
umpiring 90+ is really hard just like hitting 90+ is.
ideally I would prefer robo ump.
regarding judge I think he just accepts the called strikes and whiffs and waits for a fat pitch regardless. that is really the best strategy for him albeit it might hurt him some against elite pitchers who can spot it.
Judge had an incredible season. I expect that the MVP results tomorrow night won’t reflect this, but he was the best player in baseball this year. Tops in nearly every offensive category (at least in the AL) while playing a very good RF. Very excited to see what he can do next year — he has one of the wider range of outcomes of any player in baseball.
In the playoffs, Judge saw breaking balls over 50% of the time. It will be interesting to see if pitchers go this extreme in the regular season against him.
They could try, but most pitching staffs don’t have pitchers with breaking balls as dominant as the Indians.
You should run the numbers to see how Judge hit against the best pitchers, say the top quartile. I suspect it will show, like A-Rod before him, that he did quite poorly. Judge was essentially a non-factor in most playoff games, like A-Rod.
So…you’re saying he was worse against top-of-the-rotation starters and bullpen aces than he was against back-of-the-rotation guys and minor league call ups covering injured players? That’s a bold theory.
Alex Rodriguez had a 116 wRC+ over 300 PA in the playoffs over his career, including a 224 wRC+ (!!!) over 68 PA in the 2009 season.
Aaron Judge had a 114 wRC+ over 57 PA this last postseason.
For some context, Francisco Lindor had a 118 wRC+ and Christian Yelich had a 115 wRC+ this past regular season, so they were about that good in the playoffs.
So that’s how we can expect Judge to hit against the top pitchers I suppose. I’m sure Cashman will take it.
Why don’t you run the numbers, Jim? It’s your claim. I watched Judge all season and when he was hitting well there were several games of the sort where the Yankees had 4 hits and Judge had two of them, a double and a home run, or the Yankees got on base 6 times and Judge had two walks. Maybe you went into the playoffs with a preconceived notion of Judge and how he fairs against top pitchers, and then watched the Indians series. If so, there are a few things that come to my mind about that. First is that it’s an incredibly small sample size. Second is that everyone is expected to perform worse against top pitchers. Thirdly, the strike zone in that series was awful; not only was Judge the victim of several key low calls, he was also victimized by inside and outside calls. If I recall there was one at bat where there was both a ball 2 inches inside (and borderline low) called a strike as well as a ball 4 inches outside called a strike. Fourthly, if Judge strikes out 30% against all pitchers, he might strike out 40% or more against the best pitchers. This is going to look ugly, and there will be variance such that he might have some stretch were he strikes out 8 times in 10 plate appearances. That should not in itself be taken to mean that he has a weakness against pitchers pitching well, however.
Anyway, to me, two things stand out that could actually improve for Judge next year. Firstly is that there is reason to believe Judge isn’t getting a fair strike zone (other people in these comments have posted numbers suggesting this). Umpires might adjust. Secondly, I think Judge could make an adjustment on low and away breaking balls in the strike zone. He has swung through them an awful lot. I don’t know whether he has thought he’s swinging at fastballs on those pitches or not, but I think if he can learn to make some good contact on those pitches it will make it much harder for pitchers to so exclusively play the game of trying to throw perfect breaking pitches low and away (and if it’s too low and still called a strike, even better). He seemed to be hurting pitchers more on their keep-him-honest fastballs last season. If he can tweak his approach a little so that he doesn’t end up trying to pull that low and away breaking ball just because it’s slower than the fastball, and instead hits it the other way, I think he can make such an exclusive low and away breaking ball strategy unfavorable for the pitchers. With his strength and swing path, many of those pitches could be hit for home runs, especially at Yankee Stadium.
Would he get better if he decided to make more contact and sacrifice some exit velo, ala Stanton this year?
