Kohei Arihara Has Brought His Unique Pitch Mix to MLB
This recent tweet from Daren Willman caught my eye:
#Rangers Kohei Arihara threw 7! different pitches in the first inning. pic.twitter.com/X86J6NpvFH
— Daren Willman (@darenw) April 20, 2021
Partially it was because I was already intrigued by Kohei Arihara, but the pure absurdity of this fact stands out, too. He threw seven different kinds of pitches in his most recent start against the Angels. Seven! In one inning! The only other player to accomplish this feat in 2020 was Arihara’s countryman Yu Darvish, who coincidently did so in seven different innings last season. Since 2008 (the first season with public PITCHf/x data), there have been 909 instances of a pitcher throwing at least seven different pitch types in a single inning. When you think about how many innings of regular season baseball have been played since 2008, you can appreciate how rare this type of occurrence is.
If you are not familiar with Arihara yet, let me fill you in. The right-hander came over from NPB after playing six years for the Hokkaido Nippon-Ham Fighters. He was posted going into his age-28 season and signed a two-year deal with the Rangers for a total of $7.44 million, $1.24 million of which was the posting fee awarded to his now-former club. At the time of signing, this seemed to be a very good deal for Texas. Arihara lacks the explosive fastball and mind-bending breaking pitches frontline starters in MLB today; his strikeout rate his final season in Japan was 19.4%. But his repertoire was deep. Eric Longenhagen wrote up a scouting report back in December outlining his pitch mix and fastball velocity, comparing him to other high-profile players from NPB. Eric concluded that Arihara profiled as a back-of-the-rotation type, placing a 40 FV on him. Getting a guy who can fill the back of a rotation for about $3.7 million per year (when accounting for the posting fee) seems like a solid deal to me.
Arihara’s seven pitch mix consists of a four-seam fastball, a slider, a cutter, a sinker, a changeup, a splitter, and a curveball. He does not lean on any one pitch very often. In his four starts so far, his most-used pitch is his four-seamer, which he’s thrown 30.2% of the time per Baseball Savant. He has only thrown seven curveballs (2.3% of all pitches), with batters seeing his other five pitches on at least 23 occasions. As I mentioned above, the number different pitch types he’s thrown is unusual. The only pitcher to throw more in 2021 is Darvish, while six pitchers have thrown six different pitches:
| Player | FA (%) | FC (%) | FS (%) | SI (%) | SL (%) | CU (%) | KC (%) | CH (%) | No. of Pitches |
|---|---|---|---|---|---|---|---|---|---|
| Yu Darvish | 19.2 | 37.3 | 5.3 | 4.3 | 20 | 6.1 | 6.9 | 0.3 | 8 |
| Kohei Arihara | 30.5 | 13.6 | 7.5 | 13.3 | 20.1 | 2.3 | 0 | 12.7 | 7 |
| Corbin Burnes | 1.7 | 50.4 | 0 | 13.3 | 11 | 10.7 | 0 | 13 | 6 |
| Joe Musgrove | 13.4 | 24.8 | 0 | 4.4 | 27 | 19.9 | 0 | 10.6 | 6 |
| Aaron Civale | 29.6 | 20 | 0 | 4.6 | 11.4 | 16.1 | 0 | 18.2 | 6 |
| Kyle Gibson | 12.5 | 14.8 | 0 | 39.4 | 19.9 | 6.1 | 0 | 7.4 | 6 |
| Brett Anderson | 2.2 | 10.5 | 0 | 47.6 | 12.4 | 0 | 10.5 | 16.9 | 6 |
| Chris Bassitt | 21 | 25.2 | 0 | 33.7 | 5.1 | 5.9 | 0 | 9.1 | 6 |
Among these pitchers, only Civale and Musgrove have thrown their most common pitch less frequently than Arihara. Among qualified pitchers who throw fewer than six pitches, only Jordan Montgomery, Sandy Alcantara, Anthony DeSclafani, and Charlie Morton have thrown their most common pitches at a lower percentage than Arihara has thrown his fastball. And no other qualified pitcher has thrown at least six pitches 7.5% of the time like Arihara has (for those curious, Darvish only throws three of his eight pitches that often):
| Player | No. of Pitches Thrown At Least 7.5% of the Time |
|---|---|
| Kohei Arihara | 6 |
| Corbin Burnes | 5 |
| Danny Duffy | 5 |
| Marcus Stroman | 5 |
| Joe Musgrove | 5 |
| Anthony DeSclafani | 5 |
| Brandon Woodruff | 5 |
| Aaron Civale | 5 |
| Brett Anderson | 5 |
| Nathan Eovaldi | 5 |
| Zach Eflin | 5 |
| Pablo Lopez | 5 |
| Jordan Montgomery | 5 |
| Dylan Bundy | 5 |
| Adam Wainwright | 5 |
Clearly the breadth of Arihara’s arsenal and his willingness to go to basically all those pitches at least some amount of the time is unparalleled. Is that enough to offset his lack of bat-missing stuff? His average four-seam fastball velocity sits at 92.1 mph, a tick and a half lower than the league-wide figure of 93.5 mph. Only the splitter has generated a whiff rate significantly above league average (66.7%) and he reserves that pitch almost entirely for use against right-handed hitters, with 22 thrown to right-handed hitters versus just one to a lefty. He has yet to show any of his pitches can reliably fool hitters when they have the platoon advantage. His changeup does get significantly more fade than average, so that should help fortify his line against left-handed hitters. And while the stuff does not look impressive on paper, he does generate movement via seam-shifted wake on his sinker, slider, changeup, and splitter, which helps offset the lack of raw pitch movement with deception.
