Armed with a six-pitch mix that features mid-to-upper 90s velocity and two wicked breaking balls, Cavalli is finally looking like the frontline starter that the Nationals envisioned he would become when they selected him in the first round of the 2020 MLB Draft. Among qualified pitchers through August 24, Cavalli ranks 18th in innings pitched (148.2), 16th in fWAR (3.1), and 13th in K-BB% (20.2%). My run estimators agree with the surface stats: his 3.81 hdERA (Huronalytics descriptive ERA) and 4.02 hpERA (Huronalytics predictive ERA) both clear the qualified-starter averages by a comfortable margin (SP mean hdERA = 4.17, SP mean hpERA = 4.43).

His season can be broken down into four different segments that showcase how he has tinkered with his release point, his arsenal, and his pitch usage to maximize his success. In the following sections, I will take a look into what changes he has made, what else he could do to continue improving, and what a potential contract extension could look like. Cavalli is under team control through 2029 and looks like D.C.’s first ace since Max Scherzer left town, so it would be in the Nationals’ best interest to keep him around into the 2030s.

Subscribe now

March 26 - April 7: Effectively Wild

Through his first three starts, Cade Cavalli’s ERA was 2.51. That’s really good! However, things under the hood painted a bleaker picture. While he was able to limit damage on balls in play, Cavalli was not generating enough whiffs. His strikeout rate was under 18%, and his walk rate was almost 15%.

His issue against righties was command, not stuff. Against RHH, his fastball had a whiff rate of 31%. However, he simply could not throw the ball where he wanted: an 87 Loc+ against same-sided hitters and a 23% walk rate against a 10% strikeout rate. He survived against left-handed hitters by getting whiffs with his breaking balls and changeup, as LHH whiffed on just 8% of their swings against his fastball.

Coming into 2026, Cavalli added a sweeper to his arsenal to complement his curveball. Early in the season, he struggled to command this new pitch, throwing it in-zone only 36% of the time while getting only an average amount of whiff and chase. His attempts to command this new pitch may have bled into his ability to locate his curveball, as the curveball was in-zone even less than his sweeper and was not getting much chase (20%). His sweeper (87 Loc+) and curveball (82 Loc+) were his two worst-located pitches in this stretch, and for Cavalli to be his best version of himself, he needed to be able to command both pitches. Something had to change.

April 13 - June 25: Sliiiiiide to the Left

Starting on April 13th, Cavalli shifted his position on the rubber by about seven inches. For his first three starts, he set up on the extreme third-base side of the rubber. From his fourth start on, he has been in the middle of the rubber. This coincided with his average arm angle dropping 3°.

His slot continued to drop. His arm angle began the year around 40°, sat around 37° in April, and settled around 34° by mid-June. His fastball added about two and a half inches of arm-side run throughout this period. His mechanical adjustments were not a one-and-done change; they started with the rubber move, and his slot kept dropping for the subsequent two months.

Over the following 10 weeks, Cavalli’s early trend of shaky command vs. RHH was completely reversed. He went from a Loc+ of 87 vs. RHH through his first three starts to a Loc+ of 110 over this period. The fix was one-sided, as his Loc+ vs. LHH went from 96 to 94 over the same stretch, basically unchanged.

In particular, he was able to locate his sweeper (114 Loc+) and curveball (103 Loc+) far better. He upped his curveball usage by 7 percentage points, mostly at the expense of his sweeper, which saw its usage drop by 12 percentage points. His curveball’s zone% rose to 49%, and the whiff rate on it was 41%, which is elite (league average whiff rate on curveballs is 31%). His strikeout rate jumped to 25% and his walk rate plummeted to 6%.

Unfortunately, Cavalli was hit hard over this period. He was able to mitigate damage against RHH with his sinker and curveball, but his fastball, which was his most-used pitch, had a .511 xwOBAcon (expected weighted on-base average on contact) against same-sided hitters. His three highest-usage pitches vs. LHH (fastball, curveball, changeup) all had xwOBAcons above .400. Five of the seven home runs he allowed over this stretch were hit by left-handers. His changeup, supposedly a weapon against lefties, had an 11% whiff rate and an 83 Loc+ grade at 13% usage vs. LHH. While Cavalli had improved his command, further tweaks were required.

June 30 - July 22: The Barber Adds a Cutter

In college, Cade Cavalli “was the kid who cut his teammates’ hair.” While it is not active, he has an Instagram account called cavs_cutz that documented his haircutting prowess. With this in mind, it is only fitting that the barber added a cutter to his arsenal. He appeared to debut the cutter on May 5, but its usage was sparse until his last outing of June. Including that start and his next four, he threw the cutter 14% of the time. While he throws it to all hitters, its effectiveness against LHH (12% usage, .234 xwOBAcon against) was crucial considering he was not getting results with his changeup. The cutter is not a pitch used to generate whiffs, but it bridges the gap between his fastballs and his breaking balls. Its presence provides new velocity and movement bands that hitters have to account for.

Perhaps not by coincidence, his fastball, which had been crushed for 10 weeks, saw its xwOBAcon drop by almost 200 points. He continued to throw his fastball in-zone at a high rate and saw its whiff rate rise too. The revival of his fastball worked differently against righties and lefties: against RHH it became a swing-and-miss pitch (39% whiff rate), whereas against LHH it became a contact-suppression pitch (.192 xwOBAcon).

His curveball was still his best pitch for whiffs, but for the first time all season, his sweeper was getting an above-average number of whiffs. If this all seems very good, it is because it is, as this five-game stretch saw Cavalli strike out over 36% of hitters faced while walking just under 5%. His RHH problem looked fully solved: he struck out 44% of the righties he faced over these five starts. His 2.30 ERA across these starts looked a lot more sustainable than the 2.51 ERA he had over his first three starts. His one blemish was his sinker, which left-handed hitters feasted on to the tune of a .600 xwOBAcon. Having had periods where both his fastball and sinker were getting barreled, Cavalli has zigged once again.

July 28 - August 24: Circuit Breakers

For the first time all season, Cavalli’s fastball is not his most-used pitch. That honor goes to his best pitch, the curveball. Combined with his sweeper seeing its most usage since his first three starts of the season, Cavalli is throwing a breaking ball almost every other pitch. Both his sweeper (5.1) and curveball (2.1) have had positive Run Values per 100 Pitches, versus a -5.4 RV/100 on his fastball, making this decision seem calculated. Velocity is up on both breaking balls, from 84/85 earlier in the season to sitting 87 now. His curveball has seen its Stuff+ grade climb from 104 at the start of the season to 121 over the last few weeks.

