In the final days before the All-Star break, the Mariners threw 60 pitches to Gleyber Torres. I swear he didn’t make a single incorrect swing decision.
Fastball jammed inside on a 2-2 count? He’ll spoil it.
Slider dropped in the chase zone, low and away? Easy take.
Fastball pinned to the top edge in a full count? That’s a walk.
What’s a pitcher to do? Torres went 5-for-10 in the series with two doubles and two walks. It wasn’t the flashiest series, and it’s not really the flashiest batting line. He’s hitting .282/.390/.421 this year, which scans to me more as “very good” than “great.” But if there were an award for “most annoying at-bat,” I’d submit a nomination for Torres. Read the rest of this entry »
In one sense, we are halfway through the season. In another sense, though, we are not — we are 59.5% of the way through the season, if you count up all the games played. Regardless! There’s no baseball today, so what better time to take stock of the general state of the sport?
To that end, I’ve identified seven trends that have defined the 2025 season to this point. In no particular order:
On last Monday’s episode of the Rates and Barrelspodcast, Derek VanRiper raised a curious contradiction. “[Riley Greene is] first percentile in squared-up percentage, but 97th percentile in barrel rate, which — I’m sure there’s an explanation, I don’t know what it is just yet.” In response, Eno Sarris asked, “How can he barrel it without squaring it up?” It was a great question. In colloquial use, a squared-up ball is synonymous with a barreled ball. So what’s going on here, exactly?
The first thing to know: A squared-up ball is not necessarily a well-hit ball, as Davy Andrews highlighted when these stats were first made public last June. To understand why, one must first become acquainted with the Statcast definition of squared up. The MLB glossary entry for squared-up rate defines it thusly: “A swing’s squared-up rate tells us how much of the highest possible exit velocity available (based on the physics related to the swing speed and pitch speed) a batter was able to obtain – it is, at its simplest, how much exit velocity did you get as a share of how much exit velocity was possible based on your swing speed and the speed of the pitch.” If a hitter generates 80% of their possible exit velocity on a given swing and the ball is put in play, the batted ball is considered squared up.
We might quibble over the simplicity of that definition. In any case, as Davy showed, squared-up balls can be hit at super low speeds — if all it means is that a hitter channelled 80% of the potential exit velocity, then 80% of a half-swing is not very much exit velocity.
It’s also possible to do damage without making frequent flush contact; Greene shows us how. As Ben Clemens wrote just a couple of weeks ago, Greene is posting yet another excellent offensive campaign despite one of the higher strikeout rates among qualified hitters. He’s doing it unconventionally, swinging a ton in early counts to maximize damage. He’s also unconventional in another sense: He barrels the ball a ton while hardly ever squaring it up.
Part of the explanation for how this works is tied to the nature of swinging hard. When the bat speed statistics first dropped, it immediately became clear that there is a strong negative relationship between bat speed and the ability to square the ball up, at least by the Statcast definition. Click over to the bat tracking leaderboard, and the first thing you’ll see is this image, which shows the negative correlation between these two variables:
That’s no surprise. By the Statcast definition of a squared-up ball, slow swingers will always come out on top, because swinging slower allows for greater barrel accuracy. But it’s not all bad news for hard swingers. They also tend to produce the most valuable type of batted ball: a barrel.
Naturally, bat speed is correlated — positively — with barrel rate. A barrel, by the Statcast definition, is any type of batted ball where the expected batting average is at least .500 and the expected slugging percentage is at least 1.500. Barrels tend to be clustered in a pretty narrow exit velocity/launch angle range, somewhere north of 100 mph in terms of exit velocity and between 15 and 40 degrees or so of launch angle:
As the scatterplot below shows, the relationship between bat speed and barrel rate is extremely tight:
Greene’s average bat speed — 75.2 mph — is in the 91st percentile, so on some level, a high barrel rate and a low squared-up rate is to be expected. Even so, the spread between these two metrics is striking. His barrel rate is higher than his squared-up rate! Only one other hitter has a lower squared-up-minus-barrel rate — Aaron Judge. And that gives a hint into how, exactly, Greene is pulling this off.
