The Curious Case of Clay Buchholz

Prior to a neck strain that landed him on the disabled list on June 18, 2013, Boston Red Sox starting pitcher Clay Buchholz had a phenomenal season, posting an ERA of 1.71 in 84.1 innings pitched. During this period, Buchholz had struck out 24.9% of the batters he had faced during that period, while walking only 8.9% of batters faced during that period. Although he had an abnormally high left on base percentage of 84.1% during that span (~72% is normal and is largely luck-based), he still posted a FIP of 2.47 and a SIERA of 3.40, leading to talk of an American League Cy Young Award before suffering a neck strain. Buchholz did not play again until September 10, 2013. From that point on, Buchholz started four games, posting an ERA of 1.88, a FIP of 3.88, and a SIERA of 4.33. One may waive this off as small sample size, but even at a basic level, there was a noticeably uptick. Other things that differ before and after the injury include a downtick in groundball rate (from 48.8% to 44.1%), an uptick in fly ball rate (from 30.1% to 36.8%), and his home run to fly ball rate more than doubling (from 3.2% to 8.0%). For contextual purposes, ideally a pitcher should want his groundball rate fairly high, as there is a zero percent chance that a ball goes out of the stadium via a home run if it is hit for a ground ball, whereas a fly ball is infinitely more likely to leave the stadium, although one may see more hits given up via groundballs. It was not just his fly ball rate increasing, but also his home run to fly ball rate skyrocketing, which suggests he either struggled locating the ball in the starts post-injury or harder hit contact was made against Buchholz post-injury. Again, even if you want to dismiss it as small sample size, those are figures worth noting at the very least. For the most part, his velocity was down nearly across the board for the two different periods in 2013.

Before
Velocity of Buchholz in 2013 Prior to DL Stint


After
Velocity of Buchholz in 2013 after Returning From DL


What is the point discussing Buchholz’s 2013? Well, it is important to establish the baseline prior to any injury where Buchholz looked like a 1/2 pitcher, rather than what he has become this season. 2014 has just been a disaster for Buchholz, posting an ERA of 6.20 in 101.2 innings pitched, which sounds like a major downtick in terms of performance, when it is less than that ERA figure suggests. Buchholz has a FIP of 4.62 and a SIERA of 4.39. This is in part due to an abnormally high BABIP this season of .340 (career BABIP of .287, which is more in line with the law of averages when it comes to BABIP) and a left on base percentage of 62.6% (should sit roughly around 72%), both figures are essentially based on luck, or at least a heavy portion. Before going too in-depth on the overall 2014 season thus far, it should be noted that injuries have broken this season up, similarly as to what was seen in 2013. After pitching three innings against the Atlanta Braves on May 26, 2014 in which Buchholz gave up six earned runs and walked eight, Buchholz was then placed on the 15-day disabled list with what was termed a hyperextended left knee. Prior to the injury, Buchholz was disastrous, posting an ERA of 7.02 in 50 innings pitched. The ERA is relatively deceiving, posting a FIP of 4.90 and a SIERA of 4.70, due to a left on base percentage of 63.2% and a BABIP of .384. Those figures do not completely excuse the complete change from early 2013 Clay Buchholz and early 2014 Clay Buchholz. Going from a FIP of 2.47 to 4.90 is a very significant jump. Looking at pitch type and velocity changes, there are some noteworthy things that stand out comparing his complete 2013 season to the April 5, 2014 through May 26, 2014 season. His fastball usage is down from 49% in 2013 to 46.9% in the 2014 period. Fastball velocity fell from 91.9 mph on average in 2013 to 91 mph in the 2014 period, so roughly one mile per hour. His cutter usage jumped up from 23.6% in 2013 to 29%, which sounds well because the cutter is generally one of the more effective pitches in baseball, but his cutter velocity dropped with increased usage, from 87.5 mph on average in 2013 to 86.8 mph in 2014. Let us look at batting averages per quadrant regarding only Buchholz’s cutter.

cutter2013
Batting Averages on Buchholz’s Cutter in 2013


cutter2014a
Batting Averages on Buchholz’s Cutter from 4/5/14 to 5/26/14


As you can see, despite increased usage (when factoring in the length of the 2014 period), contact is being made more frequently on the cutter in the early part of this year as opposed to the 2013 season. All three horizontally-centered quadrants and the bottom-center quadrant are showing that his cutter is less effective, despite less than a one mile per hour drop between 2013 and the 2014 period. Despite only a 0.4 mph reduction from 2013 to the 2014 period, Buchholz’s curveball usage dropped from 14.2% to 10.2%. Let us take a look and see if it is a similar situation to the cutter, where he began to give up more hits in 2014.