There is another wrinkle in this stretch. He has begun to throw his sweeper more to LHH (12% usage after hovering around 6% for most of the season), and it has been a great tool to limit damage (.189 xwOBAcon) while also missing bats (36% whiff rate).

Cavalli has been hit hard again over this period (.532 xwOBAcon on his fastball, .393 xwOBAcon overall) due to some location regression, but he is still not walking hitters the way he was early on. He has allowed eight home runs over these six starts, and a .230 BABIP is propping up his overall run prevention. An interesting trend: his curveball is dominating RHH (.188 xwOBAcon) while getting hit hard by LHH (.468 xwOBAcon). Whether this iteration of Cavalli is a phase or the finished product is something the season’s final month may answer.

Further Improvements

The decomposition of Cavalli’s hpERA shows where Cavalli has thrived and where he can continue to improve. His stuff grades well (106 Stuff+) and he has an 81st-percentile ground ball rate. What has held him back is his command. With a 99 Loc+ on the season, he is below the qualified-starter mean of 102. During his five-start July peak, he had a 104 Loc+. If he had this type of command with the current level of stuff (109 Stuff+ in the latest period), he would move from the top quarter of starting pitchers into roughly the top 20 starters. Being able to prove that the command he showed in July is a level he can consistently get to is the next step.

Further tweaks to Cavalli’s pitch mix could bear fruit. His fastball is the hardest pitch to justify at its current usage against both righties and lefties. Against LHH specifically, there are two changes he could make. The first is upping cutter usage in place of the fastball. This keeps a hard pitch in the profile and should prevent some of the damage that his fastball routinely gives up. The second would be throwing his sweeper more. Sweepers usually have a platoon split that favors throwing them to same-handed hitters, but Cavalli’s sweeper is more of a “turbo slurve.” Cavalli’s curveball is his best pitch, but upping its already high usage could let hitters get too familiar with it.

It will be interesting to see if he keeps his sinker usage low, or brings it back against RHH in place of his fastball. While there may be fewer whiffs, there should also be less damage. The other open question is if the current home run spike he is enduring in August is simply noise or if it is the result of too many breaking balls in the strike zone.

Keeping Cavalli in D.C.

Cavalli is under team control through 2029. He will finish 2026 with between two and three years of service time. The Nationals front office should be interested in locking him up with a deal that buys out at least two of his free agent years.

Comparable extensions signed by starting pitchers with between two and three years of service time:

• Tanner Bibee (3/22/25): five years, $48M ($9.6M AAV), sixth year $21M club option

  • 3.3 fWAR in 173.2 innings in 2024

• Germán Márquez (4/2/19): five years, $43M ($8.6M AAV), sixth year $16M club option

  • 4.1 fWAR in 196 innings in 2018

• Blake Snell (3/21/19): five years, $50M ($10M AAV)

  • 4.7 fWAR in 180.2 innings in 2018

Cavalli is on pace to finish 2026 with ~3.7 fWAR over ~177 innings. He has a $4M club option for 2027 that the Nationals can choose to decline, although declining would then enter him into the arbitration process. Should the Nationals pick up his option, they could then extend him for another four years and $56M, tacking on a club option for 2032 at ~$25M. This would buy out up to three of Cavalli’s free agent years and pay him a guaranteed $60M over five years ($12M AAV).

The argument for Cavalli accepting this deal is simple: it guarantees him a life-changing amount of money after he missed two full years of baseball. This is his first full season in MLB, and long-term arm health is never guaranteed. An extension at this number would top the class of extensions signed by starting pitchers with similar service time. There is risk on both sides with this deal, but it is the kind of risk both should be willing to take.

Cavalli has provided Nats fans with hope for the future thanks to his breakout 2026 season. He has tinkered all season to keep himself ahead of opposing hitters, and his changes have mostly paid off. It seems likely he will keep experimenting as he looks to unlock his next level. D.C. should be excited to see Cavalli headline the rotation over the next three seasons (and hopefully longer), and it is not out of the question that he gets the Game 1 start whenever the Nats make it back to the postseason.

Subscribe now

Sources: MLB, Baseball Savant, FanGraphs, my website Huronalytics (huronalytics.vercel.app)

Subscribe now

The Pitchers

RHSP Luis Perales

It is not hard to see why the Nats traded for Luis Perales. They traded a more polished starting pitcher in lefty Jake Bennett (96 Stuff+ in 2026) to the Red Sox to chase the upside (108 Stuff+) of a starting pitcher whose fastball sits in the upper-90s and can touch 101. However, his best pitch has turned out to be one that he has only started throwing with more frequency recently. Over his first 11 outings, Perales threw his splitter 14% of the time, which was the lowest usage of any pitch. His strikeout rate sat at 18%, his whiff rate was average, and his chase rate was below average. Over his last 10 outings, his splitter usage has risen by over 20 percentage points, as he is now throwing the pitch 35% of the time, most of any of his pitches. The splitter generates whiffs on 44% of swings against it, and when it does get put in play, over 60% of batted balls are on the ground, leading to a .270 xwOBAcon. Since he increased its usage, his strikeout rate has jumped to 24%, both his whiff and chase rates have increased five percentage points, and his xwOBAcon has dropped .032 points. While his ERA has increased over this timeframe, his SIERA, which is a better indicator of true skill, has dropped by almost a full run.

What holds Perales back is his average strikeout rate combined with his 13% walk rate. Fellow Red Wing Riley Cornelio has walked 12% of the AAA hitters he has faced this season, but he has also struck out 36% of them, making the walks less impactful. Perales suffers from below-average locations on everything but his splitter. He actually is in-zone at about a league-average rate, but his misses are too often uncompetitive, with 16% of his pitches being thrown in waste zones compared to the league average of 12%. As of now, his profile carries legitimate reliever risk, but if he can continue to generate whiffs with his new splitter-heavy approach while simultaneously cutting down on his walk rate, there is front-end starter potential to dream on.

LHSP Jackson Kent

Jackson Kent is currently the inverse of Perales: 93 Stuff+ and 108 Loc+. While he does not overwhelm hitters with elite stuff, he locates exceptionally well. He has above-average extension at almost seven feet which helps his velocity play up a bit higher than what the radar gun reads. Through 13 starts, he has generated whiffs at a very high rate (32%) while suppressing damage when balls are put in play against him (5% barrel rate). He has different approaches based on hitter handedness, as he mixes in all four pitches against righties while attacking with mainly his slider and fastball against lefties. I would like to see him mix in his changeup to lefties more, as he has good feel for it and it would help him cover the inner third and set up his slider.

