Baseball’s Most and Least Homerable Pitches

By now you might’ve read that, Sunday afternoon, Clayton Kershaw gave up a home run on his curveball. Clayton Kershaw does not often give up a home run on his curveball. According to some sources, it was the first time Kershaw’s curve had been hit out in the regular season, ever. This is a disputed instance, from April 2011. We do know, for certain, Kershaw gave up a playoff dinger on his curve in 2009. Matt Holliday hit it, but unfortunately for him, the same game for Holliday became differently eventful in the later innings. Also, 2009 was before Clayton Kershaw became Clayton Kershaw. But anyway! The point is that Kershaw’s curve doesn’t get taken yard. It got taken yard, so that’s interesting.

Lots of people have come at this from the Kershaw side. Not a lot of attention has been paid to the Brandon Hicks side. Hicks is 28 and a former prospect, and a big reason why he’s never gotten regular big-league playing time is because he hasn’t been able to consistently hit non-fastballs. Since 2002, 1,308 position players have batted in the majors at least 100 times. Hicks has posted the third-lowest contact rate, at 59%. According to PITCHf/x, Hicks has swung at 172 non-fastballs and whiffed at 60% of them. Hicks is an all-or-nothing sort, and maybe that’s precisely the sort that was going to take Kershaw’s curve out. But prior to the homer, it’s not like Hicks looked comfortable against Kershaw breaking balls.

clip1321.gif.opt

clip1320.gif.opt

clip1322.gif.opt

clip1323.gif.opt

Ultimately, the game got to the seventh, and after striking out the first two times up, Hicks quickly fell behind Kershaw 0-and-2. The first time, at 1-and-2, Kershaw came with a slider. The second time, at 0-and-2, Kershaw came with a curve. This time, Kershaw came with a curve again, but it was a curve in a spot some might describe as “a pretty bad spot”. There are even bad spots for a curve when pitching to Brandon Hicks, and while Kershaw tried to bury the pitch down and a little in, he only got the “in” part right, and even there only barely.

kershawhicks1

kershawhicks2

It was a meatball. Let’s just say it.

dingdonger.gif.opt

The first incredible thing you realize is that Brandon Hicks homered off Clayton Kershaw. The second incredible thing you realize is how very notable that is. It’s not like this is the first time Kershaw has missed his spot with his curveball. It didn’t get hit out when it was perfect, and it didn’t get hit out when it was imperfect. Hicks jumped on a mistake, but there have been previous mistakes, and Kershaw has survived them anyway because his curve is so naturally impossibly good. As obvious as this particular mistake was, hitters in the past have failed to drive similar mistakes, and that tells you something.

Thinking about Hicks and Kershaw and the curveball got me thinking about something bigger. So we know that Kershaw’s curveball doesn’t get hit out of the yard very often at all. We have a suspicion that it’s one of the best pitches in baseball, in that regard. But what do the leaderboards actually look like? Which have been the most and least homerable pitches throughout baseball during the PITCHf/x era? Just where does Kershaw’s curve rank?

The research was simple, using the PITCHf/x leaderboards available at Baseball Prospectus. I examined the whole era, 2008-2014, and I kept starters and relievers separate, and I set a minimum of 1,000 pitches of each type thrown. This counts only the regular season, so we’re missing that Holliday homer off Kershaw’s curveball, but it’s not like that should change things very much. Of course, different pitches are used in different situations and put in different places, so this isn’t a true measure of a pitch’s homerability, but the information ought still be useful.

Let’s begin with the least-homerable pitches thrown by starting pitchers.

Player Pitch HR%
Clayton Kershaw Curve 0.05%
Chris Carpenter Curve 0.08%
Wade Davis Sinker 0.09%
Stephen Strasburg Curve 0.13%
Clay Buchholz Cutter 0.15%
C.J. Wilson Curve 0.17%
Brett Myers Curve 0.17%
Jose Fernandez Curve 0.17%
Chris Tillman Curve 0.18%
Joe Kelly Sinker 0.18%

Well would you look at that? Sure enough, we’ve got Kershaw’s curveball atop the list, or on the bottom of the list, depending on how you like to sort. It’s a list with seven curveballs on it, which isn’t a total shock since curves get a lot of non-swings and also a lot of groundballs. Kershaw shows up having allowed one curveball dinger and more than 2,000 curveball non-dingers. Carpenter’s curve was also magical, and it’s somewhat surprising to see Wade Davis in there, although I suspect there are some difficulties separating sinkers from other fastball types. In any case: that’s the list. About Brett Myers, here are the most-homerable pitches thrown by starting pitchers.