2013cu
Batting Averages on Buchholz’s Curveball in 2013


                                                                                                                    2014cua                             
Batting Averages on Buchholz’s Curveball from 4/5/14 to 5/26/14


Although we are dealing with a small sample size, as the curveball is only used 14.2% of the time in 2013 and 10.2% in the 2014 period in discussion, Buchholz’s curveball appears to not have lost any of its effectiveness, but perhaps has become more effective, despite the slight downtick in velocity (from 77.4 mph in 2013 to 77 mph in the 2014 period in question). Even if you look at whiff rates (the number of swings and misses), little has changed.

Whiff2013Cu
Whiff Rate on Buchholz’s Curveball in 2013


                                                              Whiff2014CuA                                           
Whiff Rate on Buchholz’s Curveball from 4/5/14 to 5/26/14


As you can see, we are dealing with a far smaller sample size in the second image, but for the most part, the swings and misses are fairly consistent. Why exactly is there a decrease of usage of a pitch that, for the most part, seems to be fairly effective? Pitch selection is not an injury issue, but a decision issue. Bob Tewksbury, one-time the mental skills coach on the Boston Red Sox, left the team in January of 2014 to assume a position with the Major League Baseball Players’ Association. It was made public that Buchholz reached out to Tewksbury after his hyperextended left knee injury on May 26, 2014, and talked to Tewksbury, who was in the Boston club house the very next day. Could the shift in certain pitch usage be a mental issue, attributed to the departure of Tewksbury? At this point, that is the obvious reason. His changeup usage is down from 10.6% in 2013 to 8.2% in the 2014 period in discussion, despite velocity actually increasing on the pitch in 2014. Buchholz’s average changeup velocity was 80.3 mph in 2013, whereas it rose to 82.2 mph in the 2014 period. Perhaps the pitch is less effective, which is why he is throwing it less, so let us take a look.

bachange bacurve2013
Batting Averages on Buchholz’s Changeup in 2013


                                                                        BAchange14a                                 
Batting Averages on Buchholz’s Changeup from 4/5/14 to 5/26/14


It would be a quick assumption to assume that the same thinking process behind fewer curveballs would be the logical reason that Buchholz is throwing the changeup less during this period, but that would be incorrect, as whiff percentages are down across the board, even taking into account the 2014 relatively small sample size.

Whiffchangeup2013
Whiff Rate on Buchholz’s Changeup in 2013


WhiffChangeup2014a
Whiff Rate on Buchholz’s Changeup from 4/5/14 to 5/26/14


As you can see, Buchholz’s changeup in 2013 delivered a large number of swings and misses, being quite effective, while his whiff rate is disappointingly low in the 2014 period, even taking into account the sample size. How could a pitch that has seen a major jump in velocity (80.3 mph to 82.2 mph) actually produce fewer swings and misses? I can only partially answer that question, as Buchholz appears to have left the pitch up far too often on average in May of 2014.

VertCh13
Vertical Location of Buchholz’s Changeup in 2013


VerCh14a
Vertical Location of Buchholz’s Changeup from 4/5/14 to 5/26/14


  Zero is essentially the very center of the strike zone, so you can see that the changeup was left down, or below the center of the strike zone, consistently in 2013. It started out seemingly well in 2014, although it was higher than any single month in the previous year, then things fell apart in May 2014, as his changeup was kept well above the center consistently. I will address the state of the changeup later, but for completeness, we must look at pitch that has seen its percentage usage double. The use of the splitter has seen a noticeably jump, being used only 2.5% in 2013 to 5.8% in the 2014 period in discussion. Velocity has seen a slight uptick from 85.1 mph in 2013 to 85.6 mph in the 2014 period in discussion. It is pretty clear that Buchholz should lower his splitter rate back down to 2.5% after viewing batting averages against the pitch.