Judge racks up an obscene number of barrels. Already, he’s mashed 60 this year, good for a 25.9% barrel rate. Like Greene, his squared-up rate is low — not as low, but comfortably a standard deviation below the mean. But also like Greene, Judge is amazing at converting his squared-up balls into barrels.
Nobody comes particularly close to Judge in this metric. Nearly 40% of his squared-up balls are converted into barrels, by far the highest rate in the league. (The league average is 13.6%.) As you might have guessed, Greene also excels here, ranking fifth among all hitters with at least 150 plate appearances:
So that’s the first part of this equation. Greene might not square the ball up that often, but when he does, it’s frequently crushed. The other part of the equation? Greene hits a ton of foul balls.
Greene’s 315 foul balls rank fifth among all hitters. When Greene makes contact with the ball, it goes foul 56% of the time. That mark ranks 11th out of all hitters with at least 150 plate appearances; besides Cal Raleigh, nobody else in Greene’s squared-up-to-barrel cohort fouls off nearly as many balls:
Minimum 150 plate appearances. Foul balls divided by pitches that end with contact.
All of those foul balls — in addition to his seventh percentile whiff rate — contribute to the squared-up percentage denominator, sinking Greene’s squared-up rate to the very bottom of qualified hitters. Importantly, foul balls are not part of the barrel rate denominator. The barrel rate that shows up on the Savant player page popsicles is a measure of barrels per batted ball event. A bunch of foul balls do nothing to affect a hitter’s barrel rate, but they’ll go a long way toward tanking a squared-up rate.
It isn’t necessarily intuitive to think that a hitter could be so good at barreling the ball and so bad at squaring it up. But breaking it down in this fashion, I think it starts to clarify this ostensible conundrum. Barrels are hard to come by. Even Judge, the barrel GOAT, hits one just over a quarter of the time he puts a ball in play. To be a barrel king like Judge or Greene, you don’t need to crush that many baseballs, at least on an absolute basis. But you better make sure that when the ball is in play, it gets smushed.
More than anything, I think these two data points paint a compelling picture of the modern hitter. Greene, perhaps more than any other hitter, goes for broke, almost like the anti-Luis Arraez. His swing tilt is the steepest in the sport. He mishits a bunch of pitches. He whiffs a ton. But when he connects, he does damage. And even though those damage events are relatively infrequent, they’re valuable enough to make him one of the better hitters in baseball.
On Tuesday night, Ranger Suárez aced his toughest test of the season. Taking on a red-hot Astros team (they’ve posted a 135 wRC+ over the last 14 days) that demolishes lefties (they’re first in wRC+ vs. left-handed pitchers), Suárez hardly broke a sweat. He tossed 7.1 scoreless innings, allowing just three hits, before Cooper Hummel parked his 99th pitch of the game into the right field seats.
The start may have ended on a sour note, but Tuesday’s performance was the cherry on top of an unbelievably sweet run for the 29-year-old southpaw. In his last nine starts, Suárez has allowed eight earned runs. That’s good for a 1.17 ERA over nearly a third of an entire season.
But for some reason, I’m never quite ready to believe. Maybe it’s because his primary fastball sits below 91 mph, or the absence of gaudy strikeout rates, or the lack of a single pitch that grades out as even average by Stuff+ or PitchingBot. Mostly, I think it’s because I perceive Suárez as a fundamentally streaky pitcher. He’s certainly on a run at the moment, and he’s gone on these runs before. During his breakout 2021 campaign, he compiled a 1.24 ERA in his final eight starts of the season. And there were the first three months of 2024: a 1.85 ERA over a 99 inning span. Read the rest of this entry »
Below are two changeups. Can you tell which one is better?
