Jordan Lyles' 2019 season is a perfect reflection of the Pirates' starting rotation overall: Historically dominant out of the gate, and historically hittable ever since.
He salvaged his career as a reliever last year by revamping his pitch mix -- four-seam fastballs high in the zone, hammer curves at the knees -- and appeared to translate those results to an expanded starting role with Pittsburgh this spring. Signed to a one-year, $2.05 million deal over the winter, he was touted as one of the biggest free agent bargains in the game and another example of the Pirates finding value from a veteran pitching reclamation project.
And then, he hit the wall. Hard.
Already nearing his innings pitched total from last season (75 frames in 2019, 87.2 in '18), Lyles has been crushed since late May while also dealing with a lingering left hamstring injury. Over his first eight starts, spanning 47.2 innings pitched, Lyles had a 1.89 ERA with 9.3 strikeouts, 3.2 walks and 0.9 home runs per nine innings pitched. His fielding independent ERA (FIP) -- a more accurate measure of pitching performance that's based on strikeouts, walks and homers allowed -- was 3.34. Since then, Lyles has lasted just 29.1 innings in seven starts. He's still missing bats (10.1 K/9), but he has lost the strike zone (4.6 BB/9) while serving up a staggering 2.5 home runs per nine frames. Lyles' ERA over that seven-start stretch is 10.13 (second-worst in the majors among pitchers with 20+ innings pitched since mid-May) and his FIP is 6.14. Even in a rotation stretched too thin by injury, Lyles must have a short leash at this point.
Let's take a closer look at the differences between a dominant Lyles, and the guy getting drubbed over the past couple of months.
In terms of pitch type, nothing drastic has changed for Lyles. During his superb start to the 2019 season, he threw fastballs about 50 percent of the time and curveballs about 31 percent. He also mixed in the occasional slider and changeup (thrown about 9 and 10 percent, respectively). Since late May, Lyles has thrown his fastball a little more often (56 percent) and a little harder (93.1 mph, compared to 92.2 mph). He's still dropping plenty of curves (31 percent), and showing hitters slightly fewer sliders (7 percent) and changeups (7 percent).
What has changed is the effectiveness of those offerings. According to Fangraphs, Lyles' fastball was worth +1.4 runs more per 100 pitches thrown compared to an average MLB pitcher during his first eight starts. Since then, it has been -1.4 runs below average. His curveball was +1.5 runs above average per 100 pitches early on, and -2.8 runs below average per 100 since then. His slider was effective in short doses in April and early May (+4 per 100 pitches), and wretched ever since (-5 runs below average). Ditto for his changeup (+0.5 early, -5.3 since late May).
So, none of Lyles' pitches is working lately. But let's dig a little deeper. What has changed about those pitches -- particularly on his two most important offerings? Through his first eight starts, Lyles' fastball averaged about 4.6 inches of horizontal movement in on right-handed hitters compared to a pitch thrown without spin. His heater "rose" (dropped less) 9.5 inches compared to a no-spin pitch. During his difficult stretch, Lyles is averaging 3.9 inches of tail in on righties and 9.3 inches of vertical movement. The spin rate on his fastball has dipped from 2,185 revolutions per minute to 2,151 rpm, according the MLB Statcast. Essentially, Lyles is throwing his fastball harder -- but that extra zip has come at the expense of movement. And his harder, flatter fastball is getting hammered. Lyles had a .316 opponent slugging percentage against his fastball through early May, and a .613 slugging percentage since then.
With his curveball, the good version of Lyles averaged about 9 inches of drop in the zone compared to a pitch thrown without spin. Over the past two months, he has averaged 8.3 inches of drop. His spin rate on the curve has declined from 2,517 rpm to 2,495 rpm. It's a similar story to his fastball: The pitch isn't moving as much. Lyles limited batters to a .241 slugging percentage on curves while dominating, and .651 since then.
With his two bread-and-butter pitches, Lyles has lost movement and spin as the innings have piled up. That decline in stuff and performance, coupled with some injury concerns, naturally raise the question of whether Lyles should return to a bullpen role. Maybe, in shorter stints, his pitches would regain their movement and he could get back to being an asset for a team that still has designs on contending for the NL Central.
That could be a winning approach, despite how thin the Pirates are in the rotation. Maybe Lyles would be a good "opener" or multi-inning reliever.
But here's one complicating factor: Lyles hasn't exactly shut down batters from the start this year. Check out his on-base-plus slugging percentage (OPS) against by time through the lineup. He has struggled mightily when facing batters for the first time in a game, settled in while navigating the lineup for a second time, and faded as his pitch count climbed:
Considering that Lyles hasn't thrown more than 100 innings in a season since 2014, and that his career high is just 141 frames, the Pirates' best bet is to shift Lyles back into a relief role or have him piggyback with someone like Steven Brault. But for Lyles to succeed in any role moving forward, he'll need to stop being a punching bag right after taking the mound.
MORE PIRATES
• Frame job: When it comes to pitch framing -- the subtle art of stealing strikes on pitches located on the fringes of the strike zone -- the Pirates have two players at the opposite end of the spectrum. Jacob Stallings, at least so far, is awesome at expanding the zone for his pitchers. Elias Diaz? Not so much. According to StatCorner, Stallings is getting about +1.7 more called strikes per game compared to the average catcher. That's the fourth-best rate among all backstops who have received at least 1,000 pitches, trailing only Max Stassi (+3.0 per game), Austin Hedges (+2.1) and Tucker Barnhart (+1.9). That stat considers extra called strikes on pitches thrown outside of the zone (a good thing), as well as called balls on in-zone pitches (bad). Despite limited playing time, Stallings' pitch-framing has saved +3.1 runs above average. Diaz, by contrast, gets -1.8 fewer called strikes per game compared to the average catcher. That's second-worst in the majors, ahead of just Isiah Kiner-Falefa (-1.9). Diaz's pitch-framing has cost the Pirates -12.2 runs, per StatCorner, which translates to about one win in the standings. With Francisco Cervelli possibly taking off the chest plate for good and the Pirates lacking quality catching prospects, Stallings has at least solidified his roster spot. As for Diaz, he'll need to rediscover his 2018 power stroke or make serious defensive improvements to be a good long-term starter.
• Up a Crick: Kyle Crick's 2019 season is a high-wire act that would make The Flying Wallendas squeamish. While Crick's ERA is 3.41, his FIP is 5.33. Crick is still missing bats, with a career-high strikeout rate (27.1 percent of plate appearances). But his walk rate has doubled from 2018 (9 percent) to 2019 (18.8 percent). Crick is throwing fewer pitches within the strike zone (50.2 percent in '18 and 45.6 percent in '19, per Fangraphs) and getting hitters to expand their zones less often (29.8 percent chase rate in '18, 28.9 percent in '19). His fastball is at the heart of his control woes. Crick threw his four-seam fastball in the zone 54 percent of the time in '18, and just 49.8 percent this year. His chase rate on the four-seamer has dipped from 29.7 percent to 18.8 percent. Crick has one of the game's nastiest sliders, but he won't survive with just one pitch for long. He has to re-discover his fastball.