Splitter2013
Batting Averages on Buchholz’s Splitter in 2013


 Splitter2014A
Batting Averages on Buchholz’s Splitter from 4/5/14 to 5/26/14


  It does not take additional information to show that the increased usage of the splitter has been largely trouble to Buchholz. Why throw a curveball instead of a splitter during this period? That is Clay’s seemingly poor decision making, not injury. Why do I constantly use the April to May of 2014 period for Buchholz? Well, as previously stated, in his start on 5/26/14, he suffered a left knee injury, so I wanted to clear out the possibility that injury had derailed Buchholz as much as possible, even if it meant using a small sample size. What is notable is that Buchholz improved from the ERA of 7.02 during the 4/5/14-5/26/14 period to an ERA of 5.40, with a FIP of 4.36 and a SIERA of 4.05, in 51.2 innings pitched since returning on June 25. Since returning his fastball usage in terms of a percentage is down to 43.1%, also seeing a bump in velocity to 92.3 mph on average. His cutter usage is down to 23.9% of all pitches, also seeing a bump in velocity to 88.7 mph (up from 86.8 mph earlier in the year prior to the knee injury). His curveball has made up 17.3% of all pitches thrown since returning, which is great to hear as it was underutilized, despite being so effective earlier in the year. There has been a slight bump up in terms of curveball velocity from 77 mph earlier in the year to 77.7 mph since returning. The changeup usage is especially interesting, as it, despite being a somewhat lackluster pitch in April and a poor pitch in May, has increased in usage in terms of a percentage up to 14.2% (significantly up from 8.8% earlier in the year and the 8.2% in 2013). Buchholz’s average changeup velocity sits at its highest ever since returning from the left knee injury, at 82.4 mph. There appears to be a slight improvement since then in terms of opposing batting averages and whiff rates.

2014ChangeupbaB
Batting Averages on Buchholz’s Changeup from 6/25/14 to 8/3/14


 2014ChangeupWhiffB
Whiff Rates on Buchholz’s Changeup from 6/25/14 to 8/3/14


For the most part (excluding Sunday’s start, Buchholz’ only start in August), Buchholz has managed to keep the ball down in terms of vertical location upon reaching the plate.

VertCh14b
Vertical Location of Buchholz’s Changeup from 6/25/14 to 8/3/14


Make no mistake, his start on August 3, 2014 was bad and the changeup location indicates that, leaving the ball above the strike zone. If this trend were to continue, we should experience some trouble, but disregard that for the time being. Buchholz’s splitter now makes up only 1.6% and has seen a bump in velocity, averaging 86.2 mph since returning. As previously mentioned, the splitter has not been a good pitch for him, so this usage reduction would likely indicate success.

Buchholz’s 2013 season produced a groundball rate of 47.7%, whereas his stint in 2014 prior to injury produced a groundball rate of 42.3%. While a higher groundball rate may translate to more hits, it reduces the risk of a home run to zero. Since returning from injury in 2014, Buchholz’ groundball rate has shot back up to 47.2%, which is good. In 2013, Buchholz’s percentage of fly balls sat at 31.8%. It subsequently shot up in the start of the 2014 season to 37.4% before dropping down to 32.1% since returning on June 25, 2014. One thing that has not been the same since the 2013 season is his home run to fly ball ratio. In 2013, it sat at a relatively low 4.5%. As expected, it increased to 10.3% in the first two months of 2014. One would imagine that it would follow the trend of everything normalizing, but it has not, as it has sat at 11.8% since returning from the left knee injury. Every other batted ball metric has normalized to near 2013 levels since returning, except for the home run to fly ball rate. At this point in time, it is worth noting and being slightly concerned over, but I would give Buchholz the remainder of the season and see if we notice some regression to the norm.

Sorting out what of this is injury and what of this is mental errors, now that Bob Tewksbury has departed the Boston Red Sox, is somewhat tough to do. There are no reports in 2014 that indicate that the neck injury was of concern for any part of the 2014 season. It also appears the left knee injury has healed well, as he is continuing regressing to his true talent level. I would assign the departure of Tewksbury as playing a big role in Buchholz’s missteps of 2014. Obviously, it is tough to place an objective number, but after taking some time off to heal in 2014 and talk with Tewksbury, Buchholz has performed well, barring his most recent start on August 3, 2014. Maybe the Red Sox need a new ‘mental skills coach’ or Clay can hire a sports psychiatrist to deal with his on the field issues, as others in athletics have done so.

Where does Buchholz’s play out long-term? Well, that is tough to say as well, but barring any mental on the field difficulties and injury, he should be the same player he was from 2009-2012. His hot start last year was more of an aberration than anything, as his strikeouts per nine innings sat at 7.98, uncharacteristic of his 6.83 career strikeouts per nine innings rate. Even if we were to look over his 2014 season independent of injury or Tewksbury, Buchholz has been relatively unlucky and should regress to that norm of an ERA between 3.00 and 4.50, as mechanically he has been his usual self since returning. Is he a staff ace? He is not. Is he a 1/2 starter? He is still not even that. Is he a mid-rotation starter? Yeah, I would peg him in about that role as a 3/4 starter, perhaps a back of the rotation starter on a team with a great pitching staff. What is rough about his current situation is that Buchholz, a man who has had mental sports performance issues, will have to endure criticism as he is now the number one pitcher on a big-market popular team, even though he should not be.