How to Argue About Clutchness

This probably isn’t a problem for most people, but I’m plagued constantly by the memory of frustrating baseball arguments from days past. I probably get into these arguments more than most people, partially because of my (and I hope it doesn’t sound immodest to say this) vast knowledge of the sport, but mostly because I’ve lived most of my life in New Jersey, which his home to the most stubborn, tendentious people you’d ever have the misfortune of meeting.
One such argument took place probably close to 15 years ago, when I ruined what was supposed to be a relaxing Friday evening down the shore by getting into a shouting match over the issue of Alex Rodriguez vs. Derek Jeter. I preferred A-Rod, who would go on to finish his career with a slugging percentage more than 100 points higher than that of his Yankees teammate. I was arguing against someone whose case rested on Jeter being “more clutch.”
If you’re old enough to remember what “analytics bloggers” like me thought about that argument in the 2010s, you can understand my quickness to anger and probably imagine the colors my face turned. When the dust settled, Jeter — who, it turns out, was actually an exceptional hitter all along — did finish with a better career postseason wRC+ than A-Rod. But it was close: 121 to Jeter, 116 for Rodriguez.
But the foundation of my position has stayed with me: A better hitter who performs the same in pressure situations — or even a little worse — ought to be more productive in the clutch than an inferior player who raises his game. How does that line of reasoning hold up? We’re in a bit of a slow period between the trade deadline and the sharp end of the playoff race — plus Dan Szymborski just wrote about the White Sox and their world-historical screwing of the pooch, which is the only thing in baseball I actually care about right now — so now seems as good a time as any to find out. Maybe this article will provide a useful public service to the stubborn and argumentative among you at some point down the line.
So let’s pit the clutchest hitters in baseball this season against the best. Who would you rather have up with the game on the line? I’ll use wRC+ as a proxy for general hitter quality and, um, Clutch (which rates how much a hitter raises his game in higher-leverage situations) as a proxy for clutchness.
We almost got some great symmetry here. There are 15 qualified hitters this season with a Clutch rating of 1.00 or better, which, according to the FanGraphs glossary, is “great.” Sort the same leaderboard by overall wRC+, and you get 15 players with a wRC+ of 148 or better. Which, if you close your eyes and cross your fingers, is basically a wRC+ of 150.
Only one player appears in the top 15 in both metrics, which from where I’m sitting basically makes him the favored child of the gods of fortune. Maybe it’s better to be lucky than good, but it’s best to be both. Anyone want to guess who this golden god of a man is? You got a name in mind?
Jurickson Profar. What a world.
Profar notwithstanding, situational stats don’t bode well for the context-neutral mashers. Four of the remaining 14 top hitters in wRC+ — Aaron Judge, Bryce Harper, Shohei Ohtani, and Yordan Alvarez — are actually in the bottom 15 in Clutch. Steven Kwan, no. 11 in wRC+ this season, is also down near the bottom of the Clutch leaderboard. Taking Profar out of the equation for both sides, here are your 28 contestants.
