Model Holmes: New York’s New King of Sinkers Is on a Tear

The Yankees had the third-best bullpen in baseball last year, but you’d be forgiven for thinking that had changed this year. Last season’s two best relievers, Jonathan Loáisiga and Chad Green, have combined for 0 WAR and an ERA above 5.00, and Green will miss the rest of the season due to injury. Their highest-paid reliever, Aroldis Chapman, has lost more velocity and is recording strikeouts at a below-average clip. Their big speculative offseason addition, Miguel Castro, is below replacement level.

Naturally, they have the second-best bullpen in baseball in 2022. Michael King, who I recently wrote about, is the headliner so far this year, but he’s hardly alone. Clarke Schmidt, who profiles as a starter long-term, has looked good. Wandy Peralta is a competent lefty specialist. And that brings us to King’s running mate, the other best reliever on the Yankees: Clay Holmes.

Holmes is hardly new to the majors. He toiled in obscurity with the Pirates for years, walking too many to take advantage of his grounder-inducing sinker. Then the Yankees got their hands on him, and he turned that sinker into an entire identity, filling the zone and letting the chips fall where they may.

If that’s where your knowledge of Holmes ends, you probably already think of him as an effective reliever, perhaps even one of the best in New York’s bullpen. In his 28 innings of work in the Bronx last year, he had a 1.61 ERA and 2.1 FIP, after all. The team used him in high-leverage spots, and he delivered, thanks not only to his sinker but also to a redesigned slider. That’s a satisfactory end to the story.

This season, though, Holmes has taken things to new extremes. The sinker that has long been his best pitch? He’s throwing it more than ever, 80% of the time so far this year. The huge horizontal movement that makes it so effective? He’s never gotten more of it. The grounders that make his game work? He’s never gotten them at a higher rate.

I could probably stop the article right here and pivot to video. Here, watch one of these delightful sinkers:

Here’s another:

If you’ve ever wanted visual proof that you don’t understand fluid dynamics, watch those pitches and try to tell me how they move so much. You can’t, at least not without invoking sorcery. That’s the magic of a well-thrown turbo sinker.

But just because I could stop the article there doesn’t mean I have to. There’s more to Holmes than just “the Yankees taught him how to get more movement on his sinker.” He’s been more effective this year than he was even in his spectacular 2021 run, and I think I know what he’s been doing to unlock this new level of performance – and maybe even why he’s changed year-to-year.

Here’s where Holmes threw his sinker after joining the Yankees last year:

I’m not a pitching coach, but I do remember what Mike Shannon said: sinkerballers who leave their fastball up in the zone often end up with a job at General Motors, because they’re not long for the major leagues. That doesn’t look to me like a good location for a sinking, grounder-inducing fastball.

Now, results-wise, it was an effective fastball. Did you see those GIFs?!? He could throw the thing right down main street and hitters would struggle with it. When you induce that kind of drop and run, there’s more margin of error. But this year, Holmes is locating his sinker where it should be, and reaping the benefits:

In his time in New York last year, Holmes was in the 80th percentile in terms of average sinker height across all of baseball. In other words, only 20% of sinker-throwing pitchers in baseball located theirs higher in the zone on average. Despite that, he finished in the 80th percentile for groundball rate on that sinker. Only 20% of sinker-throwing pitchers coaxed more grounders out of hitters when those hitters put a sinker into play.

That’s a strange double to think about, because as you’d expect, there’s a strong correlation between sinker location and batted ball type. Over the last two years, when batters put sinkers in the upper third of the strike zone into play, they hit grounders only 40% of the time. When they instead make contact with sinkers in the lower third, their groundball rate explodes to 64%. This isn’t hard to explain: the lower you can keep your fastball, the more likely you are to induce a grounder.

Holmes’s aggregate results broke the mold in 2021. He was getting grounders like someone who clipped the bottom of the zone with every pitch despite throwing his fair share of elevated pitches. But under the hood, he followed the same pattern. Batters put 18 upper-third sinkers into play against him and hit grounders on half of them. They put 20 middle-third sinkers into play and hit grounders on 75% of them. They put eight lower-third sinkers into play – and hit grounders on all but one. Holmes, like all pitchers, got more grounders the lower he went. He just gets so many grounders overall that he posted an admirable groundball rate despite suboptimal location.

As we saw above, he’s back to locating his sinker at the bottom of the zone this year. He’s plummeted from the 80th percentile for average sinker height to the 15th. That’s had a predictable effect on his groundball rate on sinkers; it’s gone from the 80th percentile (pretty good!) to the best in baseball, period. No one else is even close:

Not only does he lead baseball in getting grounders with his sinker, he throws sinkers more than anyone in baseball other than Tim Mayza. The result? His 83.3% overall groundball rate leads the league by a comical amount. Among pitchers with at least 10 innings pitched in 2022, the gap between Holmes and second-place Alex Cobb is the same as the gap between Cobb and 49th-place Rafael Montero.

Getting that many groundballs would make anyone an effective pitcher, but it plays better in front of an airtight defense. Last year, that didn’t describe the Yankees; they were below average per OAA, with a -11 mark that ranked 21st in baseball. They weren’t disastrous – the Red Sox checked in at 35 outs below average – but routine plays became adventures more often than any sinker-heavy pitcher would find optimal.

After an offseason where the team overhauled its infield defense, they’re 10th in baseball, two outs above average as a team. They’re better set up to turn grounders into outs, and Holmes is feeding them grounders at a rate that would make peak Zack Britton jealous. Britton’s 80% groundball rate in 2016 is the highest since 1990 – and Holmes might make a run at that mark this year.

That’s not the only thing Holmes has improved upon so far this year. Early in his career, his walk rates held him back. That changed after he arrived in New York – he walked only 3.9% of the batters he faced as a Yankee. This year, he’s improved yet again. He’s only walking 2.6% of opposing batters, an elite number. It’s a startling transition for a pitcher who walked 14.9% of his opponents as a Pirate.

What’s he doing? If you guessed “throwing every pitch in the strike zone,” you’re surprisingly wrong. His 45.1% zone rate is above average but significantly lower than the 50.3% mark he put up last year as a Yankee. But on the first pitch of an at-bat, he’s throwing in the strike zone 59% of the time, comfortably the highest rate of his career, even higher than his partial season with New York last year.

That seems like a plenty good plan to me. As we’ve already covered, batters haven’t done much of anything when they put the ball in play. If you’re not afraid of damage on contact, why not just pump it down main street and see what happens?

Only, that’s not what Holmes is doing. If you break the plate down into two regions, the dead-red heart and the fringes, you’ll see that my earlier dismissal of sorcery might have been premature:

Uh, yeah, that’ll play. Holmes has stopped throwing meatballs. Instead, he’s dotting corners. But he’s improved by so much at throwing in the periphery of the zone that he’s still throwing more strikes overall. If you wanted to sit on a fastball down the pipe against Pittsburgh-era Holmes, you could do it. He’d also frequently miss, putting you ahead 1-0 and giving you better pitches to hit in the future. Now, you can’t sit on that middle-middle pitch as much, but he’s also missing less. It’s an impossible puzzle to solve for hitters.

Will he keep up his 0.42 ERA and 1.57 FIP? Most certainly not. As good as Holmes looks right now, he won’t avoid allowing a home run all year, he won’t keep up his .241 BABIP allowed, and he probably won’t continue to strand 93.8% of the runners that reach base against him. But don’t let those shiny and unsustainable headline numbers distract you from the central fact about Holmes: he’s turned himself into one of the very best relievers in baseball, one of the new leads in a Yankees bullpen that looks much-changed from its past iterations, and yet is as dominant as ever.