Freezing, with Clay Buchholz

Though it’s not yet set in stone, Clayton Kershaw is probably going to win the ERA title, and he’s probably going to win the National League Cy Young Award, because Cy Young Awards frequently go to the guys with the ERA titles. Yet Kershaw isn’t the only starter with an ERA under 2, after you lower the minimums a little bit. There are actually three of them, one of whom is Kershaw, who is demonstrably and understandably amazing. One of them is Jarred Cosart, whose ERA is a hell of a lot more promising than the rest of his numbers. And the third is Clay Buchholz. Kershaw stands out because his adjusted ERA is nearly half the league average. Buchholz stands out because his adjusted ERA is two-thirds Kershaw’s. And Buchholz, now, is back from injury.

Any ERA that low, for a starter, is unsustainable, especially for a starter in the American League and Fenway Park, but Buchholz’s xFIP- is way improved. His FIP- is way improved. Something that’s helped him prevent runs is that he hasn’t surrendered many dingers, as dingers count for runs automatically. But more interesting than that is Buchholz’s strikeout increase. Used to be that Buchholz’s strikeout rates didn’t quite match the quality of his stuff, or at least that was the perception. This year he’s taken a leap forward, striking out a quarter of the batters he’s faced. Changes in strikeout rate capture an analyst’s attention, and in Buchholz’s case, there’s something in particular that’s been driving this.

Now for the part where I give away the ending:


Bunts aside, there are two ways to get a strikeout: you can make a guy swing and miss, or you can throw a strike the batter doesn’t swing at. So, there are swinging strikeouts, and there are called strikeouts, and while it’s the swinging strikeouts we like to watch as indicators of dominance, all strikeouts count the same and there’s an art to leaving hitters frozen. In the image above, you can see Buchholz’s rates of swinging strikeouts and called strikeouts. His swinging strikeouts, this year, are more or less where they’ve been for a while. His called strikeouts are way, way up — to the point where they’re almost equaling his swinging strikeouts. The league average is that a quarter of strikeouts are called. This year, we find Buchholz around half.

This season, 232 pitchers have faced at least 250 batters. That’s an arbitrary cutoff, but look how willing you are to accept it. Buchholz ranks 164th in swinging-strikeout rate, between Alexi Ogando and Ronald Belisario. He ranks first in called-strikeout rate, well in front of Dale Thayer and Cliff Lee in second and third. Out of Buchholz’s plate appearances, 12.0% have ended with called strikeouts. The next-best mark is 10.5%. The highest mark last season was 11.0%. Buchholz’s own mark last season was 5.9%. That gets more into the chart above.

It’s not just what Buchholz is doing; it’s that he hadn’t done it before. Around baseball, there are 165 pitchers who have faced at least 250 batters in each of the last two seasons. Buchholz’s called-strikeout rate has increased by 6.2 percentage points, which is far and away the biggest increase in the majors. Chris Sale is the only other pitcher with an increase above 4 percentage points. Buchholz, suddenly, is generating a bunch of called third strikes, and as a result he’s looking more like the pitcher he was always supposed to turn into.

Digging into the platoon splits, we see increases against both righties and lefties:


Lefties are at 10%, where before they’ve hovered around 6%. Righties, though, are way up near 15%. This season, Buchholz has struck out 24 righties looking. That’s a career-high, despite all the missed time. Compared to last year, Buchholz has two more called strikeouts against righties, in four-ninths the sample. Of Buchholz’s 39 strikeouts of righties so far, just 15 have involved a swing at strike three. This feels like something truly extraordinary.

To try to get some sense of Buchholz’s strategy, we can plot his called strikeouts against the league’s called strikeouts:



Against righties, Buchholz has gotten his called strikeouts over the outer half, or near the outer edge. Against lefties, he’s alternated inner and outer edges. Know that the strike-zone boxes are included just as reference points, and they’re hastily-produced approximations.