| Name | Team | wRC+ | Clutch | Name | Team | wRC+ | Clutch |
|---|---|---|---|---|---|---|---|
| Aaron Judge | NYY | 216 | -1.97 | Daulton Varsho | TOR | 90 | 2.83 |
| Juan Soto | NYY | 190 | -0.31 | Brandon Nimmo | NYM | 122 | 1.97 |
| Shohei Ohtani | LAD | 182 | -0.99 | Corbin Carroll | ARI | 89 | 1.93 |
| Brent Rooker | OAK | 167 | -0.33 | Corey Seager | TEX | 133 | 1.82 |
| Gunnar Henderson | BAL | 165 | -0.32 | Jordan Westburg | BAL | 129 | 1.57 |
| Bobby Witt Jr. | KCR | 165 | -0.09 | Ian Happ | CHC | 118 | 1.40 |
| Rafael Devers | BOS | 162 | -0.92 | Nico Hoerner | CHC | 91 | 1.37 |
| Marcell Ozuna | ATL | 162 | 0.12 | Spencer Steer | CIN | 100 | 1.32 |
| Vladimir Guerrero Jr. | TOR | 160 | -0.63 | Julio Rodríguez | SEA | 100 | 1.29 |
| Steven Kwan | CLE | 152 | -0.89 | Jesse Winker | WAS/NYM | 123 | 1.23 |
| Ketel Marte | ARI | 152 | 0.65 | Alec Burleson | STL | 116 | 1.12 |
| Yordan Alvarez | HOU | 151 | -0.95 | Ryan McMahon | COL | 102 | 1.10 |
| Bryce Harper | PHI | 149 | -1.69 | Gavin Sheets | CHW | 82 | 1.08 |
| Freddie Freeman | LAD | 148 | -0.58 | Bryson Stott | PHI | 91 | 1.02 |
I don’t know about you, but if I had to choose a list of hitters to win me a game, or a World Series, I’d take the list with Judge, Soto, Ohtani, and Harper.
In high-leverage situations, defined as a plate appearance with a leverage index of 2.0 or higher, the Good list and the Clutch list are very close, but the guys on the Clutch list have been better this year on the aggregate.
| Total | PA | BB% | K% | AVG | OBP | SLG |
|---|---|---|---|---|---|---|
| Clutch 14 | 499 | 11.0% | 15.8% | .325 | .403 | .620 |
| Good 14 | 523 | 15.7% | 17.8% | .311 | .417 | .528 |
So the better hitters walk more and have a higher OBP, but the clutch ones hit for more power. And if you drill down on individual performance, clutch hitters dominate the top of the list as well. I’ll warn you, this is where things get a little weird.
| Top Six | High-Leverage wRC+ | Good or Clutch |
|---|---|---|
| Daulton Varsho | 298 | Clutch |
| Juan Soto | 263 | Good |
| Jesse Winker | 239 | Clutch |
| Ian Happ | 215 | Clutch |
| Alec Burleson | 208 | Clutch |
| Corey Seager | 207 | Clutch |
| Bottom Six | High-Leverage wRC+ | Good or Clutch |
| Yordan Alvarez | 125 | Good |
| Rafael Devers | 123 | Good |
| Spencer Steer | 110 | Clutch |
| Corbin Carroll | 106 | Clutch |
| Bryson Stott | 97 | Clutch |
| Bryce Harper | 4 | Good |
Because holy crap, what happened to Harper?
Now, a lot of those tendentious New Jerseyans I talked about earlier are Phillies fans, and I still hear their baleful cries. The Phillies, who looked like the 2001 Mariners until about a week before the All-Star break, are currently in the midst of a slide that’s a couple bounces away from rivaling the one the Dodgers went on in 2017. The public, suffice it to say, is not taking it well.
Some of that reaction stems from the fact that a lot of the people who are acting like the Phillies will never win again are still twitchy from watching the Eagles start 10-1 and then look like they were never going to win again. That kind of — and this is going to sound mean and I don’t care — football mentality is ill-suited to baseball, which is a sport where you can’t call different plays or blitz more, and where just trying harder is often counterproductive.
But there’s been criticism of Harper coming up small in big moments and, yeah, turns out that’s right on the money.
In 43 high-leverage plate appearances this year, Harper is hitting .152/.326/.182. The five hits he’s managed include four singles and a double. It boggles the mind for three reasons: First, it’s hard to imagine anyone being that bad in any sample. Second, Harper’s been awesome in general this year; his struggles in high-leverage situations only entered into this article because he’s been one of the 15 best hitters in the majors this season. And third, we all remember this, right?