First up is Griffin Jax’s power changeup. He throws it over 92 mph; at two inches of induced vertical break (IVB) and 15 inches of arm-side run, it almost resembles a filthy left-handed slider:
Second is Hunter Gaddis straight change, floating up to the plate with 10 inches of IVB at an average velocity of 78 mph:
So, which one would you take? That’s a bit of a trick question: By whiff rate, these two pitches are virtually identical. Jax’s changeup ranks second in all of baseball with a 57.3% whiff rate; Gaddis’ is right behind him in third. Stylistically, they’re opposites; by the results, they are indistinguishable.
If you were creating the perfect fastball or the perfect slider in a video game, it’d be a straightforward process. Crank up the velocity, max out the vertical break, and those pitches will generally improve in a roughly linear fashion. Not so with changeups. Pitch models struggle to accurately grade these pitches because their quality can’t as easily be captured by velocity and movement in a vacuum. The Cole Ragans changeup, for example, gets a roughly average grade by Stuff+ despite performing like the best changeup in the sport over the last handful of seasons. It’s slow, and it barely moves — what makes it so good?
Lawrence Butler does a lot of things well, but he cannot hit a high fastball. Entering play on June 2, Butler had just one career barrel against an elevated fastball: A deep fly out off an 87.5 mph Trevor Williams “heater” in the dog days of 2023. In 2025, he’s whiffing on over half his swings at high heaters, per the Baseball Savant-defined shadow zones at the top edge of the strike zone. (That’s attack zones 11, 12, and 13 for the Savant search heads.)
Most of the hitters with high whiff rates on top-rail four-seamers have steep swing planes. (Aaron Judge and Luis Robert Jr. are two notable examples.) Not Butler: His 31 degree swing tilt is actually a bit flatter than the major league average. Butler’s primary issue is timing — his average attack direction on these pitches is oriented 18 degrees toward the opposite field; his zero degree attack angle is perfectly flat. Whatever the reason, it’s a clear hole, and certain pitchers are primed to exploit it. Read the rest of this entry »
If Tarik Skubal locates his fastball up in the zone, opposing hitters are probably cooked. Nearly half the time they swing, they come up empty. If they manage to put it in play, they’re unlikely to do much damage — of the six hits he’s allowed on elevated heaters, five have been singles. That lone extra-base hit? An absolute tank! A 429-foot home run, off a perfectly executed 98-mph heater on the first pitch of the game, courtesy of one Zach Neto, who is currently making a case for low-level stardom.
After a rough rookie campaign in 2023, Neto broke out in his sophomore effort, posting 3.5 WAR by playing a competent shortshop and clubbing enough home runs (23) to cover up his mediocre on-base ability. Even after missing the first few weeks of this season with a bum shoulder, the 24-year-old has managed to take another step forward in 2025: His 139 wRC+ ranks second among all shortstops with at least 150 plate appearances.
That improved line is fueled by a power surge. In just 37 games, he’s homered nine times and hit 10 doubles. The barrel rate has literally doubled, jumping from an 8.4% rate last year to 16.8%. As a result, his .589 expected slugging (xSLG) ranks eighth in baseball, just below big-time sluggers like Aaron Judge, Shohei Ohtani, and Kyle Schwarber.
Those three guys are hard swingers, perennially topping the bat speed leaderboards. But Neto doesn’t fit that profile. He stands at a slender 5-foot-11; even with a slight uptick in bat speed year-over-year, his 71.7-mph average swing speed falls below the big league average. The Angels shortstop isn’t posting elite power numbers because he’s swinging the bat hard. It’s because he’s maxing out the aggression in his approach, selling out for power and mostly succeeding. Read the rest of this entry »
If motor preferences were the final word on pitcher performance, Nathan Eovaldi would be sitting on a beach somewhere.
