Who Will Follow Dallas Keuchel’s Lightly Traveled Path?

As we anticipate Dallas Keuchel finding a new home this offseason, we should probably be wondering more about where he will be locating his sinker in 2019 than where his moving van is headed.

Sinker location has been key to Keuchel’s success, and his mostly-consistent five-year run has been nearly unique this decade. When he shaved more than two runs off his ERA from 2013 to 2014, Keuchel had a dramatic breakout that may have been most notable for what didn’t change. He didn’t throw substantially harder. He wasn’t notably better at throwing strikes or getting whiffs, chases or freezes. Keuchel did induce more grounders, but his biggest improvement came in his avoidance of hard contact. His rate dropped from 29.3 percent to 19.7 percent, and in the four seasons that followed, Keuchel continued to demonstrate that skill — as long as his sinker was down and away from righties.

Over the last ten seasons, it has not been all that common for a pitcher who was not already averse to hard contact to experience even a modest improvement in hard contact rate. Since 2008, there have been a total of 534 pitchers who have pitched at least 90 innings in three consecutive seasons and had a hard contact rate no lower than 10 percent (not percentage points) below the major league average in the first of those seasons. Only 22.7 percent (or 121 cases) resulted in a year-to-year decrease in normalized hard contact rate of at least 10 percent. (Note: For the purposes of this study, hard contact rate was normalized according to the major league average rate for a given season, due to substantial year-to-year fluctuations. The average rate was assigned a value of 100.) Going back to 2009, the normal major league hard contact rate has hovered around 30 percent, so we are typically looking at cases in which pitchers decreased their hard contact rate by at least 3.0 percentage points.

Keuchel’s 9.6-percentage point drop in 2014 represented a 29.0 percent decrease in normalized hard contact rate. Only six other pitchers in the past decade (or 1.1 percent of the pool, as described above) had met or exceeded that threshold. Even if we define a hard contact rate breakout more generously as a 20 percent decrease in normalized hard contact, only 32 pitchers (or 6.0 percent) make the cut.

While a one-year improvement in hard contact rate on the maginitude of Keuchel’s (or even a little smaller) is rare, it can make a difference for the pitcher who achieves it. Since Keuchel’s breakout season of 2014, he owns the lowest cumulative hard contact rate of any qualified starting pitcher with a 24.6 percent mark. Among the 16 starters who have compiled a rate of 28.0 percent or lower over that period, 12 of them have a HR/FB below the major league average of 12.0 percent. The four exceptions to this trend made all or most of their home starts in homer-friendly parks, and one of them — Aaron Nola — was extremely stingy with home runs (8.3 percent HR/FB) on the road.

Which Factors are Associated with a Sustained Hard Contact Rate Breakout?

While only 32 pitchers in our pool had a year-to-year decrease in normalized hard contact rate of at least 20 percent, just eight pitchers out of that group sustained a decrease of at least 20 percent from the year prior to the breakout to the year after the initial breakout. There was not a single common factor that explains each of the eight sustained hard contact rate breakouts, but the pitchers generally fell into one of two categories. The improvement in hard contact rate was either associated with locating pitches less frequently in the heart of the strike zone or changing one’s pitch mix. The only pitcher who did not fall into either category was Anibal Sanchez. His 2013-14 breakout was likely fueled by an increase in average four-seam fastball velocity and movement.

Breakouts Through Improved Location

When Keuchel, Lance Lynn and Kyle Lohse have been at their best, they were not only limiting hard contact, but ostensibly doing so by improving their pitch location. The table below shows the percentage of pitches each starter located in the heart of the strike zone in the season prior to the breakout (Year 1), the initial breakout season (Year 2) and the following season (Year 3).

The trend is clearest for Keuchel and Lynn, and the drop in Heart% for Keuchel is particularly notable for his 2015 Cy Young season. It’s also worth pointing out that when Keuchel’s ERA and HR/9 spiked in 2016, his Heart% crept back up to 18.3 percent. There is not an apparent trend for Lohse, but while he did not achieve a substantial decrease in Heart%, he did increase the rate at which he located pitches on the edges of the strike zone. From 2010 to 2012, Lohse’s Total Edge% rose from 24.4 percent to 28.2 percent to 28.4 percent.

A fourth pitcher, Felipe Paulino, clearly fits into this group, but zone location rates are unavailable for his pre-breakout season of 2009. However, these Pitch% heatmaps representing his 2009-2011 seasons should suffice to show his his growing inclination to vary his location.

Breakouts Through a Change in Pitch Mix

Remember A.J. Burnett’s resurgence when he went to the Pirates in 2012? He cut back on walks while inducing more grounders, but a decrease in hard contact was also an important factor in his improvement. He threw more sinkers and fewer four-seam fastballs, de-emphasizing his worst pitch in terms of ISO in the process. Burnett did not cut back on his curveball, but he did throw it with more horizontal movement. In effect, he took a pitch that was already effective at limiting extra-base hits and made it even better.

Though unheralded in fantasy, Scott Feldman was consistently good at avoiding home runs from 2009 to 2014. In that initial breakout season, Feldman made his cutter a more prominent part of his arsenal. He owned a 4.34 ERA over that stretch, but if he had been better at stranding runners (68.4 percent LOB%), he could have been more relevant in deeper leagues.

Andrew Cashner’s surprisingly strong 2017 season was fueled in part by the introduction of a cutter. It continued to be an effective pitch for him in 2018, but his fastball got hit much harder. In spite of that, Cashner’s 32.5 percent hard contact rate last season was more than three percentage points below the major league average.

2018 Hard Contact Rate Breakouts

If the past decade is any indication, we shouldn’t expect pitchers who have had a difficult time limiting hard contact to suddenly become proficient at it. We should also temper our expectations for those pitchers who took a step forward in this regard in 2018, as extended hard contact rate breakouts are exceedingly rare.

The following pitchers were among the best at limiting hard contact last season, so their early performance bears watching this spring.

Zach Wheeler stopped throwing his sinker in 2018, and his hard contact rate fell by eight percentage points. Concurrently, he posted a career-low 8.1 percent HR/FB and 3.31 ERA. Wheeler also got better at inducing whiffs and chases, so he could repeat as a top 25 starter.

Anibal Sanchez had a great comeback season in 2018, even though he did not experience dramatic improvements in his strikeout and walk rates. He did, however, lower his hard contact rate by nearly 10 percentage points in a season where the major league average went up by more than three percentage points. Sanchez followed the Feldman/Cashner route by throwing more cutters, and he also improved his command of his changeup.

Zach Eflin turned heads with his increased velocity and surging strikeout rate, but he decreased his hard contact rate from 33.3 to 29.1 percent. Cutting back on his sinker usage helped Eflin to reduce his rate of hard-hit balls and put up an overall 11.2 percent HR/FB, as well as a 9.9 percent HR/FB away from Citizens Bank Park.

Jalen Beeks‘ 12.5 percent HR/FB was below the major league average, even though the Rays’ rookie had the unenviable position of being in the AL East. It helped that he allowed hard contact on only 29.2 percent of hit balls. If he can improve his cutter (.211 ISO), he could take a step forward and have value, whether as a starter or as a “primary pitcher” following an opener.