OMG BRYCE HARPER!!! WOW!!! THE SWING OF HIS LIFE!!! pic.twitter.com/9UWmROAuNV
— FOX Sports: MLB (@MLBONFOX) October 23, 2022
I was in the building for that home run. It’s one of the clutchest things I’ve ever seen in person, and not unrepresentative of Harper’s playoff career. In 49 career postseason games, he’s hitting .276/.383/.613. Since joining the Phillies, he’s .286/.455/.643 in the postseason and .278/.417/.493 in 295 high-leverage regular-season plate appearances. Before this annus horribilis of the clutch, Harper was a .299/.433/.546 hitter as a Phillie in high-leverage regular-season situations.
I’m not unreceptive to the possibility that Harper is pressing in big situations this season. He’s more conspicuously aware of his context and leverage situation — both in-game and relative to the pursuit of a championship — than any hitter I’ve ever seen. And the line between rising to the occasion and trying to do too much is impossibly thin.
But if I had the game on the line and the choice between having Harper or Varsho at the plate, I’d still pick Harper. Clutchness is not intangible; in baseball, it’s actually highly quantifiable. But only in retrospect.
You want to know how I know? Because Harper led the league — literally led the league — in Clutch last year. In fact, let’s do the same experiment using last year’s stats: Take the top 15 in the league in wRC+ and Clutch in 2023, and compare their high-leverage performance in 2024.
This time, there are three crossovers: Harper, Carroll, and Ronald Acuña Jr. So by taking them out, we’re reducing the sample to 12 good hitters and 12 clutch hitters. Five hitters — Freeman, Soto, Ozuna, Harper, and Ohtani — appeared on the Good list in both 2023 and 2024. Only two — Stott and Carroll — appeared on both Clutch lists.
Anyway, taking that longer lead time into account, here’s how the best and most clutch hitters of 2023, respectively, are doing in high-leverage situations in 2024.
| Type | PA | BB% | K% | AVG | OBP | SLG |
|---|---|---|---|---|---|---|
| Clutch | 410 | 8.8% | 21.5% | .230 | .302 | .368 |
| Good | 463 | 15.1% | 14.7% | .298 | .402 | .490 |
This time there’s no comparison. The good hitters from 2023 are miles ahead of the clutch hitters.
I could’ve saved myself a lot of trouble by doing this math all those years ago. The reason I lost that argument is that it was mostly backward-looking. You can tell who has been a better hitter in clutch situations, but past returns are not a guarantee of future performance. Harping on the past can be cathartic for purposes of venting on Twitter or talk radio, but it’s not a constructive predictor of the future. If you want to find the most clutch hitters of tomorrow, you’re better off looking at the best overall hitters of today.
Michael is a writer at FanGraphs. Previously, he was a staff writer at The Ringer and D1Baseball, and his work has appeared at Grantland, Baseball Prospectus, The Atlantic, ESPN.com, and various ill-remembered Phillies blogs. Follow him on Twitter, if you must, @MichaelBaumann.
Great stuff! Would be interesting to see a more formal review of how Clutch or WPA/LI changes for players year to year. Does past clutch lead to more clutch.
This is math that has all been done before. With the possible exception of Jim Leyritz, it does not.
Select list of players in the top/bottom 30 Clutch all time with WPA in parentheses.