An interesting comp for Kent is former National MacKenzie Gore. Gore is another lefty who has average stuff but above-average locations. Gore’s arsenal is larger than Kent’s, as he also throws a cutter, sinker, and sweeper, but the main four pitches share similar movement profiles. Kent misses more bats at AAA than Gore does in MLB, but that gap will likely compress when he is in the big leagues. Kent currently projects as a back-end starter, but if he can find a tick or two more of velocity, his projection increases.

LHRP Erik Tolman

The fact that Erik Tolman is this close to the big leagues is a testament to his perseverance. Not only has he had Tommy John surgery twice, he also blew out his right knee in a freak accident in 2024, tearing his ACL, PCL, LCL, and hamstring. In 2026, he is missing bats at an elite level in Rochester, with a Z-Whiff rate of 28% (97th percentile). His changeup grades out as his best pitch: 113 Stuff+, 112 Loc+, 59% ground ball rate. He throws it 11 mph slower than his fastball while making it drop eight inches more than one would expect given his arm angle.

I believe Tolman would be in the big leagues already if he could find a way to limit his walks, especially against left-handed hitters. Considering he is supposed to have the platoon advantage against them, he has walked an unfathomable 22% of lefties faced. His fastball command holds back the rest of the profile: he throws it in the “shadow-in” parts of the zone at a below league-average rate regardless of count, which limits how effective his changeup and slider are, as he throws those at the bottom edge in the counts that matter. Tolman has the stuff to be a medium-to-high leverage arm if he improves his fastball command.

RHRP Jack Sinclair

Jack Sinclair sits 96 from a low slot with a plus sweeper (111 Stuff+, 105 Loc+). In terms of whiffs, he has actually been much more effective against left-handed hitters, as his fastball (29%) and changeup (40%) generate whiffs against lefties at an above-average rate. While he has struck out 28% of the lefties he has faced, he has also walked 15% of them, and when the ball is put in play, he has been crushed, with 16% of batted balls being barrels. Against righties, the strikeouts are down (24%), but so are the walks (9%) and barrels (5%), as he opts for a sweeper-sinker combination instead of the fastball-cutter-changeup trio he prefers against lefties. Sinclair looks like a low-to-medium leverage reliever at the moment, although his five-pitch mix does lead one to believe he could be stretched out as a starter. The pitch that could decide his future is his cutter, which grades as a strong offering by stuff (109 Stuff+) but is often thrown in poor locations (79 Loc+). Against lefties it has been the source of much of his damage allowed.

LHRP Will Dion

Will Dion, one of the four players acquired from Cleveland in the Foster Griffin trade, generates whiffs (29%) and strikeouts (32%) at a high rate despite having a fastball that sits 91 and tops out at 93. What immediately stands out about Dion is his release point. With an average arm angle of 65°, Dion throws from a slot similar to that of Toronto’s Trey Yesavage. His fastball has an extreme cut-ride profile, playing well at and above the top of the zone. He works his splitter off of his fastball extremely well, and both pitches have hitters whiffing and chasing at above-average rates. What pulls everything together is his ability to locate (114 Loc+, 96th percentile) and limit walks (4%, 93rd percentile).

The main worry with Dion is his low ground ball rate. He carries an extreme fly ball profile, which combined with his lack of velocity, could lead to an inflated home run rate in the big leagues. An interesting comp for Dion is Tampa Bay’s Ian Seymour, another lefty with below-average stuff but above-average locations who features a changeup as his putaway pitch. Seymour is having some success in the big leagues with this formula, which is proof of concept.

RHRP Yovanny Cruz

The Nationals may have finally found their closer of the future. Since debuting in pro ball in 2017, Yovanny Cruz has played in five different organizations. The Nationals, who acquired him as part of the Luis García Jr. trade, will be the first to give him an extended look at the major league level. Cruz has elite stuff (120 Stuff+), but his fastball has been hit hard to the tune of a .473 xwOBAcon. Like Perales, Cruz has a secondary pitch that is his most effective offering. In Cruz’s case, that pitch is his slider (128 Stuff+). When he locates it, it gets whiffs and ground balls at an elite rate.

Despite a solid walk rate of 8%, Cruz’s locations are below-average (92 Loc+). When he gets ahead, he does a good job of putting hitters away, but when he falls behind, he struggles to get back in the zone. When behind, 49% of his pitches are in-zone (league average 57%), and 28% are in the chase zone (league average 21%). Hitters are much less susceptible to chase in non-two-strike counts, rendering a pitch that could be competitive when he is even or ahead in the count more likely to be taken for a ball. If all things go right for Cruz, a comp for him is Twins reliever Jeff Hoffman. Hoffman also features a heavy fastball-slider-splitter arsenal with the slider generating whiffs and the fastball getting crushed. Cruz has the stuff to become a high-leverage reliever, but he needs the locations to follow suit.

RHRP Jake Bird

If Jake Bird only had to face right-handed hitters, he would be an above-average reliever at the major league level. Unfortunately, that is not the case, and lefties give him all sorts of trouble. Acquired alongside Cruz, Bird is a veteran who has over 250 career MLB innings with the Rockies and Yankees. Against righties, he uses a sweeper-sinker pairing that gets whiffs and prevents damage. Against lefties, he features his curveball and fastball more frequently, but still throws his sinker too often (-3.8 expected run value per 100 pitches). While the whiffs still exist, lefties have a .477 xwOBAcon against him, which is in the 1st percentile. The Nats will be tasked with figuring out how to mitigate the quality of contact against him, as Bird showed he was a competent reliever for Colorado in 2025 before being traded to New York.

The Hitters

1B/DH Abimelec Ortiz

Abimelec Ortiz has a chance to make Nationals fans quickly forget the pain of trading Luis García Jr. Ortiz can flat-out rake, with his 125 Hitter+ sitting in the 93rd percentile. While he is defensively limited to first base, there are very few warts in his AAA profile. He swings at the right pitches, makes contact at an above-average rate, and destroys the ball when he makes contact. With a max exit velocity of over 116 mph, Ortiz’s raw power is rivaled by few. Combining that power with pulled fly balls has led to great expected outcomes for Ortiz. In his limited MLB at-bats he has continued to showcase the tools that got him to the big leagues. He has shown elite bat speed, albeit in a small sample, and I am all in on Ortiz being the Nats’ first baseman for years to come. His ceiling is Atlanta’s Matt Olson. While he likely will never be of the defensive caliber that Olson is, his bat can match Olson’s.

The one thing that Ortiz can improve on is his ability to hit left-handed pitching. While he still makes hard contact, his overall contact rate drops substantially. He is especially susceptible to lefty sweepers, with a whiff rate of over 50%. The Nationals’ usage of platoons could benefit him, as seen by the usage and subsequent production of Luis García Jr. before he was traded.