I wonder how sustainable a high BABIP is for him. Hitting the ball hard probably helps with that and what helps too is that he has a pretty conservative batted ball Profile. he has neither a high GB rate nor a really high FB rate. his average LA was around 15 degrees which is good too and most importantly he has a healthy pull rate but isn’t extreme in that regard either (a tad above league average).
so he tends to hit hard high line drives which have so much power that they go out.
most other hitters Need to elevate and pull more to hit homers which tends to be bad for BABIP but he can just hit hard high liners up the middle.
I could see a guy like bellinger who pulls and elevates more be more succeptible to a lower babip.
however while his BABIP indeed might be real for the batted ball Quality the batted ball Quality itself might not be sustainable. the pitchers might find some spots were you hit a Little softer or Pop up a Little more for example.
Miguel sano was a Monster the first 6 week this season as shown in this good article https://www.fangraphs.com/community/miguel-sano-is-the-three-true-outcomes-hitter-weve-been-waiting-for/ where he hit a high LA, everything hard but without Pop ups but he obviously couldnt sustain that. now injuries were part of the issue but that combination of high LA, low Pop ups and super low soft contact is probably not sustainable.
we are always waiting for the TTO hitter that has cracked the code and can Combine high HR and high BABIP so that he can beat the System and hit 300 with a 30% K rate. but Long term that “never seen batted ball Quality” will often Regress and the BA will be more in the Adam dunn territiory for the 25+% K hitters Long term if they cant lower their Ks..
Yes, sure, his batted ball quality might not be sustainable. But it’s not the system that says Judge can’t hit .300 while striking out 30% of the time. The system says it can be done. The system says Judge can do it if he sustains his batted ball quality.
As to Judge’s prowess, Judge’s numbers thus far say that for him there is the possibility to sustain it. He has shown the ability to sustain that batted ball profile for more than 6 weeks. And he has shown the ability to hit the ball harder than anyone else. Of course, his track record is short, so there’s uncertainty.
no of course there is no rule it can’t be done but it really has not been done. the career K% leader in K percentage among the .300 career hitters is mike trout with a bit over 21%, there is no other guy with even 20%.
now times with Ks have changed and many old timers are on that leaderboard but I chose 2K PAs as a cut off so plenty of players in the last 15 years have reached that cutoff (basically 3.5 seasons).
maybe a guy could hit .300 in a single seasons but over mutliple seasons it is hard even with just 25% Ks. the majority of 300 hitters are around or under 20% Ks even in these days (now of course there aren’t many 300 hitters anymore…)
Fair play, what’s of the however long (9 weeks?) where he was unable to make contact with any pitch, seemingly at ant velocity, over the heart of the plate. I took as much seeing that, as I did from his incredible results before and after
“I took as much seeing that, as I did from his incredible results before and after”
What did you take from that? Pitchers figured out how to get him out and then forgot? There’s a lot of variance involved in his batting style? He was injured? He’s prone to long lapses in concentration?
i mean, sure he could make less contact, but he’s already at a 67% contact rate. pitchers already go heavily low and away, and below 40% in the zone with him.
but given his physical strength and fairly flat swing, it seems reasonable that his on-contact quality should remain. there is real concern if he suddenly pops up a lot more and changes his approach, but that isn’t the case at least for now.
“At 40% strikeouts, 2017 Aaron Judge would’ve still finished with a wOBA of .374, essentially matching the season put up by Edwin Encarnacion.”
Jeff you left us hanging… what would it have looked like if his K-rate had only been 25%?
“Well, let me just plug that in and…”
*computer explodes*
also interesting what his numbers would look like at the same K rate and a 310 babip
I see Judge turning into a .250 or less BA with 40+ HR power
I think another risk is that he has one skill that is so good and another that is replacement Level (contact). Yeah Overall he still is great but if his one great skill Drops a Little or his Ks get worse it might get ugly.
not in a year or two but I could see him dropping off in his late 20s to early 30s like dunn. now judge hits the ball harder than dunn and his plate discpline helps too (albeit dunn had even better plate discipline)
but I still prefer when a guy has grade 50 contact and 60 power vs 30 contact and 80″+” power (as 80 isn’t really doing judge justice).
but as of now he is great a and I expect him to be at least very good (5 win or so) next year.