David Phelps is Ready to Break Out if the Marlins Let Him
There are plenty of ways to characterize David Phelps‘ success in 2016. Among pitchers who threw at least 80 innings, he posted the 5th-best ERA (2.28), 6th-best xFIP (3.15) and 7th-best FIP (2.80). That’s a big deal, although it’s a decidedly smaller deal considering the bulk of Phelps’ innings came from the bullpen.
The reason for Phelps’ success — the cause to the effect, that is — is fairly obvious:
Phelps added more than 3 mph to his four-seamer and sinker as well as a tick or two to each of his off-speed pitches.* In an August edition of his NERD game scores, Carson Cistulli quipped, “As with most other pitchers, Phelps at 94-95 [mph] is markedly different than Phelps at 91.” Indeed, Cistulli. Phelps looked like a changed man.
It’s easy to attribute his sudden late-career success to his almost-full-time move to the bullpen. It’s how the narrative typically plays out: a pitcher’s velocity plays up better in short spurts. It’s why we expect failed starters can become elite relievers. It’s a cognitive bias, but it’s a bias we have because it tends to be true. This shorthanded logic, however, undersells Phelps’ gains both under the hood and on the mound.
It was the perfect test of the null hypothesis: deploy Phelps in the rotation for an extended period of time. In the month of August, Phelps appeared in games exclusively as a starter, leaving and returning to the bullpen on July 30 and Sept. 12, respectively. He didn’t once step out of a bullpen in August. And his velocity? It stuck.
Phelps sustained his velocity as a starter (although I acknowledge he wasn’t forced to bear the burden of a starter’s workload). And he demonstrated that he could be just as, if not more, successful in a starting role, notching identical rates of strikeouts per nine innings (11.84 K/9) and weighed on-base averages against (.256 wOBA) as well as a better strikeout-to-walk differential (22.5 K-BB% to 21.3%) and xFIP (2.97 to 3.22).
Granted, it’s a small sample, but Phelps was essentially a 5-WAR pitcher in his August rotation stint, and he was a 4-WAR pitcher when prorating his 87-ish innings to a full 200. Should he come even remotely close to replicating his 2016 season, he would easily be the Marlins’ best starter.
Unfortunately, the Marlins apparently have no interest in finding out if this golden ticket will actually let them into the chocolate factory. Michael Hill, president of baseball operations for Miami, told the South Florida Sun-Sentinel in December he’d keep Phelps in the bullpen if he had his way. Accordingly, FanGraphs has projected Phelps for only 19 innings as a starter, instead allocating starts to Justin Nicolino and his projected 4.9 K/9 (ugh).
Just know that all of Phelps’ pitches benefited from the added velocity, registering whiffs on swings about 50 percent more often across the board. His sinker and curve induce ground balls at elite rates and didn’t allow a single home run in 2016. His walk rate leaves much to be desired, but pitchers with big strikeout rates, such as Tyson Ross and Lance McCullers, make due with unsightly walk rates (although it’s worth noting those two induce better ground ball rates, too). He has his flaws, but he has four legitimate offerings that now all play up with velocity, regardless of his role. Obviously, he’s no guarantee to succeed. Walking four hitters per nine innings is something like walking a tightrope in this day and age. But what do the Marlins really have to lose?
Perhaps the Marlins will see the error of their ways. And the inevitability of poor performance (from any of Miami’s mediocre starters) and injury (ditto) could open the door for him to re-enter the rotation. Until then, it’s hard to rely on Phelps for much of anything, really. He could be one of those highly effectively middle relief arms, but his expected role effectively caps his upside.
Phelps is a name to remember in deep leagues and dynasty formats, even if he’s 30 years old. He’s being drafted outside the top 400 in National Fantasy Baseball Championship (NFBC) drafts, establishing a potential return on investment that’s very difficult for me — and, I encourage, also you — to ignore.
* * *
*Does anyone know anything about Phelps’ offseason regimen? Any indication of what might’ve manifested his velocity bump? I scoured the internet (aka I searched Google for, like, eight minutes) and couldn’t find anything. Any information is appreciated!
Two-time FSWA award winner, including 2018 Baseball Writer of the Year, and 8-time award finalist. Featured in the print edition of Baseball Prospectus since 2022 (2026 edition out now). Biased toward a nicely rolled baseball pant.