Eovaldi throws from a low slot, releasing his pitches from an average arm angle of 30 degrees. (Zero degrees is fully sidearm; 90 degrees is straight over the top.) Many low-slot pitchers have a supination bias. There are downsides to being a supinator — their preference for cutting the baseball tends to produce crummy four-seam fastballs — but they usually have no trouble throwing hard breaking balls; they can also more easily harness seam-shifted wake to throw sinkers, sweepers, or kick-changes. Low-slot supinators, like Seth Lugo, can basically throw every pitch in the book. High-slot pronators like Ryan Pepiot or Lucas Giolito don’t have that sort of range, but make up for it with excellent changeups and high-carry fastballs.
Eovaldi is, tragically, a low-slot pronator. Not many low-slot pronators make it to the big leagues. The pronation bias blunts their ability to throw hard glove-side breakers, and the low arm angle obviates the pronator’s nominal advantage, killing the carry on their fastball. As Tyler Zombro of Tread Athletics (now a special assistant of pitching for the Cubs) said in his primer video on motor preferences, “I know in stuff models and just off of Trackman alone, this arsenal with this slot is not that attractive.” Read the rest of this entry »
In the introduction to their 2023 Saberseminar presentation, Scott Powers and Vicente Iglesias hit on a fundamental truth about pitching: The variable that bests predicts the outcome of a pitch is the location where it crosses the plate. For a case study, look no further than this tweet from MLB.com’s David Adler about Yoshinobu Yamamoto’s splitters.
Hitters have swung 74 times at Yoshinobu Yamamoto's splitter this season …
If Yamamoto buries his splitter arm side, he’s probably getting a whiff. If it’s on the edge of the zone, it’s likely a foul ball. If it catches plate, it’s getting put in play. The location dictates the outcome.
Given this truth, pitchers who command the ball best ought to dominate. But there’s a catch. As Powers and Iglesias noted, the location is also the variable with the least predictive reliability. If you see a pitcher throw a fastball 98 mph, you can be pretty sure he is going to do it again. A dotted backdoor slider, on the other hand, does not guarantee an entire game of dotted backdoor sliders. Command is both the most important and the least reliable quality for a pitcher.
Scott Powers and Vicente Iglesias, 2023 Saberseminar
Nobody can nail the corners with every pitch. But pitchers can at least minimize the variance of their locations, finding relative reliability within the chaos of command. And in 2025, there is perhaps nobody more reliable than Jacob deGrom.
deGrom’s flat attack angle fastball and firm slider have (justifiably) built his reputation as a stuff monster. Even after easing up on the gas pedal this season, deGrom is still a darling in the eyes of the models. His overall Stuff+ is in the 80th percentile for starters with at least 30 innings pitched, fueled by his depth-y 89-mph slider. PitchingBot likes deGrom even more, ranking him in the top 10 among those pitchers. Over at Baseball Prospectus, the StuffPro model believes deGrom wields four pitches — his curveball and changeup, in addition to the heater and slider — that all grade out as plus.
deGrom’s unbelievable precision came to my attention while writing about Hunter Gaddis for a piece that was published on Monday. As part of my effort to discern whether Gaddis owed his early-season success to slider command (the verdict: inconclusive), I created a version of the Kirby Index for sliders to see where he landed. That metric measured the variance in release angles and release points and distilled those figures into a single score that captured command ability. Originally, it was designed for fastballs, which tend to be thrown to all parts of the strike zone. It perhaps works even better for sliders, which generally are thrown to fewer targets. Gaddis’ rank among his fellow pitchers was nothing remarkable, but deGrom’s name sitting at the very top caught my attention.
Minimum 50 sliders thrown to right-handed hitters.
As I wrote earlier this year, a more straightforward implementation of the Kirby Index would be to just measure the variance of the actual pitch locations. For this story, I calculated the standard deviation of the vertical and horizontal locations of a given pitcher’s sliders; once again, deGrom found himself at the top of the pack. Look at how much distance there is between him and the next closest pitcher:
Minimum 50 sliders thrown to right-handed hitters.
Random tangent here, but you have to admire Luarbert Arias for refusing to throw his junky 82-mph slider anywhere but inside the strike zone.