Top
1. Tony Gwynn (57.9)
2. Pete Rose (24.1)
5. Ichiro Suzuki (14.3)
7. Dave Parker (31.2)
8. Omar Vizquel (-18.6)
10. Yadier Molina (4.1)
13. Mark Grace (38.9)
20. Ryan Howard (29.8)
21. Tim Raines (50.4)
24. Eric Hosmer (14.0)
25. Adrian Gonzalez (36.0)
26. Eddie Murray (54.7)
27. Michael Brantley (17.4)
28. Juan Pierre (-2.4)
30. Johnny Damon (18.8)
Bottom
1. Sammy Sosa (24.7)
2. Mike Schmidt (53.7)
4. Jim Thome (50.2)
6. Marcus Semien (-3.4)
7. Giancarlo Stanton (26.9)
9. Alex Rodriguez (58.5)
10. Andre Dawson (31.6)
11. Gary Carter (16.1)
12. Barry Bonds (126.8)
15. Adrian Beltre (19.6)
16. Mike Trout (52.1)
18. Andruw Jones (14.4)
22. Frank Thomas (60.3)
24. Mike Piazza (42.7)
25. Jeff Kent (23.7)
27. Ivan Rodriguez (-7.2)
29. Aaron Judge (27.8)
30. Bobby Bonilla (22.4)
Overall, there does seem to be a trend where the clutchest hitters tend to be contact-oriented low-walk low-strikeout hitters, whereas the un-clutchest hitters are often power-oriented high-walk high-strikeout hitters.
I believe that for majority of the players, their true talent Clutch is not meaningfully different from league average, and this makes leaguewide search for ‘clutchness as a repeatable skill’ a hopeless endeavor. But that does not necessary imply the absence of players with true positive/negative clutch skill.
I wonder how much of that is that clutch situations are usually with runners on, and batting average goes up with runners on.
It seems likely to me that much of the difference in BA with runners on is because the changed infield alignment makes groundballs slightly more likely to be a hit with runners on than with empty bases.
Are groundball guys systematically more “clutch” on average?.
WPA/LI has a bias in favor of home runs (relative to WPA and Linear Weights) which is why the “unclutch” leaderboards tend to be dominated by sluggers. This blog post goes deep on the topic. The tldr is that Home Runs occur slightly more often in low leverage situations and that they are more valuable (relative to other events) in those low leverage spots. Together, these factors causes the HR to be worth ~0.016 wins / ~0.16 runs more by WPA/LI than it is in WPA.
IIRC the Book indicates that clutch skill does exist, that there is a statistically significant corallation between clutch over years. But that it’s such a small effect in such a noisy data set that there’s no real point in worrying about it in lineup construction and any given batter who was clutch one year (or even over a reasonable number of multiple years) is more likely to have been lucky than to have actual “clutch skill”.
It absolutely exists. But we are miles from quantifying it.
I got into it with Dave Cameron in the comments during a playoff series where Papi was going insane and DC said that there was no such thing as clutch.
About 2-years later, he came around and acknowledged that clutch should be recognized. We just… Don’t know how to verify it, the sample size is always going to be such that there is a ton of noise.
But I don’t think that makes it completely meaningless to the point you don’t need to worry about batting order. You still want your best hitters getting the most PAs and opportunities.
Why would he recognize it but also not know how to verify it? Interested to see this post you referred to here.
Surprised Baumann didn’t mention that in Harper’s 2015 MVP season he similarly had the second-worst clutch in the league.
People who believe that “clutch” is a repeatable skill do not understand statistics or math, which makes them representative of most people.
That’s not true. Use math to prove to me that clutch isn’t repeatable.
People that make a statement shooting down the idea of clutch without actually having the means or the evidence to prove it’s nonexistence, do not understand sports, neuroscience, anxiety, or physical activities in general. But the very fact that people have been proven to have the opposite of clutch, the yips, unable to perform under pressure, etc., suggests that there is indeed some underlying wiring in our brain that can lead to some people performing better in pressure situations than others, given comparable training and preparation.
But all that aside, I don’t want any opinions, I want you to use math to prove, unequivocally, that clutch isn’t in any way a repeatable skill. You can’t, but if you can, then you can go on being a judgy jerk.
Russell’s teapot, but for baseball
You are proving my point about people not undertsanfing how statistics work. You can not prove something doesn’t exist. You can only fail to prove that it does exist over and over again. That has been done on this very site several times. Craig Calcetalerra had an article a decade or more ago.
People have been looking for clutch for a long time and nobody has been able to find it is a repeatable skill, which means it almost certainly is not.