Player Pitch HR%
Brett Myers Fastball 1.64%
Derek Holland Curve 1.62%
Tommy Hunter Sinker 1.60%
Derek Holland Changeup 1.57%
Sean O’Sullivan Fastball 1.56%
Pedro Martinez Fastball 1.54%
Scott Olsen Fastball 1.49%
Hector Noesi Fastball 1.48%
Collin Balester Fastball 1.48%
Chris Capuano Slider 1.44%

Myers was a man of several extremes. Holland’s breaking ball has proven occasionally problematic in the past, and you can see here one of the reasons why Hunter has taken to the bullpen after not working so great as a starter. When you’re a starter, you don’t want even your sinker to be homer-prone. I also just noticed a second Derek Holland pitch in the top four. That’s two Derek Holland pitches that’ve been more homerable than Hector Noesi’s fastball, which is about the least-flattering way to put anything.

Turning now to the least-homerable pitches thrown by relievers.

Player Pitch HR%
Craig Kimbrel Curve 0.00%
Matt Lindstrom Slider 0.07%
Javy Guerra Fastball 0.09%
Al Alburquerque Slider 0.09%
Jake Diekman Sinker 0.10%
Jamey Wright Curve 0.12%
Brandon Lyon Cutter 0.13%
Javier Lopez Sinker 0.13%
Fernando Rodney Sinker 0.14%
Mark Melancon Cutter 0.14%

Craig Kimbrel. He’s thrown 65% of his breaking balls for strikes. Batters have swung at 41% of them overall. More than half of the swings have missed. Three of every four curves hit in play have been on the ground. Not even 30 Kimbrel curves have been hit in the air, ever, and that makes it tough to hit the pitch out. It’s almost impossible to make contact with the pitch. Then there’s the extra impossible step of hitting it fair. Then there’s the extra impossible step of hitting it fair beyond the infield. It’s just — yeah, so, 0.00%. He’s good. The other pitches have been good, too. They haven’t been Kimbrel-good.

And, the most-homerable pitches thrown by relievers.

Player Pitch HR%
Chris Jakubauskas Fastball 1.33%
Louis Coleman Fastball 1.29%
Carlos Villanueva Changeup 1.25%
Matt Reynolds Fastball 1.25%
Edward Mujica Fastball 1.21%
Joel Peralta Fastball 1.20%
Tyler Walker Fastball 1.19%
Dan Wheeler Fastball 1.19%
John Grabow Fastball 1.17%
Bob Howry Fastball 1.14%

There’s not a lot that’s interesting about this table, but you can definitely see why, when the Cardinals picked up Mujica, Yadier Molina was like, yeah, let’s only throw the splitter from now on. You also now remember who Chris Jakubauskas is, or, alternatively, you’ve now learned about Chris Jakubauskas. You wouldn’t expect to see a side-armer like Coleman on here, but Coleman has been an extremely extreme fly-ball pitcher in his time, and his slider is his money pitch. His slider isn’t the pitch on this list.

To wrap it all together: on Sunday, Brandon Hicks became the first player, maybe, to hit a home run off Clayton Kershaw’s curveball in the regular season. Hicks, until then, had been almost entirely hopeless against quality non-fastballs. This Monday through Wednesday, the Giants will coincidentally play host to the visiting Braves. Will Hicks now become the first player ever to hit a home run off Craig Kimbrel’s curveball too? No, almost certainly not.




Print This Post



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.


37 Responses to “Baseball’s Most and Least Homerable Pitches”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Abe says:

    Honestly, I was expecting the least homerable pitches to be a segue into the week’s most wild pitches.

    +6 Vote -1 Vote +1

  2. Brian says:

    Re: Kimbrel and his knuckle curve…

    Per baseballsavant.com, he’s thrown 1,183 sliders/knuckle curves in his career and surrendered just ONE extra-base hit – a double to right-center!

    Breakdown:

    Sliders: 778, 1 double surrendered
    Knuckle Curves: 405, zero XBH.

    This is insane to me. Right?

    +21 Vote -1 Vote +1

  3. Josh says:

    Shoutout to you in Jeff Passan article! It warmed my heart.
    “the great Jeff Sullivan”
    http://sports.yahoo.com/news/10-degrees–which-mlb-2014-breakout-star-is-off-to-the-most-dubious-start-012416082.html

    Vote -1 Vote +1

  4. Matt says:

    Before I read the article, I thought “0-2 Hector Noesi fastball’s gotta be one of the most.” Win or lose the rest of the year, this has been a successful year for the Mariners for simply DFA’ing Noesi.

