Tyler Phillips Is at It Again

The other night, I was lying around, looking at my phone, trying to fry as many neurons as possible without using hard drugs or listening to Angine de Poitrine, and I saw something that bugged me a little. It was a highlight reel from a series of interviews with Padres closer Mason Miller and Kait Maniscalco, which started off as follows:

Maniscalco: Do you think closers have to have a couple screws loose to want to pitch in the highest-pressure situation in the game?

Miller: Quietly, yes. Outwardly, I think you can keep it together and be a fairly normal dude… I wouldn’t say anybody would say I have a screw loose quite yet.

There are two ways to read this question. First: Does it take an unusual personality type to thrive in a high-pressure environment like closing out a big league baseball game? Probably, to some extent. The ability not only to thrive under pressure but also to shake off failure when it comes is a special thing, one baseball people have tried and struggled to identify since the closer role was invented.

I would not, however, go so far as to say that closers need to have a specific personality type. No one would accuse the top three pitchers on the career saves list — Mariano Rivera, Trevor Hoffman, and Kenley Jansen — of having a single loose fastener among them. At the same time, you don’t have to get too far down that list (Jonathan Papelbon, Rollie Fingers, Fernando Rodney) before you hit some serious oddballs.

Which is where we get to the second interpretation of Maniscalco’s question, which trades on one of baseball’s greatest stereotypes: Relief pitchers are a little nuts.

I grew up in the 1990s, which was something of a golden era for weird relief pitchers. Turk Wendell was the greatest of these, but feel free to remember your favorites in the comments. Even children’s media was replete with such characters; I was especially fond of the Little Big League duo of Blackout Gatling and Jim Bowers. (Jonathan Silverman’s portrayal of Bowers is also memorable for being one of the very few instances of an actor with no high-level baseball experience producing even remotely believable pitching mechanics.)

I think Miller is correct when he says a reliever can “quietly” have a couple screws loose. It’s important to remember that a person’s mental health is frequently invisible to others; we are each fighting our own private battles, and so on. But it kills me to see the best reliever in the world take a question about how crazy relievers are, while looking and sounding like a junior associate at an accounting firm talking to his boss at a company golf outing.

She should’ve asked Tyler Phillips.

The Marlins right-hander came to prominence last year for what can actually be described as bullpen antics.

Everything about Phillips’ career is weird. He was a 16th-round pick in 2015 out of Bishop Eustace Prep in Pennsauken, New Jersey, a school that’s best known for giving us Billy Rowell, one of the most famous draft busts of the 21st century. (Part of the Mike Trout legend is that Rowell flamed out so badly it convinced some scouts that the quality of play in South Jersey was such that Trout couldn’t really be that good — he was just beating up on weak competition.)

In its 76-year history, Bishop Eustace has produced six major league players, four of whom — Phillips, Zac Gallen, Justin Hagenman, and Devin Smeltzer — were all in school at the same time. Phillips floated around the Rangers and Phillies systems without much fanfare for nine years, until July 2024, when the Phillies had an unexpected hole in their rotation. Phillips threw six scoreless innings in his third career start, and pitched a complete-game shutout against the Guardians in his fourth. (More weird trivia: Only 16 pitchers threw complete-game shutouts in the majors in 2024. Four of them were Phillies, but Cy Young runner-up Zack Wheeler was not among those four.)

It was a major local headline — Phillips had grown up across the river and had been a Phillies fan as a kid — but any hopes of a long-term local-boy-made-good situation were dashed when Phillips allowed eight runs in 1 2/3 innings in his next start. He was out of the rotation shortly thereafter, and on his way to Miami as a roster crunch casualty on the eve of last season.

That’s where Phillips turned into quite a good high-volume reliever in 2025: He had a 2.78 ERA in 77 2/3 innings over 54 appearances, which is how he found himself as the subject of a postgame interview in which he gave one-sentence answers about how much he hates hitters.

