There’s a lot going wrong with Orioles pitching right now, but Kremer is trending in the right direction.
Let’s not sugarcoat it: the Orioles are playing bad baseball. They can’t hit. They can’t hit with runners in scoring position. They can’t hold a lead. Most importantly, they can’t build a lead, because their starting pitching has been so bad.
The Orioles rotation currently ranks last in the AL with a 5.52 ERA. Not only are the results bad, but in terms of stuff—that is, the velocity, shape and spin of pitches—this has been deemed one of baseball’s worst rotations, ahead of only the Rockies, Angels, Cardinals, White Sox and Diamondbacks. When your pitchers have bad stuff, their only hope is to locate it perfectly, meaning there’s precious little room for error, and precious little reason for Orioles fans to expect real improvements in starting pitching this season. Right??
How does this stuff/location thing apply to one Orioles pitcher whose results are lately trending upward—Dean Kremer? The 29-year-old right hander had a poor start to the season, going 1-4 with a 7.04 ERA through six starts ending in April. But his last two outings have been something different: on May 2, Kremer shut out the KC Royals over seven innings, and on May 8, he allowed two runs on just three hits over seven innings to the Minnesota Twins.
If we take a look at Statcast data, the Stuff+ pity party thing doesn’t perfectly apply to Kremer the way it does to other Orioles pitchers whose metrics are currently a deep sea of blue. For instance, even with his struggles this season, Kremer has what Statcast acknowledges is a great curveball. It turns out this is typical for Kremer in the sense that it’s not typical. Here’s what I mean: over his career, Kremer has showed an interesting pattern where each season he’s had at least one plus pitch working for him, and each time, it’s been a different pitch or combination of pitches. In 2024, his cutter and new splitter were his main weapons (a +1 Run Value for each). In 2023, it was his fastball (+3 RV) and his sinker (+7 RV). In 2022, it was his cutter (+10 RV).
What’s the story behind such eclectic (erratic?) pitch mixing/effectiveness? It’s possible Kremer loses feel for his pitches. Or that he tinkers a lot. Or that hitters adjust to his arsenal, and coaches give him something new to work with.
It might be all three, in fact. For instance, since 2024 he’s been working with a higher arm slot, and it’s helped his curveball, which despite his reputation as a prospect, was actually his worst pitch in ’21-’23 and is now his best. In 2024, he dropped the changeup, too, and worked in both a cutter and a splitter. Hence adjustments to pitch mix and windup can be seen with Kremer.
Has the higher arm slot hurt his fastball? I’m not enough of a pitching maven to know, but it’s true that compared to other years, Kremer’s four-seam fastball has lost effectiveness. Whereas as recently as 2023 the pitch ranked in MLB’s 90th percentile in Run Value, now it’s down to 3rd. What gives? Is it velocity? This is not strongly suggested, as his average fastball velocity barely budged, from 93.8 mph in ’23 to 93.4 mph now. Nor does it seem related to spin.
There’s room to think an inconsistent delivery could be at issue. Over his stretch of bad games in April, Kremer’s vertical release point on his fastball kept getting lower and lower (I believe this means he was not extending as much). But over his last two, it’s picked back up noticeably.
Perhaps relatedly, Kremer may be fiddling with his grip, as his offspeed pitches are varying in spin to a noticeable degree. For instance, in his good start on May 2 against Kansas City, his splitter spun at a rate of 1178 RPM. On April 22, a day he got clobbered by Washington, it was down to 996 RPM.
Besides inconsistency and experimentation, there may be strategy at work. Kremer is using his fastball less and less in his mix, from a high of 75% in his first start on March 29 to just 28% against Kansas City last week and 24% against the Twins yesterday. Perhaps not coincidentally, his strike percentage went up noticeably over his last two good starts. On April 27, when he allowed five runs to Detroit, he landed 46.1% of his pitches for strikes. On May 2, 56%.
Lots of fiddling. You might ask, couldn’t Kremer have practiced some of this during spring training? The righty himself is aware. “I wish I would do it Opening Day, you know?” he said after his good start on May 2. “But it doesn’t seem to have unfolded that way. But hopefully, from here on out, we can put together a consistent start.”
That same day, manager Brandon Hyde said something that turns out to be true. “April’s over for Dean and it’s May,” Hyde said, “and historically, I don’t know the numbers actually, just going from my recall the last few years, he gets better as the season goes on.”
In fact, Kremer has a career 6.24 ERA in March/April and opponents hit .291 with an .861 OPS off of him. By May, the ERA drops to 3.72 with an opposing average of .235 and a .698 OPS.
Anyhow, it turns out there are several reasons Dean Kremer got off to a bad start this spring. One, he always does. This may be because the Orioles seem to keep giving Kremer new assignments to learn, year after year. Two, he’s working on commanding his offspeed pitches, as fluctuating spin rates show. Three, he’s working on his fastball delivery. (It’ll be worth following the effectiveness of the four-seamer, and whether Kremer continues to use it less and less with greater effectiveness.)
Better late than never, I guess? In two consecutive outings, Kremer tossed seven scoreless innings and has lowered his ERA by nearly two points. With Zach Eflin on the plane to Anaheim, it seems he’s close to rejoining the roster. It’s too soon to feel anything like “optimism” about this rotation, but with Eflin nearly back and Kremer on a roll, there is a chance to watch something more high quality soon.