All of his called strikeouts against lefties have come on fastballs, sinkers, or cutters — or, some variation of heat. Most of the inside ones have been sinkers tailing back over the inner edge. Most of the outside ones have been cut, with Buchholz targeting the back door. A sample, featuring James Loney:


His three most inside called strikeouts against righties have come on curves. For much of the rest, we find those back-door sinkers, with Buchholz taking full advantage of his horizontal movement. A sample, featuring Jayson Nix:


If you take a look at Z-Swing%, Buchholz, rather unsurprisingly, is in the group with the lowest rates. Batters have taken a lot of strikes, and not just in two-strike counts. More interestingly, from last year, Buchholz’s O-Swing% is unchanged. His Z-Swing% is down from 63% to 57%. So batters have swung less often, but only at pitches in the zone, which is an encouraging thing to see for a pitcher. Something else that most certainly helps: Buchholz has been given a more friendly zone this season. One in eight pitches out of the zone have been called strikes, against a league average of one in 14. Implied is that Buchholz has pitched to good receivers, and implied also is that Buchholz has done well pitching to spots right near edges. That’s where called strikes are found, and that’s where called strikeouts are found.

Looking back, Buchholz has generated a remarkable amount of called strikeouts. The big question, then, is what to make of this for the future. Swinging strikeouts tend to be more stable year-to-year, and those are taken to be a better indicator of pitcher ability. Called-strikeout rates bounce around. Since coming off the DL, Buchholz has generated just two called strikeouts over two starts. He didn’t get any on Sunday. But called strikeouts still aren’t random, and the correlation isn’t that much weaker than the one for whiffs, so it’s worth considering that Buchholz might have a new skill, now. He might have figured out a way to boost his strikeouts while leaving his contact rates virtually unchanged. It feels like there would be more reliable ways for a pitcher to get better, but if you think I actually really understand pitching, man, thanks, I appreciate it, but, nope. If Buchholz is developing mastery of the edges, then he’s got his job figured out.

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.

17 Responses to “Freezing, with Clay Buchholz”

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

    it’s really interesting to me how wide those called K plots are? Wouldn’t you expect more variation in high/low strikes, due to differing batter height?

    I knew about the ‘lefty strike’ outside off the plate, but look at all those called Ks to RHB both inside and outside, where the top and bottom of the zone seem to be firm.

    Am I misinterpreting the plots?

    Vote -1 Vote +1

    • Andrew says:

      He mentioned that the strike zone boxes were hasty approximations, but yeah, it does seem very wide to me too. I’d be interested in seeing gifs of some of those outlying pitches, just to see the circumstances.

      Vote -1 Vote +1

    • Tim says:

      Maybe it is because Pitchers tend to be pitching to swings above and below the zone. You catch Hitter looking on the corners and swinging on high heat and splitters in the dirt, etc.

      Vote -1 Vote +1

  2. Bryan says:

    Awesome piece, Jeff, thanks.

    Vote -1 Vote +1

  3. That top graph should get its own spin-off xGraphs site.

    Vote -1 Vote +1

  4. MGL says:

    Are those circles in the called strikes one pitch each? Are we talking about 20 called strikeouts from each side of the plate? I mean, can we really conclude a whole lot from sample sizes that small?

    Vote -1 Vote +1

  5. derp says:

    Do we all remember what happened to Vance Worley?

    Unless we see an uptick in Swinging strikes this will almost certainly be a fluke occurrence.

    Vote -1 Vote +1

    • Spit Ball says:

      Way different repertoires and talent levels. I would give to you that Buccholtz given his age and fragility level is unlikely to keep it up. Did I add AL east? But Worley and Buck? C’mon

      Vote -1 Vote +1

  6. pft says:

    Buchholz issue going forward will be durability. He has had problems staying healthy and pitching 200 innings, something he has yet to do. With only 14 starts he will barely break the 100 IP mark.

    He probably needs to go back to pitching to contact more. K’s are nice but build up the pitch count, and reduce IP.