OK, I was watching a Marlins’ game last season when Phelps was pitching (yes, I live in Florida), and I believe the announcers said that throwing harder was simply a function of being a reliever, of not having to pace himself. So that after his success as a reliever his coaches told him to keep throwing hard, even when starting.
Reminds me of Hector Santiago. His velocity increased a couple notches at the beginning of this year and I heard the a similar story, that he just decided to throw harder every pitch instead of pacing himself. His velocity declined later in the summer though. Keeping it up is difficult, it turns out.
Also, Carrasco
Been thinking the same on Phelps. Definitely a guy worth taking a late round shot on.
The Marlins are right to keep him in the bullpen where he is useful. Look at the big picture instead of August as a whole. Here are his average 4 seam velocities in his August starts: Aug 5 (95.05), Aug 10 (94.36), Aug 15 (94.15), Aug 20 (94.18) and Aug 26 (93.70). And SD shelled him on Aug 26. As a RP he provides a lot of value to Miami. As a SP he provides very little long term value.
So as bpqa said earlier with Santiago be careful of sample sizes this small. I think Phelps is a wasted pick if he is a SP. If he is a RP he has value.
I’m reluctant to say SD shelled him; 6 K’s, 2 hits (0 HR) in 3.2 innings. That’s just bad sequencing. Three walks didn’t help, but I’ve seen much worse than a single-game 1.36 WHIP.
I’ll be the first to harp on small samples all day, but velocity is not the same as BABIP; Phelps is entirely in control of velocity. You don’t need a large sample size to tease out luck like you would BABIP. Phelps threw hard, and he threw effectively. I don’t really care if he can’t sustain it all year (although I wish he could); it’s mostly about the fact that even a couple of months of Phelps in the rotation would likely be a massive marginal upgrade for a team that pretends like it wants to contend.
I don’t know how we are defining value when you say he is more valuable as a RP, but, to attest: Phelps accumulated ~0.6 WAR in ~25 innings as a starter and ~1.3 WAR in ~62 innings as a reliever. Thus, he was more valuable on a per-inning basis in the rotation. That is why he is an experiment worth trying, given he has the potential to be significantly better than any pitcher in the Miami rotation, even if he does lose steam midseason.
(I acknowledge that this could rip open a discussion about reliever leverage and WPA’s effect on value — something I’d prefer to avoid — but let’s just consider that, in a neutral context, Phelps has the potential to deliver a lot more value as a starter given bulk innings.)
Looked back at Phelp’s career and he added a great deal of velocity last year. Don’t know where it came from but it helps me understand why he suddenly became extremely valuable.
Phelp’s lost 1.3 MPH from the 1st start to the 5th even though he pitched 20 innings in his first 4 starts. Start extending this to 7 innings per start and I believe he loses another 1.5 MPH after 5 starts. Now consider to enter the season with him at SP means you extend him in ST so he enters the regular season at 93 and may be down to 91 and back in the bullpen by the end of April. Maybe he can stay above 92 for a couple of months. I don’t know. From a fantasy perspective I say go for it. I can monitor his starts and drop him when his velocity drops. But from a club perspective he proved he was very valuable out of the pen last year and I would keep him there.
As per getting shelled I didn’t really look at the WHIP. I noticed he was yanked in the 4th inning after giving up 3 runs (7.36 ERA).
Yeah, I got you. I guess my argument would be if they gave him ample time to stretch out, he could maybe keep it up for longer. (Something I didn’t articulate in the post, so that’s my fault.) He was in the bullpen all year through July 31, then jumped into a starter’s role August 4. Seems a bit abrupt. So that would give me cause for optimism.
I also get the bullpen thing from a real-baseball perspective; they KNOW it’ll work in relief. I guess I’d argue that if they really think they can contend, then they should take calculated risks where they might pay off, because they’re not really that close to contending… Phelps would be one of those risks. But if they don’t plan to contend then I’d understand not forcing the issue.
Similar theme with Danny Duffy last year, but Duffy couldn’t maintain as the season finished. If Phelps actually started all year, doubt he’d be able to sustain that reliever level velocity as a starter all year as well. But clearly for the Marlins, it’s worth finding out!