3B/1B Yohandy Morales

When he makes contact, balls frequently leave Yohandy Morales’ bat at high exit velocities. With 23 home runs, a Hard-Hit% in the 96th percentile, and a Barrel% in the 90th percentile, there is no doubting his power. He rarely pulls the ball in the air, which is an issue the Nats believe their player development staff can fix at the individual level. With his great surface-level stats in AAA, including a 135 wRC+, fans have been clamoring to see Morales get a chance in D.C. What more does he have to improve?

What Morales has to improve is his ability to consistently make contact. His overall swing decisions are fine (98 SD+), but when he swings at pitches in the strike zone, he is only making contact with 73% of them (8th percentile Z-Contact%). What is even more troubling is that he whiffs on nearly 30% of in-zone fastballs that he swings at. Combined with his fly ball profile skewing toward the opposite field, it may be that Morales has a timing issue and/or a long swing. While Morales may ultimately end up being an above-average major league contributor, it is understandable why the Nats have continued to let him develop in AAA as he works through these issues.

OF Andrew Pinckney

The one player to rival Abimelec Ortiz’s 99th percentile maximum exit velocity? Andrew Pinckney, come on down. In May, Pinckney hit a home run at 119 mph, which ties him with Pittsburgh’s Oneil Cruz for the hardest-hit ball in 2026. While his overall profile has plenty of room for improvement, Pinckney has shown elite bat speed (in a tiny sample), which can lead to some loud contact (108 BB+, 80th percentile xwOBAcon). However, those sounds do not occur as frequently as he or the Nats would like due to a combination of his below-average contact rate (91 CT+) and above-average ground ball rate (56%). Pinckney has the tools to be a successful MLB hitter, as shown by his home run off of Cy Young candidate Cristopher Sánchez in his full debut last week, but his inconsistency at the plate (31% strikeout rate) has him grade out closer to a league-average AAA hitter so far (101 wRC+) as opposed to a good one.

INF Seaver King

A recurring theme is emerging. Seaver King hits the ball hard but has flaws elsewhere in his profile. For King, this is his chase rate (37%, 19th percentile). He makes above-average contact on his in-zone swings, which is an improvement from Morales and Pinckney, but unlike the other two, who chase at an average to below-average rate, King expands the zone far too often. His swing decisions grade out 20 points below an average MLB hitter, and over half of that deficit sits in the chase row. Getting into good counts has not been the problem; giving them away has. Like Morales and Pinckney, he also struggles to pull the ball in the air. The hard contact is real, with a Hard-Hit% in the 84th percentile, and it is the reason he has still produced (116 wRC+) despite the poor swing decisions. If he were called up right now, he would likely be overmatched by superior MLB arms. If he can improve his chase rate, King could be an impact player in 2027 and beyond.

2B/UTIL Phillip Glasser

Unlike the last three hitters, Glasser has no issues making contact. His main issue is doing damage. A .300 batting average in 2026 is impressive, but a 24% Hard-Hit rate and only one home run are worrying. Glasser’s offensive profile is similar to that of Detroit’s Zach McKinstry, who is currently sporting an 80 wRC+ in the big leagues. Without bat speed data in AAA, it is impossible to say for certain, but it would not be a surprise if Glasser’s is below average. Sacrificing some contact ability for increased damage would improve his future outlook, as right now he profiles as a AAAA player. Making it to the big leagues at all is an achievement in itself, and Glasser should get there, but if the power never comes, he might not stick around.

The Future Is Now

While Rochester is not currently full of impact players for future Nationals teams, there are absolutely arms and bats who can help. Some may be just a tweak or two away from unlocking potential, and the new Nationals regime has already shown it is capable of doing so. With the three new arms acquired via trade already in the major league bullpen, Jackson Kent set to make his Nationals debut this week, and Ortiz and Pinckney getting their first tastes of the big leagues, the future is becoming the present. The next contending Nationals team already has some of its building blocks in place, but plenty of spots remain open, and the final two months of the season could serve as an audition for the names above to claim one.

Data courtesy of MLB, FanGraphs, and my website (huronalytics.vercel.app).

Subscribe now

The Top Line

• Hitter+: the composite, scaled like wRC+ (100 = league average, roughly 40 points per standard deviation). It is a weighted z-blend of BB+ (65%), CT+ (28%), and SD+ (7%). It is a measure of what the hitter does as opposed to wRC+, which only measures results.

• wRC+ / xwRC+: park-adjusted offensive output (results) versus its expected, or deserved, value. The gap between them, like the gap between wOBA and xwOBA, is the luck signal; BABIP is another luck tell.

The Main Three Components

• SD+ (Swing Decisions): the run value of swing/take choices, scored by zone and count. It rests on five attack zones, from in to out: heart (middle of the plate), shadow-in (edge strikes), shadow-out (edge balls), chase (clearly off the plate), and waste (well off it). Good decisions mean swinging at the heart and shadow-in and taking the shadow-out, chase, and waste.

• CT+ (Contact): a leverage-weighted contact rate (did you hit the ball when you swung?), weighted so two-strike and heart-zone whiffs cost the most. Reported across the same five zones.

• BB+ (Batted-ball / damage): how good the contact is when made, a 58.5/41.5 blend of two inputs:

• con+ (from xwOBAcon, the Statcast exit-velocity-and-launch-angle model): how hard and optimally the ball is struck.

• sp+ (from xwOBAsp): how well-placed the ball is, scoring each batted ball by its launch-angle-and-spray location against league outcomes. League average xwOBAsp is .375.

• SACQ%: the binary cousin of xwOBAsp, the share of batted balls landing in “quality” zones (league wOBA at least .500).

Supporting Context

• Bat Tracking: bat speed, attack angle and ideal-attack-angle rate (the swing’s loft), and squared-up rate (barrel accuracy). These partly explain the “why” behind BB+.

SD+: Swing Decisions

SD+ asks the first question of a plate appearance: did you swing at the right pitch? The leaderboard shows each hitter’s overall mark plus the five attack zones as percentiles (red = strength, blue = weakness). Reading left to right, the first two zones (heart, shadow-in) reward swinging; the last three (shadow-out, chase, waste) reward taking. These percentiles take the count into account: a swing on a pitch in the shadow-out zone with two strikes is penalized less than a swing on that same pitch with zero strikes.