Arihara has only struck out 17.1% of the batters he has faced, which is far enough below the league average to question the sustainability his 2.21 ERA through four starts (his FIP sits at 3.05). He certainly is being aided by a low BABIP (.259) and HR/FB (5.6%). That is not to say there isn’t anything encouraging to be found in his stats so far. The strand rate is average. While he may not be keeping hitters off the base paths via the strikeout, he has limited free passes. His walk rate sits at a minuscule 3.9% (tied for 11th amongst qualifiers), giving him an elite K/BB of 4.33.
I tend to think the depth of his arsenal and the deception derived from the seam orientation of some of his pitches will allow him to mitigate results on contact. The 2.21 ERA he has accumulated thus far is too rosy of an outlook for Arihara, but run suppression numbers in the range of his xFIP (3.93) seem attainable. With enough bulk, that is great production from a back-end starter. Given his unusual style relative to most MLB pitchers in 2021 and his plethora of offerings, I have found Arihara to be a joy to watch.
Carmen is a part-time contributor to FanGraphs. An engineer by education and trade, he spends too much of his free time thinking about baseball.
What is the difference between a sinker and splitter? How different is the pitch movement that PITCHf/x can definitively tell the difference?
you are asking the right question, just instead of sinker (which is faster) vs splitter (slower) its changeup vs splitter
in fact, there is basically no difference, they both have low spin and both are similar speeds and both are similar movement profiles. Eno Sarris has some writing on this and groups them as the same pitch.
For some pitchers (Tyler Clippard), the velocity and movement is different enough that they are indeed two distinct pitches. However, many pitchers that throw both don’t have much of a difference in velo and/or movement, and I feel it’s a little ridiculous for that pitcher to throw both. However, I suppose there might be a psychological component where the pitcher believes that both pitches are worthy offerings or that more pitches = harder to hit.
it could be as simple as he has days where he controlling one the pitches better than the other and prefers having both in case he doesnt have the feel for one of them that day?
Take a look at Alex Cobb on BaseballSavant. He throws a splitter and a sinker.
This year, his sinker has 18.2 inches of vertical drop.
This year, his splitter has 31.1 inches of vertical drop.
Average velocity and spin is lower on the splitter than the sinker.
a sinker is a type of fastball, a splitter is a type of off-speed pitch.
The question I’m interested in is why a pitcher would throw both a splitter and a changeup, since they’re similar pitches.
Different kinds of changeups have different movement depending upon the grip.
A circle changeup can be thrown with more arm side run (horizontal movement). So a left-handed pitcher can throw a circle changeup that goes down and away from a right-handed batter, and vice versa.
The reason to throw a splitter is because it looks more like a fastball, then breaks more sharply downward as it gets to the plate. It can often be thrown faster than other changeups. Because of that, it can have less horizontal break than other changeups.
One reason to throw both is if you want different horizontal movement profiles. A left-handed pitcher may not want to throw a circle changeup to a left-handed batter because it can move down and in towards the batter. Instead, the pitcher may prefer to throw a splitter, which may have less horizontal movement.
Different movement.
i dont know why the sinker continues to get thrown in the conversation since as you and others have pointed out it a completely different pitch in the fastball family with a velo similar to a 4 seamer while the splitter despite its name is more a off speed pitch.