Corbin Carroll Has Lefties in a Blender

Corbin Carroll is the best left-on-left hitter in baseball right now.
It’s been a tremendous year for Carroll, however you want to split up the data. He has a 152 wRC+. He has 2.6 WAR. It’s looking like his best-ever season — a perfect follow-up to his best-ever season last year.
That Carroll appears to be taking another step forward in 2026 isn’t quite newsworthy. But the way he’s doing it certainly is:

Carroll is all of a sudden crushing lefties. He was just average against them the first few years of his career. Last year, he was fairly good in left-on-left matchups, though that improvement seems to stem more from his overall growth at the plate rather than a specific step forward against lefties.
This year? Carroll is a top-five hitter in baseball against lefties and the best lefty batter in the mix. If his 219 wRC+ against southpaws were to hold, it would be the third-best left-on-left season since we began tracking in 2002:
| Batter | Year | PA | wRC+ |
|---|---|---|---|
| Barry Bonds | 2002 | 178 | 278 |
| Barry Bonds | 2003 | 161 | 224 |
| Corbin Carroll | 2026 | 78 | 219 |
| Drake Baldwin | 2026 | 84 | 203 |
| Yordan Alvarez | 2026 | 75 | 198 |
Yup, Barry Bonds is the only one who’s done this over a full season. The names directly behind Carroll are also from 2026. That alone tells us he is probably a bit over his skis (as does the .500 BABIP). He probably won’t maintain the magnitude of his left-on-left excellence, but his success against southpaws does appear to be more than just a small-sample fluke.
The first place my mind goes is to his stance changes. As Chad Young noted for RotoGraphs last month, Carroll has moved off the plate by about three inches in 2026, the farthest change in that direction among all batters. Only three players stand farther from the plate.
This is actually the second year in a row that he has altered his stance. Last April, Esteban Rivera detailed the changes Carroll made ahead of 2025. That overhaul was more about his mechanics, with a new setup, load, and bat path, than where he was standing. This came with all sorts of mostly positive consequences and fundamentally reshaped his “hot zones.”
In 2024, Carroll was, frankly, terrible on pitches over the inner third of the plate. Esteban expected that last year’s mechanical overhaul would lead to better inner-third outcomes, and he was spot on. Carroll raised his xwOBACON on inner-third pitches by 125 points. This year, he’s improved again, adding another 146 points:
| Year | Inner | Middle | Outer |
|---|---|---|---|
| 2024 | .238 | .469 | .372 |
| 2025 | .363 | .573 | .487 |
| 2026 | .509 | .497 | .354 |
This makes sense. By stepping back, Carroll has taken his bat path with him and changed the pitches he can cover. The actual zone hasn’t changed, of course, just his relation to it. He’s now more likely to get the barrel to an inside pitch than he was before. The tradeoff is he’s less likely to cover the pitch away. We can see that where he’s done damage this year has shifted:

What’s interesting is that plot for this season shows more red — more success — on up-and-away pitches. He’s simply stopped swinging at them, and pitchers haven’t proved capable of throwing strikes to him in that region very often. Where he used to make decent (albeit infrequent) contact, he’s now mostly taking pitches and advancing counts in his favor.
This underlines a broader approach change. Carroll is swinging about four points less this year overall. He’s not necessarily swinging at “better” pitches, in the way we might define a zone independent of the batter’s approach within it. But he’s isolated a section of the zone to dominate while giving up on another, banking that pitchers can’t consistently hit the part of the zone he’s abandoned. To this point, it’s worked.

Carroll isn’t hitting righties nearly as well this year, though it’s not clear if that has anything to do with the shift. He has a 120 wRC+ against them, down from last season’s mark of 148. Again, he’s simply laying off pitches they throw up and away, and they haven’t been able to land that pitch consistently this year. His walk rate against righties is up to a career-best 13.6%, and many of those new walks are coming on out-of-zone pitches up and away. That said, he’s also striking out against righties at the same career-high clip that he did last year (24.7%). The pitches he’s whiffing more on this year are low and away, and up and in, but he’s also just whiffing more against righties this year, broadly. His whiff rate against right-handed pitchers is 29.3%, up from last year’s mark of 24.9%. When he does make contact, though, he’s finding the barrel — his ISO still starts with a two — so it’s possible this tradeoff will ultimately work in his favor against righties.
