How did Ben Brown and Kyle Harrison transform their pitches?

Case studies are how most pitchers learn what pitch design can do. They also carry traps: early-season samples are tiny, and most pitchers change several things at once. With those caveats up front, here is what Ben Brown, Kyle Harrison and Tatsuya Imai actually did in 2026, checked against full-season data from Baseball Savant and the MLB Stats API [1][2].

Ben Brown: a two-seamer built for one matchup

The problem

Ben Brown's 2025 with the Cubs was rough: a 5.92 ERA in 106.1 innings. His four-seam fastball, thrown 55.6% of the time, allowed a .315 batting average and a .566 slugging percentage [3]. Against right-handed hitters he leaned on that four-seamer 56.8% of the time, with only a knuckle curve as a complement [1]. Righties knew what was coming.

The design

Over the offseason, Brown worked with Clay Holmes at a training facility near Nashville and learned Holmes' two-seam grip. Holmes helped him kill spin efficiency on the new pitch. Brown's stated goal was a clear distinction from his four-seamer [3].

Here is the interesting part: Brown's two-seamer looks nothing like Holmes'. His best spring version had about 7 inches of vertical break and 17 of horizontal, while Holmes' sinker sits around -2 and 20 [3]. MLB.com later described Brown's pitch as a "riding" two-seamer that drops less than comparable two-seamers and tunnels with his four-seamer at the same speed [4].

The verified numbers

Source: Baseball Savant [1]. The two fastballs come out at the same speed, but the two-seamer has about 3.4 inches less vertical break and 6.4 inches more arm-side run. That separation is the whole point.

The reality check

Brown posted a 1.85 ERA over 68 innings (20 games, 8 starts) with a 24.5% strikeout rate and 7.2% walk rate, then went on the injured list on June 21 with a neck strain and was moved to the 60-day list in August [2]. Sixty-eight innings is a small sample. He also gained about 0.8 mph [4], and his usage changed alongside the new pitch, so the two-seamer is one of several changes. The lesson still holds: you do not need to copy your teacher's shape. Brown's version worked because it was different from his own four-seamer and aimed at the matchup that was hurting him.

Kyle Harrison: the rubber move that held, and the ERA that did not

The changes

Kyle Harrison arrived in Milwaukee as a hard-throwing lefty without a reliable breaking ball. Boston had given him a cutter and fewer four-seamers; Milwaukee removed the cutter, leaned on the four-seamer, and moved him across the rubber [5]. Writing in April, Jake McKibbin of Brewer Fanatic credited the rubber move with changing the angle of his slurve and noted the Brewers had also raised his fastball's ride and turned his changeup into a kick change [6].

The Statcast data confirm the physical changes [1]. His release moved about 0.9 feet toward the first-base side compared with 2024, consistent with a rubber move. His four-seam arm angle went up from 26.1 to 33.0 degrees, and its induced vertical break rose from 10.8 to 14.7 inches. His changeup's vertical break went from about +1 inch to -3.7 inches as its spin dropped from about 1,900 to about 1,300 rpm, the signature of a kick change. For more on why going up worked for him, see should I lower my arm slot.

What held all season

The slurve gain was real. He threw it more (27% to 32% usage), its expected wOBA fell from .356 in 2025 to .249 for the full 2026 season, and its hard-hit rate dropped from 45% to 28% [1]. McKibbin's April article had the slurve's expected wOBA at .197; it rose as the sample grew, but the improvement held.

What did not

Harrison carried an ERA of about 1.57 through May over 51.2 innings. From June on, it was about 6.33 over 64 innings, including a stint on the injured list for left forearm tightness from July 9 to August 5. He finished at a 4.20 ERA in 115.2 innings with a 27.9% strikeout rate and 7.1% walk rate [2].

That is still a useful season. It is also a perfect example of the main lesson in pitch evaluation: the shape stabilized much faster than the results [7].

Keep your own case study

Log every grip change in Pitch Lab

Pitch AI's Pitch Lab logs every grip change so you can see what it did to the shape. Record velo, IVB, spin, axis and release for each version and watch the movement chart as your new pitch settles.

Tatsuya Imai: elite stuff, costly walks

Tatsuya Imai came from Japan with a rare pitch: a slider that breaks to his arm side. Analyst Chris Langin argued before he signed that Imai deserved more than $200 million, mainly because his slider's NPB results (a 53.2% whiff rate against right-handed hitters in 2025) outweighed its average stuff grade. Langin also flagged command, especially against left-handed hitters, as the main risk [8]. Imai signed with Houston for three years and $54 million with opt-outs [9].

After two starts, Langin found only four MLB pitchers out of 603 who met his criteria for an "arm-side gyro" pitch like Imai's [10]. Driveline's computer-vision analysis noted that Imai seems to get underneath the ball instead of over the top like most reverse breakers [11].

The full season [1][2]: the slider held its shape all year, about 86.6 mph with about 1 inch of vertical break and 4.2 inches of arm-side movement, with 93% of them breaking arm-side. It produced a 39.7% whiff rate and a .263 expected wOBA over 660 pitches. Imai himself finished with a 4.93 ERA in 76.2 innings, a 28.7% strikeout rate and a 15.1% walk rate, with a month on the injured list for right arm fatigue.