    Vote -1 Vote +1

  5. CKrome says:

    Seeing Pedro Martinez’ fastball on that “most homerable” list … weird.

    Vote -1 Vote +1

    • Belloc says:

      Not weird when you consider that the data starts in 2007. Martinez didn’t pitch much in his final three seasons, and his fastball hovered around 88 mph. He was a completely different pitcher at that point in his career.

      Vote -1 Vote +1

    • Schuxu says:

      Even Felix Hernandez grand slammed off of that pitch.

      Vote -1 Vote +1

  6. Bip says:

    I’m not even convinced that was a bad pitch. I’m totally certain that Kershaw and Butera had a clear scouting report on Hicks that said in big letters “CANNOT HIT THE HOOK.” Matchup aside, Kershaw has the kind of big curveball where “hanging” it isn’t necessarily a bad thing. The difference in velocity and movement between his curve and his fastball is so huge that when he throws his curve through the middle of the zone, hitters often just watch it go by. A query on Baseball Savant backs this up: over his career, about 50% of his curveballs in the zone have been taken, or the swing rate on his curves in the zone has been only about 50%. Whether this is normal for curveballs I don’t know, but if he is not routinely punished for tossing a curve down the middle, then for him you couldn’t really call it a bad pitch.

    Vote -1 Vote +1

    • Belloc says:

      In his career, more than 61% of the time Kershaw throws his curve ball out of the strike zone. And hitters only bat .126 against the curve, with an OPS of just .292. Unless there’s a two-strike count, a hitter probably shouldn’t bother to swing at a pitch that is so difficult to hit and will most likely be called a ball.

      With a two-strike count, a hitter isn’t likely to take a hanging curve, because he is less likely to be sitting on a fastball. The only way a hitter takes a hanging curve ball with two strikes is if he is sitting on a fastball and gets completely fooled.

      Vote -1 Vote +1

      • Bip says:

        Baseball Savant is great. I’m testing what you are saying right now. Here is what I’m finding. Each of the below breakdowns are with two strikes.

        On curves down the middle: 17% take rate, 17% hit rate, 17% whiff rate. We’re only talking about 47 pitches here, but still certainly a good pitch for him

        On curves middle-up in the zone (what are often called “hangers”): 31% take rate, 11% whiff rate, 11% hit rate. Still pretty good results, this time on 206 pitches.

        On curves anywhere in the zone: 25% take rate, 18% whiff rate, 9% hit rate.

        So you’re absolutely right that with two strikes, batters have been less likely to take his curve. However, the results support my argument as well, that Kershaw’s curve is so effective, he doesn’t even have to bury it in the dirt to get good results. Part of this reason is also that he only uses it about 10-15% of the time, and the slider is his primary breaking ball, so if the curve was his primary breaking ball, he likely wouldn’t have as much room to leave it up in the zone.

        Vote -1 Vote +1

      • Yehoshua Friedman says:

        With a two-strike count, why on earth should a hitter with above developmentally-challenged intelligence be sitting on a fastball? The pitcher is going to be nibbling with an advantage like that and is very likely to throw a curve or an offspeed pitch.

        Vote -1 Vote +1

  7. Twm says:

    I misread the headline and so was expecting a post on the most and least honorable pitches. Uncertain what that might mean, my interest was piqued. And now, despite a thoroughly entertaining post, I am a bit disappointed.

    Vote -1 Vote +1

    • Micah Stupak says:

      I also misread it as honorable, which I figured at the very least was a pretty Notgraphs thing to do, and now I hope to see said article soon.

      Vote -1 Vote +1

      • Dave Cornutt says:

        I read that too, and my first thought was, “What would be a dishonorable pitch? The screwball? The Eephus pitch? And where does the knuckleball fit in — is it honorable, dishonorable, or a completely different ethos?”

        Vote -1 Vote +1

  8. Hurtlockertwo says:

    As a Giants fan, this was refreshing since Kershaw has owned the Giants in the past. Very odd of the Giants to have a limited outcome second baseman, just the opposite of Scutero who hits everything.

    Vote -1 Vote +1

  9. Tommy "5 Runs All Earned" Hunter says:

    No park factor adjustment? I’m disappoint

    Vote -1 Vote +1

    • No real benefit to making things extra complicated. A ball either left the yard or it didn’t. If you start to overthink, you only overthink, and nothing ever gets done.