Anyway, measuring location densities, ultimately, could just point at pitchers who fill up the strike zone; the real test of command is a pitcher’s ability to hit his actual target. To that end, Driveline Baseball provided me with a sample of their proprietary miss distance data. Using Inside Edge tracking data, Driveline measures the distance from the intended target to the actual location of the pitch.
No surprise — deGrom’s slider miss distance ranked first among all pitchers. The league-average miss distance for sliders is about 12.5 inches; this year, deGrom is missing his target by under nine inches, nearly three standard deviations below the average. Any way you slice it, deGrom is commanding his slider like no one else in the sport.
The outcomes have been unassailable. So far, deGrom’s slider has returned a run value of -3.2 per 100 pitches thrown, the best mark for any slider thrown by a starting pitcher. Not only is he getting a bunch of swing and miss — a 38.1% whiff rate, as of this writing — it’s also grabbing a ton of called strikes. When batters do manage to put it in play, they can’t do much with it. The average launch angle on the pitch is just 2°; the xwOBA is a meek .227.
The harmless outcomes on balls in play are a function of deGrom’s targets. To right-handed hitters, he targets the classic low-away corner, breaking off the plate. Note the bimodal distribution on the heatmap — there’s a large concentration of sliders he’ll throw in the zone for strikes, and then another cluster right below the zone that generate chase.
These intentions can be seen in the filtered heatmap clusters. When deGrom throws sliders to righties in zero-strike counts, he tends to be in the zone:
In two-strike counts, he chases the swing and miss:
To lefties, deGrom shows a similar bimodal distribution, but the pattern appears reversed. In early counts, he’s aiming just below the zone; in late counts, he’s looking for called strikes. This sequence to Athletics rookie Nick Kurtz, which featured four sliders, gives a sense of the approach. On 1-0 and 2-0, deGrom tries to bait a chase, but the big lefty resists.
Down 3-0, deGrom fires a middle-middle heater in an auto-take scenario, then returns to the slider in a 3-1 count. Here, deGrom dials in his robotic precision, dotting the lower edge of the strike zone to bring the count full.
On 3-2, he goes there again. Kurtz takes it and pays the price. Though the superimposed strike zone on the broadcast says this pitch is just low, my sense is he deserves that call; if he’s consistently landing pitches within inches of his intended target, you sort of just have to hand it to him.
deGrom isn’t just painting with the slider. I calculated the Kirby Index for four-seam fastballs thrown to righties in 2025; incredibly, he also sits in first place on that list.
Minimum 50 fastballs thrown to right-handed hitters.
As nice as it would be to think that deGrom can be just as good even after dropping two ticks off the fastball, it just isn’t true. Absent improvement elsewhere, losing stuff will bring him back to Earth. But deGrom is far from stagnant. In 2019 — his last full big league season, amid the most dominant phase of his career — his fastball command measured as below average by miss distance. Six years later, it’s hard to argue his command is anything but 80-grade. And as long as the elbow cooperates, it will help him defy gravity.
Random relievers can do crazy things in small samples. Who can ever forget Nationals right-hander Justin Miller striking out 57.9% of the hitters he faced across a three-week stretch of 2018? Or Kody Funderburk’s legendary whiff explosion to close out the 2023 Twins season? Guardians reliever Hunter Gaddis is on one of these incendiary strikeout runs, and it’s driving me to madness.
Gaddis might not strike you as operating at the same level of random as Miller and Funderburk. By any set of reasonable standards, Gaddis broke out last season, appearing in nearly half of his team’s games while delivering a 1.59 ERA. But — forgive me — I didn’t really buy it. His 23.7% strikeout rate matched the league average for relievers, and his arsenal didn’t exactly justify a .205 BABIP. Given his pitch shapes and peripherals, I figured Gaddis would settle in as more of a solid middle-relief type than one of the premier backend arms in the league. And then this April happened. Read the rest of this entry »