Elite Eyes

• Nuñez, Wood, Mead, Vivas

Maybe surprisingly, Nuñez is the best decision-maker on the team. He has above-average swing rates on pitches in the zone and below-average swing rates on pitches out of the zone. Mead and Vivas are great at laying off pitches out of the zone, and Vivas does a great job swinging at pitches in the heart of the plate. Wood’s swing decisions are the most fascinating: He refuses to chase pitches out of the zone, but pairs that with substantially below-average swing rates at pitches in-zone. His approach is a weapon that allows him to key in on specific pitches, but he occasionally will take very hittable pitches because he is clearly looking for something else. He may be a rare case where swinging a bit more would benefit him, but because he’s doing so well already, maybe he’s fine just the way he is.

Solid to Average

• Tena, Lile, Young, Abrams, Millas, Crews

Young and Millas mirror Wood’s in-zone passivity but without the power to justify the takes. Lile is the opposite, as he is great at attacking pitches in the heart zones but struggles against pitches outside of the zone. Abrams is the most actionable: solid almost everywhere but with a real chase/waste problem that is single-handedly dragging his profile down. In a small sample size, Crews is chasing off the plate, and this expand-the-zone tendency is likely feeding into his ground ball rate.

Chase-Prone

• Ruiz, García Jr.

Ruiz shows good restraint on pitches just out of the zone, but swings at a very high rate of pitches in the chase and waste zones. García Jr. is fine on strikes, but expands relentlessly on pitches out of the zone. He has a 41% chase rate and 3.3% walk rate. His chase issues are the biggest thing holding him back from being an above-average hitter.

CT+: Contact

CT+ asks the second question: when you swung, did you make contact? Same five-zone layout, leverage-weighted so heart-zone and two-strike whiffs count most.

Elite Bat-to-Ball

• Mead, Ruiz, García Jr., Vivas

This group rarely misses. Mead is the cleanest version, as he is excellent in and around the zone. Ruiz and García Jr. make above-average contact on pitches that they should not be swinging at, which contributes to soft contact (and, for García Jr., ground balls). Vivas makes contact on the right pitches but has minimal power in his bat, which ultimately is holding him back.

Solid

• Millas, Young, Nuñez, Crews, Lile, Abrams

Abrams’ profile is the most curious. He is whiffing on pitches in the zone but making contact when he expands. He is missing the pitches he should crush, which could be capping how much of his power he can get to. Crews has no real holes and is hopefully trending up. Like Vivas, Millas makes consistent contact on in-zone swings, but his lack of power has not led to positive results.

Below-Average

• Wood, Tena

Wood’s whiffs are his biggest issue, but because of how much damage he does when he does make contact, they have not caused him many problems so far. Tena’s contact profile is alarming: He does not miss in one spot; he misses everywhere. His strikeout rate of 32% reflects his inability to make consistent contact.

BB+: Damage

BB+ asks the third question: when you made contact, was it any good? It splits into con+ (how hard, from xwOBAcon) and sp+ (how well-placed, from xwOBAsp), with SACQ% (spray-adjusted contact quality) as the placement check and bat speed as the physical engine.

James Wood

Wood is in a tier of his own. His BB+ is driven by elite raw contact and bat speed, and is supported by good, balanced placement (111 sp+). He has an average exit velocity of 95.5 mph (98th percentile), hard-hit rate of 58.5% (97th percentile) and barrel rate of 23.0% (99th percentile). Everything he gets his bat to is hit hard.

Above-Average Pop, With an Asterisk

• Tena, Abrams, Mead

All three pair real or solid contact quality with a caveat. Tena has legitimate power (111 con+, 69th percentile bat speed), but the lowest ideal-attack-angle rate on the team (35.7%, 5th percentile) and a poor contact rate keep that power theoretical. Abrams is more balanced and has the steepest swing on the team at 13.2°, which is geared for the pull-side power he is looking for. Mead is contact-quality over placement. His SACQ% is in the 18th percentile because his flat swing leaves too much of his contact in low-value launch zones rather than the high-value air bands, which is the main thing keeping him from stronger production.

Average and the García Jr. Conundrum

• Young, Lile, García Jr., Crews

García Jr. has the most interesting profile in this group. His raw contact is fine (101 con+, 47.4% hard-hit rate), but it is placed below-average (92 sp+, 47.2% ground ball rate). The tempting read here is “he needs to elevate,” but his swing is already built to lift (9.9° attack angle, ideal-attack-angle rate of 61.4%, which is 92nd percentile). The grounders come from chasing pitches, so the fix is the chase, not the swing. Young and Lile have been placement-over-power profiles so far this season, even acknowledging Young’s eight home runs. Both have below-average bat speed that limits their damage output. Crews’ sample size is too small to generate a BB+ score (minimum 80 batted balls required), but the shape is García Jr.-like: above-average raw contact hit on the ground.

Minimal Impact

• Ruiz, Nuñez, Vivas, Millas

These four guys share a common problem: lack of bat speed. Slow bats will do less damage no matter how good the eye or the contact is. Ruiz has a 99th percentile squared-up rate (82.8%), but his 4th percentile bat speed produces well-struck but soft contact, which his extreme air-pull profile (31.5%, 98th percentile) dresses up into a misleading slugging line. Nuñez, Vivas, and Millas have nothing behind the bat at all; their value lives entirely in some combination of swing decisions, defense, and base running.

Hitter+: Putting It All Together

Wood is the only hitter who is plus on the dominant axis (BB+ 137) without a fatal hole elsewhere (SD+ 116, and a CT+ of 92 he can carry given the power). Mead is the model of balance, above-average on all three, which is why he is a clear second despite never being elite on any single one.

Why the bottom tier is the bottom tier: Nuñez (SD+ 118, the best on the team) and Vivas (SD+ 109, CT+ 105) have genuinely excellent decisions and contact, yet land at 73 and 75 Hitter+, because BB+ is 65% of the formula and they have none (BB+ of 86 and 85). In today’s game, bat-to-ball skills without impact are a complementary profile, not a carrying one. Their value is real but it lives in swing decisions, defense, and base running, not the bat.

The hard-versus-placed divide: the con+/sp+ split sorts the middle of the roster into two camps with opposite fixes. Hard contact, poor placement (García Jr., Crews, Tena) is the higher-upside group, as the exit velocity is already there, though García Jr.’s focus should be on plate discipline rather than a swing change. Soft contact, good placement (Ruiz, Millas, Young, Vivas) is the more fragile group, squeezing value out of loft and spray that their bat speed does not really support.

Conclusion: How Sustainable Is This?

The short answer is yes, and the reason is that the breakout is not built on luck. Weight every Nationals hitter by playing time and the lineup has produced a 107 wRC+ while its skill deserves a 109 xwRC+. An offense whose results trail its underlying quality is the opposite of a mirage; if anything, the bats have been slightly unlucky as a group.