The outlier pitch performed close to the optimistic case. The walks were the limiting factor, the exact risk Langin named. Great stuff did not cover for command.

The lessons that line up

• Solve a specific matchup problem. Brown's two-seamer was aimed squarely at right-handed hitters who were crushing his four-seamer.
• Distinct beats copied. Brown's pitch rides far more than Holmes'. It worked because it differed from his own fastball, and the second fastball shape logic explains why.
• Trust shape early, results late. Harrison's slurve held for a full season; his May ERA did not.
• Command still decides how much stuff you get to use. Imai's slider was elite, and the walks still cost him.

Keep the other side of the ledger in mind too. Tread's Ben Brewster describes an unnamed lefty whose organization converted him into a sweeper pitcher; his slot and ride dropped, his velocity fell from 94 to 96 mph to 88 to 92, and he was out of baseball within a couple of years [12]. Success stories get articles. Failed conversions rarely do.

What to do with this as an amateur pitcher

• Write down the matchup that hurts you most, the way Brown targeted the right-handed hitters who were hurting him.
• If you add a second fastball, aim for separation from your main one. Our guide to the two-seam fastball and sinker covers grips, and the changeup guide covers kick change variations.
• Judge a change over a season's worth of outings. Do not let one great month or one bad one decide it.
• Keep working on command while you build stuff.

For the evaluation process behind all of this, read how to know if a new pitch is actually good. Next, we cover pitch design without a Rapsodo.

Frequently asked questions

Shape early, results late

Track the change the way the pros do

Pitch AI's Pitch Lab logs every grip change so you can see what it did to the shape, flags pitches that overlap, and ranks what to develop next, so your next pitch has a record you can trust.

References

1. Baseball Savant, MLB Advanced Media. Statcast pitch-level search and Pitch Arsenal Stats leaderboards, 2021 to 2026 seasons; retrieved and summarized by Pitch AI on October 14, 2026. https://baseballsavant.mlb.com
2. MLB Stats API. Season, monthly split and transaction data for the pitchers named; retrieved by Pitch AI on October 14, 2026. https://statsapi.mlb.com
3. Laurila D (2026). Ben Brown Pitch-Designed a Sinker Over the Offseason. FanGraphs, February 27, 2026. https://blogs.fangraphs.com/ben-brown-pitch-designed-a-sinker-over-the-offseason/
4. Greenspan J (2026). How Brown's new two-seamer is fueling a breakout season. MLB.com, June 1, 2026. https://www.mlb.com/news/ben-brown-shining-with-new-two-seam-fastball
5. Brozdowski L (2026). The Brewers Pitching Lab Must Be Stopped. YouTube, June 29, 2026. https://youtu.be/Y9qkukpL3A4
6. McKibbin J (2026). How Kyle Harrison's Move Across the Rubber Has Finally Given Him a Breaking Ball Worthy of His Fastball. Brewer Fanatic, April 30, 2026. https://brewerfanatic.com/news-rumors/milwaukee-brewers/how-kyle-harrisons-move-across-the-rubber-has-finally-given-him-a-breaking-ball-worthy-of-his-fastball-r4466/
7. McGrattan O (2023). Stuff+, Location+, and Pitching+ Primer. FanGraphs Library, March 10, 2023. https://library.fangraphs.com/pitching/stuff-location-and-pitching-primer/
8. Langin C (2025). Tatsuya Imai Deserves $200M+: The Inflection Point for Modern Pitcher Evaluation. Unfiltered Labs. https://unfilteredlabs.com/blog/tatsuya-imai-deserves-200m-the-inflection-point-for-modern-pitcher-evaluation
9. MLB Trade Rumors (2026). Astros Sign Tatsuya Imai. January 2, 2026. https://www.mlbtraderumors.com/2026/01/astros-to-sign-tatsuya-imai.html
10. Langin C (2026). Tatsuya Imai Owns a Pitch Type. Unfiltered Labs. https://unfilteredlabs.com/blog/tatsuya-imai-owns-a-pitch-type
11. Sawchik T (2026). Investigating a mystery pitch as a test case for a new (computer) vision for pitch design. Driveline Baseball, April 23, 2026. https://www.drivelinebaseball.com/2026/04/investigating-a-mystery-pitch-computer-vision-pitch-design/
12. Tread Athletics (Brewster B) (2026). Throw Harder By Changing Your Pitching Mechanics [The Right Way]. YouTube, September 24, 2026. https://youtu.be/_EilR25ydwY

Related reading

• How to throw a two-seam fastball or sinkerThe grip family behind Brown's new pitch.
• Why adding another fastball shape can helpWhen a second fastball fixes a matchup.
• How to throw a changeupIncluding kick change variations.
• Should I lower my arm slot?Part 4: why Harrison went up.
• How do you know if a new pitch is actually good?Part 2: shape first, results later.