It would be a major understatement to call 2019 a disappointing season for New York Yankees left-handed starter J.A. Happ. After the Yankees acquired Happ from the Toronto Blue Jays at the 2018 non-waiver trade deadline, he went 7-0 in eleven starts with a 2.69 ERA, .676 OPS against, and sixty-three strikeouts to sixteen walks over 63.2 innings pitched. After that solid stretch run, Happ gave up five runs in two innings in an ALDS Game 1 loss against the eventual World Series Champion Boston Red Sox.
Since signing a two-year contract worth $34 million with a $17 million vesting player option for 2021, which triggers if Happ throws 165 innings or makes 27 starts in 2020, the thirty-six year old has been more ALDS disaster than solid stretch run for the Yankees. Below are his 2018 and 2019 numbers.
Except for his walk rate, Happ’s seen his statistics deteriorate drastically across the board. Given the league-wide home run surge, his home run rate has not spiked as much as it appears, but it is still above the 2019 major league average. He has the second-worst FIP of all sixty-eight qualified starting pitchers after Seattle Mariners lefty Yusei Kikuchi. According to Statcast, Happ has seen his velocity diminish on most of his offerings as well. Happ throws an occasional curveball, but due to its infrequent use, it is not included in the table below.
|Pitch||2018 Avg. Velocity||2019 Avg. Velocity|
|Four-Seam||92.3 mph||91.9 mph|
|Sinker||90.4 mph||89.5 mph|
|Slider||85.3 mph||84.8 mph|
|Changeup||86.0 mph||86.0 mph|
The low point of Happ’s season was likely on August 9th at Toronto. He surrendered six runs over five innings on four hits, three walks, and three home runs. The 6’5 southpaw allowed a .467 xwOBA against for the game, the second-worst mark of his season other than his first start of the season against the Orioles on March 31st. The start against the Blue Jays raised Happ’s ERA to 5.48, and it would rise even higher before it would start to fall.
Over his next five starts, which include a start where he allowed five runs in four innings to the Oakland Athletics on August 21st, Happ is 3-1 with a 3.42 ERA, a .557 OPS against, three home runs given up, and twenty-nine strikeouts in 26.1 innings pitched. Here are his numbers before and after August 14th.
|3/31 – 8/9||5.48||5.67||18.6%||6.3%||.350||.335|
|8/14 – 9/7||3.42||4.09||27.9%||12.5%||.255||.261|
Thirteen walks in 26.1 innings are responsible for Happ’s slightly inflated FIP, but he’s otherwise improved in every aspect. Happ is missing a lot more bats and inducing much weaker contact. The average exit velocity on the batted balls against Happ has decreased from 89.2 miles per hour prior to August 14 to 86.1 miles per hour from August 14th on.
The Athletic’s Lindsey Adler noted in an article over the weekend that Happ feels he’s in a better place mentally now than earlier in the season. Happ also seemed to credit throwing first pitch strikes more often for his recent success, which is puzzling given that his first pitch strike rate through August 9th was 57.9% and 54.8% from August 14th on. Even if you alter the cutoff to be August 21st, when Happ had his poor start against the Oakland Athletics, his first pitch strike rate up to and including that game was 57.3% and has been 58.1% after that start.
Happ might be enjoying a better mindset on the mound, but getting strike one more often doesn’t look like what’s driving his turnaround. The keys to Happ’s recent success appear to be some changes in his pitch mix and the location of his sinkers to lefties and his four-seamers against righties.
|Date Range||Batter Handedness||Exit Velo.||Launch Angle||xwOBA|
|3/31 – 8/9||Left||87.1 mph||5.9 degrees||.295|
|8/14 – 9/7||Left||80.4 mph||4.7 degrees||.119|
|3/31 – 8/9||Right||89.9 mph||15.6 degrees||.348|
|8/14 – 9/7||Right||87.8 mph||17.2 degrees||.308|
Happ has had better success against both righties and lefties in his recent five-game stint, and he’s really been stifling lefties lately.
Below are two tables from Brooks Baseball showing Happ’s pitch mixes from March 31st through August 9th and from August 14th through September 7th.
Against lefties, Happ has dropped the use of slider by nine percent overall and thirteen percent when he’s ahead in the count. He’s using his sinker and four-seamer more often in all counts, and his sinker has become his primary offering when he’s ahead of hitters, particularly with two strikes. With batters on the ropes in two strike counts, Happ’s increased his sinker usage against lefties by eighteen percent and dropped his slider usage by twenty-three percent.