    Vote -1 Vote +1

    • Bip says:

      He probably needs to go back to pitching to contact more. K’s are nice but build up the pitch count, and reduce IP.

      That’s not true. K’s also are basically guaranteed outs. Pitching to contact means more hits, which means more batters faced, which definitely means higher pitch count. Secondly, pitching to contact wouldn’t help durability, it would (if it were true that it reduced pitches per inning) just allow him to squeeze a few more innings out of an injury-shortened season.

      Vote -1 Vote +1

  7. Eric M. Van says:

    Of the 148 pitchers with 90+ innings, Buchholz has (according to pitch/fx values/100) the 8th best 4-seamer, the 3rd best 2-seamer or sinker, the best cutter (2nd best cutter or slider), the 16th best curve, and the 22nd best changeup (minimum throwing each pitch 4% of the time).

    The next level of investigation into this phenomenon might therefore be a look at pitch selection and sequencing. It makes sense that someone with five pitches that good (and it might actually be six, if they’ve misclassified enough splitters as changeups) could freeze a lot of batters if the two-strike pitch-calling was adept enough.

    (Kershaw, since you want to know, has the 3rd best 4-seamer, 2nd best curve, and 45th best slider. Two dominant pitches, but three fewer great ones than Buchholz.)

    Vote -1 Vote +1

    • Bip says:

      What I noticed about those values, is that they basically are always highest for the pitchers with the best ERA. As long as Buchholz has that ridiculous ERA, his per/pitch values are going to look great, because those values are based partially on outcomes, and so is ERA.

      One would think that true measures of pitch quality would be a determiner of ERA, but these are not true measures of pitch quality, so causation is actually reversed. If next year, Buchholz’s KL% regresses, and his HR/FB rate increases from “ridiculously low”, then we will see those pitch values come down.

      Vote -1 Vote +1

  8. SKob says:

    Not trying to give a manager credit for making batters take more 3rd strikes, but John Farrell knows pitching. If the improvement from Clay is more game planning and sequencing than a change in stuff or luck related, I would bet there is some ability to sustain the improvement based on game plan changes made by the coaching staff! Even if that credit goes to improving Salty’s game calling, coaching should be mentioned.

    Vote -1 Vote +1

  9. Mark says:

    The platoon split chart left me a bit confused…I get the 15% called strikeout percentage against righties, but hen it goes on to say that 24 of Bucholz’s 39 righty Ks have been looking, which would make the number 62%. Am I missing something?

    Vote -1 Vote +1

    • Bip says:

      The percentage is of plate appearances. So, amazingly, 15% of his plate appearances against righties this year have ended in a called strikeout.

      Vote -1 Vote +1

  10. Jamie says:

    I have a question (or two) about the strikeout plots. Those plots are a 2 dimensional represention of a 3 dimensional space (i.e. the strike-zone). So, when they plot the pitches, is it based upon where the ball crosses the front of the plate, back of the plate, or where the catcher recieves the ball?

    Another question, what is the distance from the front of the plate to where the catcher receives the ball?

    When I look at those gifs for Buchholz, I see a ton of late movement. Basically the ball starting outside the zone and swinging back into the zone. Which triggered my question about where in space the pitches were plotted.

    Vote -1 Vote +1

    • Bip says:

      1. Front of the plate.

      2. Not sure, but it looks like 3-4 feet? Obviously the catcher can adjust that distance somewhat.

      It’s hard for viewers to really tell where the strikezone is. We’re basically limited to seeing where the catcher catches the ball, having no depth perception from the TV. The one thing we can to is kind of interpolate where a pitch crossed the plate from where the catcher caught the ball and how it moved.

      So, in the second GIF, the catcher caught the ball on the outside part of the plate, and the ball was moving towards the center of the plate at the time. Therefore, the ball was more outside when it crossed the plate than when the catcher caught it, probably.

      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>