      Vote -1 Vote +1

    • MrKnowNothing says:

      I ran them myself just now. Kimbrel’s curve still gives up zero HRs in every park. And on the moon.

      Vote -1 Vote +1

    • Jason B says:

      Your question I can take or leave, but your user name is so, so apt. It’s like the second coming of Todd Jones: 1 IP, 2 H, 1 BB, 1 ER, 1 K, SV! EVERY. DAMN. NIGHT.

      Vote -1 Vote +1

  10. Stank Asten says:

    I prefer the term dishomerable.

    Vote -1 Vote +1

  11. arc says:

    Wouldn’t this be better measured if it were adjusted for the % of time the pitch is thrown for a strike?

    I don’t know the answer; I’m just thinking that we see a lot of breaking balls on the list which may be disproportionately thrown out of the zone. In that case, they may still the least homerable, but not in a way that is necessary meaningful. The least homerable pitch is a ball in the dirt.

    Vote -1 Vote +1

  12. Rick says:

    Speaking of Kimbrel’s breaking ball, can anyone explain why pitch f/x suddenly started classifying it as a knuckle curve this year? In all previous seasons, it was typically called a slider.

    Vote -1 Vote +1

    • Bip says:

      No, PitchFX classifications are bordering on nonsense. Kimbrel’s pitch is certainly not a slider. It’s movement profile makes it squarely a curveball. He also throws it with a knuckle-curve grip.

      That said, we have to appreciate the fact that these classifications are not straightforward. The reason PitchFX was likely confused is that, although Kimbrel’s pitch moves like a curveball, it is as fast as a slider. This raises the question of whether a pitch with curveball movement and slider velocity is actually more like a slider or a curveball in terms of how the batter reacts. If a player is good at hitting curves and bad at hitting sliders, would he be comparatively good or bad at hitting Kimbrel’s pitch.

      In my opinion, it makes sense to classify pitches based on movement and *relative* velocity. Kimbrel throws so hard that a pitch in the low-80′s for him is like a pitch in the mid-70′s for another pitcher. Surely it is useful to further classify them based on absolute velocity, but for many purposes, it is not necessary.

      Vote -1 Vote +1

  13. Stephen says:

    So Kimbrels curve guves up the least amount of hrs, but whos pitch generates the most outs? I think this is a nice segway to that question. That is, if you have the time to calculate it.

    Vote -1 Vote +1

    • peopletocakeratio says:

      i’m just curious what you mean here. just on balls in play or including called strike three and swinging strike three?

      like, i could believe some pitches that get a higher percentage of whiffs/swing may actually get hit harder when put in play than some pitches that are simply tough to square up the barrel on. meaning, pitches that are “strikeout pitches” are that way because of an element of deception. but when a batter is looking for that pitch, they can square it up and line it somewhere. whereas a sinker or a cutter, which may not generate the rate of whiffs as a slider, may be thrown more often and induce a higher contact rate (albeit poor contact), leading to more outs.

      essentially, swing-and-miss only leads to an out with 2 strikes, but contact leads to balls in play in any count.

      granted, there are probably other considerations to keep in mind, but i like the idea of finding out the percent of pitches thrown (for a given pitch) which lead to outs.

      Vote -1 Vote +1

  14. RY says:

    I was watching that game yesterday, and distinctly remember that AB. When it got to 0-2 I said to myself, this is why Hicks is batting under .200, he doesn’t know that Kersh is going to throw a curve ball right now to finish him off. Which I believe he threw on the first pitch. Looks like he can pay some attention!

    Vote -1 Vote +1

  15. The Humber Games says:

    Dang, I was really excited to find out what Least Honerable pitch meant…I need to read titles more closely

    Vote -1 Vote +1

    • jruby says:

      The least honorable pitch is probably the one that most humiliates and demoralizes opponents.

      So… still Kimbrel’s curve.

      Vote -1 Vote +1

      • Jason B says:

        Maybe Gaylord Perry’s spitball? No honor in that, except among old-timey baseball types who think cheating began in approximately 1994.

        Vote -1 Vote +1

  16. Victor says:

    It would be interesting to see the breakdown for each pitch type instead of just seeing the overall result for all pitches

    Vote -1 Vote +1

  17. Beardman says:

    I’ve seen at least one other commenter point this out, but Pedro Martinez almost certainly made the list due to the time frame of the data, since he was generally pretty ineffective in ‘ 08 and ’09.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current ye@r *