Against righties, Happ has ditched his sinker for his four-seam fastball: his sinker usage against righties has decreased by thirteen percent to just five percent, and his four-seam usage is up from fifty-one percent to sixty-one percent. Happ’s upped his slider usage against righties by four percent overall and a whopping twenty-one percent on the first pitch. Maybe being more comfortable starting righties out with his slider is what he was referring to in his comments about the first pitch.
Let’s see how lefties have fared against Happ’s sinker.
|Date Range||Exit Velo.||Launch Angle||xwOBA|
|3/31 – 8/9||84.8 mph||2.3 degrees||.263|
|8/14 – 9/7||75.1 mph||-21.6 degrees||.102|
The results are much better. Lefties are making worse contact on average and absolutely pounding his sinkers into the ground. Location might have something to do with the improvement.
In the heat maps above, you can see that Happ was leaving a lot of sinkers out over the middle of the plate to lefties. Lately, he’s done a much better job of getting the sinker inside on lefties, especially low and in just off the inside corner. That location is likely to get hitters to swing over the top of the sinker for more whiffs and get in on lefties’ hands for weak contact. Here’s J.P. Crawford with a swing and a miss on August 26th.
And Happ sawing off Matt Olson for an easy double-play on September 1st.
Finally, Happ gets Matt Olson to ground out weakly into the shift in the same game.
Let’s turn to righties against Happ’s four-seam fastball.
|Date Range||Exit Velo.||Launch Angle||xwOBA|
|3/31 – 8/9||92.0 mph||27.8 degrees||.369|
|8/14 – 9/7||88.3 mph||31.8 degrees||.220|
Happ’s numbers have also vastly improved with the four-seamer against righties. It’s encouraging to see a higher launch angle here, as that means Happ has likely been seeing more lazy fly balls and less hard-hit line drives and fly balls.
Happ has made a massive shift in his four-seam fastball locations to righties. It looks like he’s targeting the upper middle part of the zone less often and looks to have taken aim at the outside part of the plate. Keeping the four-seamer away from righties rather than getting them to swing underneath the fastball might make some sense given the movement profile on Happ’s four-seamer this season.
|Season||Velo.||V-Mov||VM vs. Avg.||H-Mov||HM vs. Avg.|
|2018||92.3 mph||14.7 in.||1.6 in. (10%)||7.1 in.||0.0 in. (0%)|
|2019||91.9 mph||15.4 in.||1.1 in. (7%)||8.1 in.||1.3 in. (20%)|
Happ’s four-seamer is dropping more in 2019 and doesn’t have as much rise as it did in 2018. That alone might make the middle-up location less effective for Happ in 2019 as hitters are less likely to swing underneath the old number one. In 2019, when the ball is leaving the park at record rates, throwing pitches in a spot that frequently leads to fly balls is probably not the greatest strategy if you aren’t missing a ton of bats.
While the rise on Happ’s fastball has decreased, the arm-side run on the pitch has drastically increased. He’s getting a full inch more movement away from right-handed hitters on the four-seamer, good for twenty percent more run than the average sinker in 2019. That increase in tailing action may make the outside part of the plate a smart location against righties, who are expecting the pitch to continue towards the heart of the plate but wind up seeing it dart back outside instead. That movement could be helping Happ get more whiffs and worse contact. Watch as Happ rings up Chad Pinder on August 21st.
Matt Chapman fails to pull this four-seamer on September 1st.
Xander Bogaerts flails at a four-seam fastball on the outside corner on September 7th.
Even before the newfound arm-side run on his four-seamer, you can see that Happ targeted the outside part of the plate with the pitch frequently in 2018.
It’s possible that the location of his four-seamers against righties had become too predictable in 2019. Whether it’s the movement, location, lack of predictability, or a combination of all three, Happ seems to have figured something out.
One other encouraging late-season trend has emerged for Happ: the lefty has also regained some velocity as the season has progressed. Below are charts from Baseball Savant detailing the average velocities on his four-seam fastball, sinker, and slider in each game this season.
J.A. Happ might be coming around at the right time for the New York Yankees. As rough of a season as it’s been for Happ, his recent success seems to be supported by some underlying changes that might make those improvements sustainable. If Happ’s success continues on today against the Detroit Tigers and throughout the rest of September, it might make sense to slot him in as the Yankees’ fourth starter in the playoffs and hope for a solid four or five innings before the bullpen takes over. Germán (5.68 ERA and 5.90 FIP since May 26th) might be hitting a wall and likely profiles better as a bullpen arm than Happ. If not, Happ’s recent numbers against lefties look really good from a lefty specialist out of the pen. At the very least, Happ’s recent run has given the Yankees front office a few reasons to look at him as a useful option on the playoff roster.