The internal luck nets out, and it nets out in the team’s favor. The biggest over-performer is Keibert Ruiz, whose 116 wRC+ sits on a 67 xwRC+ built on soft contact and no walks; that production is coming down. Abrams is a milder case of the same thing, a 142 wRC+ vs. a 122 xwRC+: still one of the team's best bats, but as of now seemingly more likely to settle into the 120s than to hold a 140+ line. But the regression candidates are outnumbered by hitters who have been under-rewarded and should rise: Wood himself (162 wRC+ vs. a deserved 178 xwRC+), Mead (129 wRC+ vs. 138 xwRC+), Millas (57 wRC+ vs. 76 xwRC+), and especially Crews, whose ugly 55 wRC+ rides a .192 BABIP that will not persist. The hitters due to fall are fewer and lighter than the hitters due to climb.

Two things give the offense a high floor. The first is James Wood, a franchise cornerstone (175 Hitter+) who is elite in both contact quality and plate discipline and has actually been a touch unlucky. An attack anchored by a 22-year-old hitting like this does not crater. The second is hidden upside: Washington has been a top-five offense despite getting almost nothing from Crews, whose skills (above-average bat speed, sound contact) say he is far better than his line.

The honest caveats are about ceiling and shape, not collapse. The lineup is top-heavy: one elite bat, two strong ones in Mead and Abrams, and a long tail of average-to-light hitters, several of whom (Nuñez, Vivas, Millas) have no power to fall back on and carry value only through their eyes, gloves, and legs.

Put it together and the verdict is clear: this is a sustainably above-average offense whose hot start is backed by skill, not luck. Whether it stays top-five or settles into merely very good depends less on regression than on health and on whether Crews and Mead push the supporting cast forward. The floor is high, and the ceiling could be higher.

Data courtesy of MLB, FanGraphs, Baseball Savant, and my website, Huronalytics

Subscribe now

Four weeks into the 2026 season, the Washington Nationals pitching staff has struggled mightily. Through April 22, they have allowed the most runs in baseball (155). While there is not one specific reason for their struggles, I looked into each pitcher’s arsenal to see who I think could add pitches to improve their performance. To do so, I built a multivariate normal projection model over arm angle, extension, and each pitcher’s current arsenal, conditioned on his actual metrics. The output is a predicted velocity, induced vertical break (IVB), and horizontal break (HB) for any hypothetical new pitch, plus projected outcome rates: whiffs, called strikes, chase, ground balls, hard hit, and expected weighted on-base average on contact (xwOBAcon).

I ran it on three Washington righties: Clayton Beeter, Gus Varland, and Cole Henry. The same pattern kept coming up: each of them is missing pitches that comparable arms across the league already throw, although usually at low usage. These are pitches the model says fit and where the closest MLB comp is often already throwing the exact thing.

One note before the fun part. Projected outcome rates come from finding pitchers whose actual pitch shape is close to the predicted shape, weight averaging their results, and shrinking toward league average for that pitch type and handedness. Comps are filtered to within eight degrees of the Nats pitcher’s arm angle and then ranked on fastball shape plus arsenal overlap. It’s late April, so sample sizes are not huge, meaning we should take direction over precision.

Beeter and Varland: High Slot, Different Holes

Both Beeter and Varland throw from a higher slot, with Beeter at 52° and Varland at 50°. Neither has a comfortable answer for right handed hitters. They get to that problem from different directions, and their comps suggest they’ll get out of it differently too.

Beeter is a two pitch reliever. He throws a fastball at 96 mph with 18” of IVB and 3” of HB, and a gyro slider at 87. Early on, he has been effective vs. LHH (.134 wOBA, .224 xwOBA), but he’s been getting lit up by RHH. Righties have a .639 xwOBA on his fastball alone. This isn’t a velo or shape issue. His fastball is good. The issue is righties see two pitches that both break glove side or stay straight, allowing them to sit over the plate because nothing runs in on their hands.

Hunter Bigge of the Tampa Bay Rays is the closest trajectory comp for Beeter. Bigge throws from a similar slot (55°), has a nearly identical fastball at 97 mph with 19” of IVB and 4” of HB, and also sprinkles in a splitter and curveball. Bigge isn’t doing it perfectly yet, but he’s a working example of where Beeter could go.

Gus Varland is a three pitch reliever on paper, but he throws his fastball (95 mph) and slider (88 mph) over 95% of the time. He has thrown a changeup five times this year, all vs. LHH. His slider has been the main problem so far. His xwOBAcon on it is .478, the worst contact quality number in this study. The fastball looks better according to xwOBA (.227) but is getting hit hard 70% of the time. Against righties his xwOBA is .388. The actual wOBA has been lower than that, but he could be due for regression.

The Varland comp picture is easier than Beeter’s, because his FF/SL/CH skeleton already lines up with a handful of MLB relievers. Nick Mears of the Kansas City Royals is the tightest match. Mears throws from a 55° arm angle and has a fastball at 95 mph with 19” IVB and 8” HB. Varland’s fastball is 95 mph with 18” IVB and 9” HB. Mears’ full 2026 arsenal already includes a low usage sinker (93 mph, 14” IVB, 13” HB) and a low usage curveball (81 mph, -14” IVB, -6” HB). A sinker and curveball are the two pitches the model wants Varland to add.

Peter Lambert and Hunter Brown (both on the Houston Astros) fit the Varland archetype too, carrying five or six pitches around the same skeleton. The model’s suggestion for Varland isn’t to invent something new; it’s “do what these guys are already doing.”

Sinker: Covering the Inner Edge

Here are the sinker projections:

Neither projection is especially exciting on its own. However, the pitch’s standalone quality isn’t the point. The point is what the threat of an arm side fastball does to everything else in the arsenal.

Righties facing Beeter or Varland right now can credibly assume nothing will run in on their hands. That assumption allows them to stay over the plate. A sinker forces them to respect an option they currently don’t have to honor, which makes the fastball less predictable and turns the slider sharper as a chase pitch. There’s also a specific thing about sinkers that helps here: they have the highest called strike rate of any pitch type in baseball (22%). Hitters take them. So adding a sinker doesn’t only introduce a shape variation. It preserves the early count strike throwing that both guys already get from their fastballs, just with different movement.

The sinker also limits damage when it does get hit. Both projections sit above 50% ground ball rate (Beeter's at 54.6%, Varland's at 52.9%), well clear of where their current fastballs live. Beeter's fastball is getting hit in the air 67% of the time, and Varland's gets hit hard 70% of the time, which is exactly the kind of damage a sinker is built to cap.