Phillips throws hard — his sinker, his most-used pitch, averaged 95.4 mph last year — and the five-pitch repertoire he carried over from starting gave him the ability to show hitters different looks depending on which side of the plate they lined up on. He was almost all sinker-sweeper against righties, from a low arm slot, and sinker-four-seamer-splitter-curveball against lefties.

Everything had kind of an off-kilter movement profile; Phillips’ sinker sank, but so did his four-seamer, and both his sweeper and curveball were closer to vertical than average. The result: a 97th-percentile chase rate and gobs of groundballs and weak contact. But even with all that velocity and deception, Phillips couldn’t strike many batters out. His strikeout rate, 16.6%, was 177th out of 199 pitchers with at least 70 innings pitched. The two guys behind him on that leaderboard were Cal Quantrill and Kyle Hendricks.

This year, Phillips has raised his arm angle a couple degrees, giving all of his pitches a more normal movement profile. That’s spread the movement on his five pitches out over a wider area. He’s also using his splitter more (23% usage rate in 2026, up from 14% last year).

Phillips’ groundball rate is (admittedly in a small sample) closer to average, and he’s giving up more hard contact when batters get wood to horsehide. But that’s happening less than ever. Through six appearances, he has a 96th-percentile whiff rate and an 80th-percentile strikeout rate.

I am a child of the sabermetrics movement of the late 2000s and early 2010s, and am therefore congenitally predisposed to distrust pitchers with strikeout rates under 20%. A million things can happen when a batter puts the ball in play, and most of them are bad.

But I’m not convinced this new version of Phillips — if indeed it is a new version, and not a small-sample blip brought on by one too many scoops of pre-workout — is better than last year’s. Not only was he so good at producing weak contact and grounders, I liked that the guy who slapped himself in the bullpen and talked like a Hanson brother had an aesthetically weird game.

Fortunately for us, the Marlins are still using Phillips in a weird way. Last year, he started as a low-leverage and mopup guy and gradually worked his way into more significant appearances. From July 1 through the end of 2025, Phillips allowed just five earned runs in 37 1/3 innings, and got more than six outs just three times in 25 outings.

This year, he’s yet to make his first single-inning appearance. His one save was a three-inning stint to close out an 8-1 win over the Reds on April 9. Through the weekend, Phillips was tied for seventh among relievers with five or more appearances in batters faced per game, at a hair under nine. He’s averaging 35.5 pitches per appearance, which is sixth in the league.

And yet I wouldn’t quite call him a bulk reliever. A lot of the leaders in workload per relief appearance (Sean Manaea, Antonio Senzatela, José Urquidy) are guys who would prefer to be starting, but their performance doesn’t warrant a rotation spot. You could say the same about former Rockies first-round pick Chase Dollander, the latest hard-throwing Colorado prospect to struggle in his first contact with the majors.

But Dollander leads all relievers with an average of 16 batters faced and 64.4 pitches per outing. Phillips is still performing a recognizable relief role, even if it is one from the 1940s, but Dollander is doing something else. To be clear, he has been terrific this season after last year’s woes, with a 55.1% groundball rate and a 28.7% strikeout rate, along with a .205 opponent batting average and a 3.32 ERA. (The one blemish for him is three home runs in 19 innings, which is an occupational hazard of playing where he does.)

Dollander has entered the game in the third inning or earlier in four of his five appearances, and thrown at least four innings in four of five appearances. Under normal circumstances, we’d call that a bulk reliever behind an opener, but that’s not really how he’s being used. Dollander has followed a right-handed reliever in every game he’s pitched this year; three times he’s followed Jimmy Herget, a fellow low-arm-slot guy, albeit a softer-throwing one.

I’ve been banging this drum for as long as I’ve had a platform to place the drum on: The one-inning fastball-slider guy is not the only way to build a relief pitcher. Pitchers like Phillips and Dollander — legit power arms — can be deployed on a multi-inning basis. In Dollander’s case, that’s gotten the Rockies near-starter-level performance with shorter bounce-back time and some ability to pick matchups. In Phillips’ case, it’s given the Marlins two or three relievers’ worth of length at the cost of one set of warmup slaps.

Relievers are supposed to be weird. Teams should use them in weird ways.