Against lefties? It’s absolutely working. They can’t make that pitch up and away, or at least haven’t been able to this point. Lefties have landed just four total pitches in the up-and-away part of the strike zone this year. The only place where they can get the ball in the zone consistently is over the middle or inner thirds, playing right into Carroll’s new swing path.
It makes sense. To throw to the outside part of the plate, lefties have to cut the pitch across the zone. If they miss — or even if Carroll simply catches their pitch out in front before it has a chance to reach the target — it often finds the middle part of the plate. When it does, he is punishing them: His hard-hit rate on middle and inside pitches is 66.7%, and his barrel rate is 18.2%.
That creates essentially two options for lefty pitchers: They can pound a fastball in, right where Carroll wants them to throw it, or they can sweep a bender away, risking a hanger. Carroll, to his credit, seems to know this and has retrofitted his approach. He’s dropped his first-pitch swing rate against lefties to a mere 30.8%. He’s seeing more 1-0 counts than almost anyone else in left-on-left matchups. And once he’s in that 1-0 count, his swing rate leaps to 43.8%.
What does he swing at when he gets ahead? Mostly breaking balls. Lefties almost always throw Carroll a first-pitch fastball to try to get ahead. If they miss, they very often come right back with the bender of their choice, trying to sneak back into the count. That generally hasn’t worked for them:
Carroll is swinging at 63.2% of breaking balls he sees from lefties when ahead in the count. That’s a 13-point increase over last season, and by far his favorite pitch when ahead. He’s whiffing on these benders just 8% of the time in 2026 — down from 35.3% last year. I wouldn’t call the contact he’s making superior, as most of it’s coming on the ground, but hard grounders to the pull side can sometimes result in extra bases for a player with Carroll’s speed.
It’s worth looking at a rare left-on-left plate appearance that went poorly for Carroll this season. There was a great matchup over the weekend, when the Mariners brought in lefty bullpen ace Gabe Speier to face Carroll with the game on the line. Speier started him with a fastball up and in. Carroll anticipated this and just missed it. Speier threw another perfect fastball on the top rail that Carroll could only watch, falling behind 0-2. Speier threw a third straight fastball way up. Carroll couldn’t lay off of it, swinging through it for strike three.
Look, I can put a positive spin on anything, but I almost think that at-bat highlights what’s made Carroll so great against lefties. He’s narrowed what they throw him into essentially two pitches: The fastball up, or the breaking ball down. His decision to swing is largely based on the count and whether the pitcher has to come into the zone. To get him out, a lefty must be almost perfect like Speier was there, blowing that first-pitch heater by him to take control. And even then, Carroll was clearly ready for it — he knew what was coming and took an “A” swing. From there, he was hoping Speier would miss high to even the count. Unfortunately, Speier didn’t, and not even Carroll has good splits 0-2.
Regardless of that at-bat, Carroll has lefties in a pinch. They aren’t equipped to attack his weaknesses, limiting them to a predictable set of choices. And few can beat him at his strength, especially when he knows what’s coming. How does a lefty beat Carroll right now? Well, it seems they generally don’t.
Ryan Blake is a contributor for FanGraphs and Lookout Landing.
I know this isn’t the point of the article, but a TWO SEVENTY-EIGHT wRC+? Dear Lord.
This also isn’t the point of the article, but I’ve been wondering if I missed an announcement about Estaban Rivera moving on to something else. I always made a point of reading his work, it was uniformly terrific. Nice to see it referenced here.
That’s absolutely insane. I had to take a look at his splits that year. In 2002 (year after breaking the record)against lefties, Bonds hit 21 HRs in 125 AB and only struck out 16 times to 57 BBs + IBB + HBP. He had an OPS of 1.532, an ISO of .592, wOBA of .595 on only a .300 BABIP. If only we had statcast data back then.
Wilyer Abreu has a 185 wRC+ against lefties this year also, after being bad against them throughout his career. But now it’s righties he hasn’t been great against (85 wRC+). I don’t know how many players have been able to suddenly become good against same-side pitching, with such a big drop in performance against the opposite side, when they weren’t before.
I’d say a small change possibly, working on a certain thing…but primarily sss
Honestly, I feel like this is overkill.
Carroll has a grand total of 78 PAs against lefties this season. With .500 BABIP.
Not everything can or should be analyzed in depth.
The closest early season result this year I could find to this was
Andy Pages who was hitting wRC+ 220 in 75PAs through 4/17 with BABIP .500.
Since then, he hit wRC+ 111 in 171 PAs with BABIP .242.
