Last year, Clay Buchholz seemed poised for a breakout. He posted his best K/BB ratio of his career by far. An already solid groundball rate was buoyed by an awesome new changeup. He also posted a 2.68/3.30/3.35 in FIP/xFIP/SIERA, and all of those three were career-best marks. His ERA matched up with those metrics, at 3.26. The best part was that he was finally healthy!… until a flexor strain in July sidelined him for the rest of the year.
Since returning this year, Buchholz has been a mess. In 101.1 innings split between the rotation and the bullpen, he has a 5.42/5.41/5.47/5.29 ERA/FIP/xFIP/SIERA. He has posted the worst K-rate of his career (5.51 per nine innings), coupled with a poor walk rate (3.64 per nine). His groundball rate is now below average, down 7.5% from last year’s solid 48.3% to 40.8% this year. He’s understandably allowing a ton of homers, at 1.51 HR/9. What went wrong?
My first thought was that he was still suffering from the flexor strain or some other injury. Forearm injuries are notorious amongst pitchers. They ended Cliff Lee’s career, and they’re often thought of as a precursor to elbow problems. Thus, I consulted Jeff Zimmerman’s Pitcher Injury Factors on his website, Baseball Heat Maps. Buchholz’s velocity was normal, but something else struck me. When a pitcher’s release point becomes inconsistent, they may be trying to pitch through an injury, changing their delivery to avoid pain. Jeff’s model works best with four-seam fastball data, so Buchholz was tricky because he mainly throws two-seamers and cutters. Take this with a grain of salt since I used two-seamer data, but his release points have gotten very inconsistent late in the game this year, more so than last year:
Speaking of Buchholz’s release point, there’s something else of note here: his arm slot has been dropping. Consider this graph of his vertical release point:
And as his vertical release point has dropped, his arm has moved out more towards the side, except for a blip in the middle where I’m guessing he started pitching on the other side of the rubber (maybe in hopes of compensating for his dropping release point)–so here is what his horizontal release point looks like:
When your arm slot drops, you tend to get more horizontal movement to the arm-side and less vertical movement. Consider the horizontal movements on his pitches over the last two years:
We can ignore the slider and the splitter because they haven’t been thrown much. You’ll notice that his cutter has lost much of its cutting action. His curve has lost some bite. His four-seamer runs so much to the arm-side that its basically a two-seamer now. The change and sinker haven’t changed much.
Obviously, if you’re losing your ability to throw strikes, then that might mean you’re injured, but for Buchholz, it could also be due in part to him struggling to control his pitches that have new movements. His change in release point could even be due to injury, also. However you look at it, for his two-seamer this year, the zone percentage has gone way down from last year:
I decided to investigate this dip in zone percentage further. It turns out that this held true for his other fastballs as well, and his entire repertoire in general. Take a look at the differences in plate discipline against Buchholz the last two seasons:
His Zone% has gone down over 6%. Hitters aren’t fishing for balls outside the zone anymore, and they’re swinging less in general. But when they do swing, they’re making more contact. Buchholz is having trouble getting first-pitch strikes. And naturally, his swinging strike rate has gone down. These are some troubling trends. However, the last time that Buchholz had a Zone% this low was in 2014, a year in which he was more effective (at least in terms of the ERA estimators). Take a look at the similarities between this year and 2014:
Wow. Every stat differential is two percentage points or less. This got me thinking that, despite some tangible changes, last year might have just been an outlier, and Buchholz is just regressing to the mean. Now, here are his last two seasons again, but this time, I included his career averages:
This validates my outlier theory, as this year and 2014 are very similar to his career averages, while 2015 is not at all. What about the year that Buchholz had a 1.74 ERA in 108.1 innings? Was that year an outlier too?
2013 was very similar to his career averages, except for one stat: Z-Swing%. It was much lower. Basically, hitters were taking more strikes against Buchholz that year, which allowed him to be more effective. I’m not saying that’s the only reason he was effective that year, but it certainly seems to be a factor.
So is Buchholz as maddeningly inconsistent as everyone says he is? Or, has he just gotten lucky a few years and is really… not that good? I’m going to go with the latter. Although he does have a superb changeup, his other stuff is indeed really not that good. While I do think that there is a chance he’s injured this year or that maybe he’s struggling with his dropping arm slot, I still don’t think his true talent level is where he was last year. This year may well be an outlier on the opposite end of the spectrum, though: since he’s pitching the same way he always has, I would have expected no fewer than 6.5 K’s per nine innings, and no more than 3.5 walks per nine just by looking at his plate discipline stats.
Data from FanGraphs, Baseball Heat Maps, and Brooks Baseball. Picture from Boston.com. Stats are as of morning 8/23/16.
Thanks for reading!
Written by Alex Eisert.