The comp data backs this up. Mears already throws a sinker that is basically the Varland projection. A sinker is a normal thing for a righty with Varland’s fastball to throw. The case for Beeter is the same.

Curveball: Steal Strikes Early

The curveball is the get me over play.

Curveballs do about half their CSW (called strikes + whiffs) work via called strikes. The league average called strike rate for curveballs is 17%, and that’s the argument for both of them. Beeter’s slider is more of a chase pitch than an in-zone pitch. He has the ability to land it for strikes, but ideally its value comes from swing and miss. His fastball gets taken for strikes 22% of the time, but the fastball is the pitch hitters are sitting on. He would benefit from a curveball to steal strike one early in counts (0-0 and 1-0). He could also use it as a put-away pitch, especially vs. LHH. A curveball also introduces a new velocity band into his arsenal, forcing hitters to adjust between 83 and 96 in any given plate appearance.

The same logic applies to Varland. He would be able to drop it in for a strike early in counts, and mix it in later in counts as a chase pitch. A curveball also stretches his velocity band a lot. His current range is 88 to 95, so adding a 79 mph curve nearly doubles the band, which is a real timing wrinkle. Mears’ actual curveball is basically the Varland projection, so again, this is a thing righties at this slot can do.

The contact projections split the two. Beeter’s curve grades out well on contact (xwOBAcon 12 points below league average for the pitch type). Varland’s projects worse, 11 points above average, which reflects that not every high slot righty can throw a 79 mph curve that misses the barrel.

Beeter’s Bonus: A Changeup

Varland already throws a changeup, albeit in a limited manner. Mears throws one from a similar slot at higher usage. Beeter doesn’t throw one at all. His arsenal stands to benefit from a pitch that fades away from LHH.

Beeter’s early numbers vs. LHH have been great: .134 wOBA, .130 xwOBA on the slider. This isn’t about fixing something broken. It’s about protecting what’s working. Right now lefties see fastball and slider, and as the league adapts to him, he will have to adapt back. A changeup gives him a secondary option that takes pressure off the slider vs. LHH and lets him save it for two strike counts.

The shape fits. A differential of 13” of IVB and 10” of HB between his fastball and changeup should be large enough to be effective. It’s a chase and whiff pitch that works off the fastball’s plane. The direct comp here is Bryan Baker of the Rays. Baker throws from an arm angle of 56°, with a nearly identical fastball at 97 mph, 19” IVB, 4” HB, and an already established FF/SL/CH arsenal. Baker isn’t elite, but he’s a legitimate three pitch RHP from this slot, and his existence means the development path is real.

Cole Henry: Bring Back the Sinker and the Changeup

Henry’s situation isn’t quite like the other two, as he throws from a much lower slot (23°), and the pitches he should be adding aren’t new pitches for him. In 2025, he was a five pitch pitcher, throwing a sinker and changeup along with his fastball, cutter, and sweeper. The case is for reintroducing the sinker and changeup, as the model says the two pitches he’s set aside are the ones his slot was built for.

Against righties, Henry has a .487 wOBA / .416 xwOBA. His sweeper, the supposed glove side weapon vs. righties, has a .499 xwOBAcon. When righties square it up, they punish it. He needs another option vs. right handed hitters to set up his sweeper, and the sinker is the clean one.

The comp picture for Henry is the tidiest of the three. Casey Legumina of the Seattle Mariners throws from a 19° arm angle, has an almost identical fastball (93 mph, 15” IVB, 12” HB vs. Henry’s fastball, which is 94 mph, 14” IVB, 13” HB), and his current 2026 arsenal is four seam, sinker, sweeper, and changeup.

Low-Slot Sinker

17” of projected arm side run would be about 80th percentile movement. Henry’s slot gets there naturally. His lower arm angle with a fastball grip already produces 12.7 inches of run, and a sinker grip extends that into a pitch that starts at a righty’s hip and breaks toward his back foot. Of the three sinker projections in this piece, Henry’s is the one most in line with how his arm is already working.

The called strike rate is what I like most in this projection. At 22%, this is the highest of any projected pitch in this analysis. Hitters already take sinkers at high rates. It can be an early count strike grabber to set up the fastball and sweeper.

The 57% projected GB rate is the fix for the sweeper’s contact problem. Right now righties are lifting his sweeper and his fastball in the air. A sinker with a high ground ball rate reroutes those same balls into the dirt. When righties do connect, contact might be loud, but it will more often than not be a loud ground ball, which is a worthwhile tradeoff over loud fly balls.

Changing Up

Henry is already adequate vs. lefties (.291 wOBA, .256 xwOBA), so the changeup isn’t fixing a problem. It’s protecting what’s working and giving him a non-fastball option vs. lefties. Sweepers are usually more of a platoon pitch, with their effectiveness concentrated vs. same-sided hitters. Even if not elite, a changeup would offer a pitch that moves away from LHH with substantially more drop than his fastball. His old changeup had 19” of HB, but I would be curious to see if a changeup with less fade would mirror his fastball better.

Closing

The argument for adding these pitches is to elevate the performance of their current pitches. None of the pitches I recommend adding are projected to be elite, but their introduction into each pitcher’s arsenal would force hitters to respect an additional pitch or two. For all three pitchers, the goal is to prevent hitters from sitting on their best pitches. The model is pointing at the gap between what comparable MLB arms carry and what these three are throwing.

For Beeter and Varland, the priorities mirror each other: sinker for RHH, curveball for early count strike grabbing both sides, plus a put-away option especially vs. LHH. Henry’s slot is lower but the logic is similar. Being able to throw sinkers in will set up each pitcher’s best pitches (fastball and slider for Beeter and Varland, fastball and sweeper for Henry) away. A changeup to LHH will tunnel off the fastball and provide a pitch in a similar velocity band as a breaking ball that moves the other direction.

Caveats

All three pitchers have 130 to 200 total pitches on the season so far. Per pitch splits are working with 30 to 60 PA per side. Projected outcome rates come from league wide comp data that’s reasonably well sampled, but the candidate side stats aren’t yet stable. Same goes for the comps from other teams.

The comp pool is made up of pitchers who throw the proposed pitch. Those are by definition the ones who can throw it, which biases the comp set upward. If Beeter develops a changeup, his version might be worse than the average changeup thrower’s, because the average changeup thrower is a guy who has already developed a usable changeup.

The model assumes average command. A pitch with elite shape and bad command is worse than the projection says. A pitcher with elite command may be able to beat the projection.

The compositional argument is from first principles, not from a controlled study. “Adding a sinker makes the fastball better” is a structural claim I would defend. The magnitude of the effect is harder to pin down.