See:
https://www.tangotiger.net/clutch.html
It exists. It has been known to exist for over 20 years.
It’s also very, very small.
If it’s very, very small, for practical purposes, it doesn’t exist. i.e. no one should care about it.
If I said, batters born in May have an advantage in MLB, and to prove it showed with 100% certainty that those batters have a 0.001 higher BA than other batters, you’d say the advantage doesn’t exist. If no one would ever rationally make a decision based on a certain skill or advantage, it doesn’t exist in the real world.
https://tht.fangraphs.com/the-color-of-clutch/
According to Tango, r=.15. so whatever your observed clutch is, divide it by seven? Also, clutch advantage is somewhat less than platoon advantage.
I mean this is the sort of thing you can’t disprove, but it’s a little absurd to cite lack of understanding of math and claim you can’t prove something doesn’t exist. We do it all the time in math! Just look at Euclid’s theorem!
Lol ok buddy. Sure, I don’t understand how statistics work or basic math. Great argument. You win. You’ve never been wrong about anything in your life.
Except this, but hey, who am I to burst your bubble, I don’t even math, right?
Jackass.
Yeah, the easiest argument to make disproving clutch is that it’s not predictive. A dude can be great one month and then terrible the next, or the next year, or whenever, or vice versa. A non-predictive skill is really just throwing it to the bell curve.
But proof doesn’t matter. Believing in clutch isn’t about numbers, it’s about emotion. It takes the conversation away from it all being a question of skill to being one of intestinal fortitude. Which resonates with people, particularly (I would say) Americans. It isn’t that I’m just better, it’s that I willed it to happen. I created my own fate.
Taking out the samples and variance etc – I’ve often wondered how often in those high leverage situations a Harper type gets a Loogy and a Varsho type gets to face the tired SP w/ the manager hoping he’ll get out of it against an inferior hitter to “save the bullpen”.
Regardless as a Braves fan, I’m V glad Harper is struggling in the clutch as the karma gods punish him for his very Jersey / Philly esque stare down of Arcia in the playoffs. Homeboy could use some clutch humbling.
a humble Harper is a lot less fun to watch.
I don’t think the Jeter – Rodriguez thing was as much about actual performance than vibes. Whatever postseason good things were remembered by those who got positive pheromones from him and whatever bad postseason things Rodríguez did were remembered by those who got yucky pheromones from him. It’s not factual. It’s visceral.
The yapperati also decided that Rodriguez didn’t know how to win because he signed the biggest contract ever with the Rangers, and they proceeded to lose 90 games a year for three years until the owner got tired of paying him and shipped him off to the Yankees.
It wasn’t remotely Rodriguez’s fault, since he put up more WAR in those three years than the entire Rangers’ pitching staff combined (27.0 to 22.3), but some people love the narrative of a player getting a big contract and then losing.
His career choices weren’t designed to maximize popularity. I’m not sure what he was thinking. I guess he maximized money in the second Yankee contract.
I don’t know if he was thinking. I think that Scott Boras was (quite selfishly) thinking for him.
Until ofc he had to can Boras and go begging for that second Yankees contract to save face.
Whoever is just downvoting me for the sake of downvoting me on every innocuous comment, please keep at it you lonely idiot. Downvotes mean nothing lol. Except that you’re a loser and sad.
Re: Jeter, it also became impossible to convince lower-information baseball fans that he was in fact a poor defender after the flip play.
I am curious what the distribution is of Clutch vs. wRC+. If one is more skewed than the other, the effect of what you are measuring here may not be as much that Clutch is a poor metric, but that ordinal ranking of Clutch is a poor metric. It would be interesting to consider say, players who are 25% above league average for each instead of “top 15”. That would get more to the question of how much better/worse is being 1% better as a clutch player compared to 1% better overall.
this is one of those things that you just can’t argue with someone about. A person who is arguing that they’d rather have Mediocre Player X instead of Great Player Y because he’s more clutch is predispositioned to ignore any evidence to the contrary
Look at Pete Alonso for years he hit real well wrisp and this year he’s hitting .198 with an ops in the 600’s in those situations and he went from “super clutch” to a “choker” per social media. The whole thing can flip on a dime even within a season as the Mets had good numbers with runners on base most of the first half and in the last month they’re hitting 180something with runners in scoring position.