Yes, pretty extreme regression as things go, but I would bet Carroll’s rest of season wRC+ against lefties is going to be much closer to his career average wRC+ 115 rather than 219. (Bold prediction, I know.)
It’s always worth taking a look at when it is coupled with a known approach change; the stance change does appear to be meaningfully changing what he’s attacking. I agree that there’s likely some reversion toward his career averages, but I’m always curious to see the adjustment/counter-adjustment game playing out.
This. It’s cool to see the intentionality behind Carroll’s approach. appreciated the chance to understand the adjustments one of the best players in the game is making in order to at least try to get even better. That’s true regardless of how much the actual results are luck-driven.
Honestly, I feel like this comment is overkill. Lotta words to say you want less content lol.
Agree. We’re now comparing someone with a .500 BABIP over 78 PA to Bonds? This is pretty clickbait-y for FG.
Dude is playing like an MVP, might as well highlight his season and focus on what’s made him successful this year?
Wish you’d expanded out that leaderboard so we could see what the best full-season performances were. All the non-Bonds performances are from 2026, which kinda undercuts the significance of the leaderboard. Couldn’t you write this same piece about a hot lefty in May of every year?
Full leaderboard
https://www.fangraphs.com/leaders/splits-leaderboards?splitArr=1,3&splitArrPitch=&autoPt=true&splitTeams=false&statType=player&statgroup=2&startDate=2002-03-01&endDate=2026-11-01&players=&filter=&groupBy=season&wxTemperature=&wxPressure=&wxAirDensity=&wxElevation=&wxWindSpeed=&position=B&sort=16,1
I decided against making it 10 or 20 because of exactly what you say. The table, to me, isn’t that significant to the story. It’s just a brief bit of context. Like, some splits I don’t have a baseline understanding of what a good or bad wRC+ would be in those situations. I’m pointing to Bonds simply to say, hey, this is really, really high, look how well things are going right now — not that I think Carroll is suddenly Bonds. Honestly, I’d have included at least 10 in the table anyway, but I don’t like putting a big visual thing next to another big visual thing with only a few words in between. And the first visual thing is cooler and tells a nice story, so I kept that and shrunk the table. Just personal preference. I should have included a link, and I’ll try to do that next time, but it’s above for those interested (might take a few seconds to load).
Could this be written every year? Yes and no. You could certainly find the hottest hitter for a given split on June 2 and dive into the peripherals. I could write about Baldwin or Alvarez, who are playing just as well right now. But those would be fundamentally different stories. The post isn’t really about the best left-on-left hitters ever, or whether Carroll is set to join them. It’s a post on Carroll changing his stance and the positive and negative consequences of that decision so far.
“If his 219 wRC+ against southpaws were to hold…”
It’s the “on pace for” articles based on tiny splits 2 mos. into the season that get tiresome. The same thing happens in NFL when a player goes off for the first 5 games of the year.
Nah, it’s a reference point. It’s fine. The comments make pretty clear that everyone is on the same page about how sample size works.
Bellinger was a 180 wRC+ against lefties just last year.
is becoming loogy-proof actually worth being good instead of great against rhp?
I don’t really pay attention to reverse splits unless they are sustained over at least 500 PAs vs lefties and at least that many against right handers. Before that it’s probably just small sample weirdness.
I would guess that in this case this is a case of facing off against a lot of lefties who are having bad years. And it looks like although he did face off against Tarik Skubal once he also has had games against guys who are not playing so well: Trevor Rogers, David Peterson, Anthony Kay, Matthew Boyd, Kyle Freeland, Jose Quintana, and Robbie Ray. And then tonight he gets Eric Lauer.
I guess Martin Perez and Jesus Luzardo and Mackenzie Gore and Shota Imanaga have been adequate to good, but when you compare this to the right handers he has faced it’s not even close. He’s faced off against the Dodgers’ right handers multiple times, Paul Skenes, Nolan McLean (twice, I think)…
Yup. Like I believe that Yordan has no platoon split. But he’s the only lefty in the game right now that I’m confident about on that, based on adequate sample size.
>>If his 219 wRC+ against southpaws were to hold, it would be the third-best left-on-
>>left season since we began tracking in 2002:
I’m confused. The “What is WRC+?” page says WRC+ is available for all players going back to 1871. Surely left-on-left is not that problematic to look up. What exactly is it that began tracking in 2002?
Not sure. That’s just as far back as those particular splits are published on the site. I assume it has to do with the various data sources each tool pulls from.
This is awesome. And he hit another homer off a lefty last night, so curious what it is up to now.