With all of that: the cases for these pitches hold up, the development paths aren’t unusual, and the closest comps in MLB are already throwing the recommended pitches. If even one of these additions sticks across the three guys, the bullpen looks different by midseason.

All data courtesy of MLB, FanGraphs, Baseball Savant, and my website: huronalytics.vercel.app

Irvin has a strong tendency to locate his pitches to his glove side. That means they land on the inner third against LHH and the outer third against RHH more often than not. For some pitches, this works well. For others, it can be a real problem.

To measure Irvin’s location, I built movement-matched location heat maps for each of his pitches. The process: take his average velocity, induced vertical break, and horizontal break for a given pitch type, then find the 2,000 most similar pitches thrown by any right-handed pitcher in that matchup, regardless of pitch classification. Those 2,000 pitches form the heat map backdrop, showing the expected whiff rate or expected weighted on-base average on contact (xwOBAcon) at every spot in and around the strike zone (pitches in the “waste” zones are excluded). Irvin’s actual locations are overlaid on top, with a covariance ellipse showing where he clusters and markers for whiffs (W) and barrels (B).

Each pitcher in the comparison pool gets a “location score”: the average heat map value at their pitch locations. For whiff maps, higher means the pitcher is locating pitches where hitters swing and miss more often. For xwOBAcon maps, lower means the pitcher is locating pitches where contact quality should be the weakest. Two pitchers with the same movement profile can have very different outcomes depending on where they locate, and this framework attempts to capture that difference.

Good Locations

The glove-side tendency isn’t all bad. For pitches designed to break away from the hitter, Irvin’s natural command bias puts the ball exactly where it should go.

Slurve vs. RHH

This is his best-located pitch, and it’s not close. His slurve (labeled by Statcast as a curveball, but it does not have enough depth to really be a curveball) against RHH has a 32.2% expected whiff location score, which is 97th percentile among comparable pitchers.

On average, Irvin’s slurve breaks 16.9” to his glove side with a bit of depth (-3.5” IVB). Against RHH, that action sweeps the ball low and away, which is exactly where Irvin’s command wants to put it anyway. His ellipse is in the low-glove-side chase zone. The heat map confirms that’s the single best spot for this movement profile, and Irvin finds it naturally. While it is only a sample size of twelve pitches, the locations are as good as it gets, and Irvin’s glove-side bias is a repeatable tendency.

Cutter vs. LHH

Irvin’s decision to lean into his cutter against left-handed hitters is backed by the numbers. Cutters are generally used to generate weak contact, and Irvin’s xwOBAcon score of 0.374 on his ranks in the 84th percentile among comparable pitches. By staying up and in (dark red) instead of down and in, (dark blue), he is able to mitigate potential damage when LHH make contact against his cutter.

Sinking Sinker Results?

The sinker is where the glove-side tendency shows up most clearly but the results haven’t caught up yet. Among comparable pitchers, Irvin’s sinker xwOBAcon location score of 0.323 ranks 79th percentile, which is above average. His expected whiff location score of 12.5% (42nd percentile) is mediocre, but sinkers aren’t usually intended to be whiff pitches. They’re built to generate weak contact, and by that measure, Irvin’s sinker is doing its job.

According to the heat maps, the best xwOBAcon zones for this movement profile are down and in to RHH. Javier Assad of the Cubs leads all comps at 0.306 by throwing his sinker right there, where the arm-side run attempts to jam the barrel and produce weak contact. Irvin’s ellipse drifts glove-side, matching the locations of his other pitches. His results have been fine so far, but it is fair to question whether the contact-suppression numbers hold as the season goes on or regress toward where his locations suggest they should be.

Bad Locations

Fastball vs. LHH

Irvin throws over three-quarters of his four-seam fastballs to left-handed hitters. While he has generated a fair few whiffs with it, his expected whiff location score is 12.3%, which is 6th percentile among comparable pitchers. His ellipse sits in the center of the zone with no real tilt toward the edges or up where elevation would help. The xwOBAcon score of 0.422 (27th percentile) confirms this: lefties are seeing a hittable fastball in a hittable spot.

Irvin’s fastball sits 93 mph with 14.5 inches of induced vertical break and 8.3 inches of arm-side run. All three numbers are near the 50th percentile. An ordinary fastball can still play, but it can’t survive in the middle of the zone. It doesn’t have the velocity to overpower hitters or the ride to elevate past barrels. It needs to be located better. Seth Lugo of the Kansas City Royals throws a nearly identical fastball (92.3 mph, 15.8” IVB, 8.3” arm-side movement) and leads all comps with a 21.7% expected whiff location score. He gets there by locating up and away, which is the ideal spot for a pitcher’s fastball. Zac Gallen’s fastball leads in expected contact suppression among comparable pitchers (0.385 xwOBAcon) by being thrown higher in the zone where batted ball quality drops off. Irvin’s fastball sits in the middle of the zone, and both metrics suffer for it.

This is the core of the problem. The four-seam to LHH is 27% of all his pitches, the single most common pitch-matchup combination in his arsenal, and his locations with it do not suggest he should be getting whiffs or suppressing damage.

Changeup vs. LHH

Irvin’s changeup is his least-used secondary, and the location data suggests there’s a reason for that. Among all right-handers who throw a changeup with comparable movement (85.4 mph, 4.7” IVB, 17.0” arm-side), his location quality ranks 33rd out of 33 in both expected whiff rate and xwOBAcon.

A changeup with this much arm-side run is built to be a chase pitch down and away to LHH. That’s exactly where Michael Lorenzen of the Colorado Rockies throws it. Lorenzen leads all comps on both metrics (37.5% expected whiff, 0.265 xwOBAcon) by consistently locating his changeup down and away. Irvin’s changeup sits low and middle, which is too far from the outside edge to get whiffs and too close to the middle of the plate to suppress damage. His glove-side tendency robs arm-side pitches of the runway they need to work.

Conclusion

For Irvin, this isn’t a velocity problem or a stuff problem. He has a six-pitch mix with league-average movement across the board and a full tick of velocity back from his 2025 floor. The usage changes he’s made (more cutters to LHH, a sweeper replacing his old slider) are smart adjustments that lean into what his command already does well. But 35% of his pitches are four-seam fastballs, and the vast majority of them land in spots where this movement profile doesn’t play.

The gap between where Irvin is and where he could be isn’t about necessarily adding velocity or another pitch. The template for what good looks like already exists in his own arsenal. He just needs to apply it to the other side of the plate.

Data courtesy of MLB, Baseball Savant, FanGraphs, and my website: huronalytics.vercel.app