That doesn’t mean it’s random.
It could be Alonso with conditions A, B, and C performs at _____ level, but Alonso with conditions X, Y, and Z performs at _____.
Maybe it’s not that either. I don’t know. I’m just saying that a lot of really smart people are acting like this has been proven, when the truth is it’s anything but. And we simply don’t have the means to. Yet. But wearable neurotech is big, it’s going to be a $30 billion industry next year. Workplaces are using them to monitor employee stress, gaps in attention, etc., and eventually, they will contribute to sports.
And that eventuality isn’t far off.
I don’t remember who it was, but a writer on here once posited that a major reason it’s so hard to identify players as clutch is that being largely immune to pressure is a prerequisite for having enough of a MLB career to even be in the sample.
Yes. People forget this all the time, and not just in the endless arguments trying to pin down the phantom of clutchness but in a lot of other contexts too — “the world’s best athletes” is an incredibly non-random sample of humanity in about a thousand different ways.
This was Dave Cameron’s argument for clutch and for other psychologic factors. It was overly simplified, but still a good point.
Well there are great hitters who didn’t hit well in postseason. Barry Bonds with Pittsburgh, Mike Schmidt, Ty Cobb. SSS is probably true, but it isn’t emotionally satisfying.
Did they not hit well, or did they face an unusually dominant and productive slate of pitchers?
I don’t think we can look at the stats in a vacuum like that. There should probably still be some index to compare their performance relative to the performance of everyone else that had to face those pitchers.
Maybe it comes out similarly, but it’s still an exercise that should be run before we do anything other than hypothesize.
Oh sure it basically means nothing, but it creates a narrative.
The best argument that I heard on this topic came from Kevin Goldstein formerly of this site. Being clutch isn’t about raising your level of performance in a pressure situation it is about not falling apart. There are certain people that can perform in any situation and certain people that don’t.
Think of it from the standpoint of a 3 foot putt in golf. A 3 footer with a match on the line or for a personal best score is a lot different than a 3 footer for bogey.
Certain people are better at remaining calm in pressure situations better than others. This is true in both sports and real life situations.
Except that is not what the numbers show at all. There is huge variation from year to year with the same person (Harper for example).
A person can change a lot year to year, or even day to day, too. I wonder if that’s the next frontier after teams figure something out about biomechanics and how to not break pitchers: measuring stress levels and brain activity to produce a measure of how far the player is from their baseline state, so you know if they’re locked in and ready to clutch it up
To use your putter analogy, isn’t it possible that you miss the 3 footer for bogey just as often, but you don’t notice it because it’s not as big of a deal at the time?
First, I never miss 3 footers. 🙂
Rory Mcilroy in the U.S. Open missed 2 short putts in the last three holes with the Championship on the line. His final putting stats for the year will reflect that he make 99.5% (estimated) of those putts throughout the year. When he really needed it though with a major tournament on the line, he missed.
Can you measure quality of pitcher in these high-leverage situations? I would imagine many of these good hitters are facing elite relievers, and I wonder if it’s at a different rate than the clutch hitters (who are in their own right, still very good).
Now if Judge can win “Feared” he might pick up serious HOF votes.
The basic assumption here is that situational play is situationally blind. If it is a crucial situation in a game, you are assuming that the pitcher ignores this fact. As Maddux put it of Barry Bonds, if it matters just walk him. One of the reasons that many big names disappear in the playoffs is that pitchers refuse to pitch to them, at least with the game on the line. I have no idea what this means to the argument.