Yusei Kikuchi just looked his best and worst

Last Saturday, Yusei Kikuchi gave up seven earned runs on two home runs. It was his second straight game of allowing at least five earned runs. As far as I’m concerned, Kikuchi was every bit as deserving of his pummelings as the box scores suggest. Against the Yankees, he was missing more than a tick off his fastball. On Saturday, he threw plenty of strikes, but he got torched on his four-seam fastball and cutter.

There are lessons to be learned from both games, but especially his most recent one. Some were things Kikuchi was doing well, and some were things he needed to stop doing. Most salient for me is that Kikuchi leaned into his slider, like I’ve been imploring him to do for months. Like, he really leaned into it. That alone is worth talking about, so I’m going to talk about it!

Consider Kikuchi’s raw slider pitch count since 2019:

One thing to note is that Kikuchi’s slider from 2019 is what we now call his cutter. So, although the graph says otherwise, this is hugely unprecedented. In its current form, Kikuchi has barely thrown his slider more than 20 times in a game. On Saturday? He pushed that up to over 30 times. I don’t know what got into him, but I imagine it has to be conscious on some level. Perhaps he recognized that his slider is his best pitch.

There’s a way Kikuchi can deploy his current cutter for it to be more effective. But I’m not here to talk about that. At least not today. I want to make it painfully clear how much Kikuchi is leaving on the table by not using his slider as best as he can. I did some of that the last time I wrote about Kikuchi, but I want to hammer it home.

Kikuchi’s slider heatmap doesn’t look like that of a typical whiff pitch:

Kikuchi uses his slider like a get-me-over pitch, and it’s not. There’s no reason to throw it in the zone as he does when he gets hitters to chase sliders out of the zone at a 46.0% clip. That makes it one of the best chase pitches from any starting pitcher. He struggles to command it at times, but he should be able to keep it at or below the bottom of the zone much more than he does. By tossing sliders in the zone, he’s allowing too many balls into play, and he’s missing out on a lot of strikes.

To illustrate where Kikuchi’s slider produces swings and misses, here’s a heatmap of his slider’s whiffs:

This is pretty much the direct opposite of his slider’s location heatmap. Swinging strikes aren’t the only way to throw earn strikes, but a buried slider for a ball is often a better outcome than a slider in the zone that’s put into play. When Kikuchi throws his slider in the zone, it gets put into play 35.1% of the time. That’s really high! When it’s out of the zone, his ball in play percentage drops to a mere 8.6%.

If that isn’t compelling enough, consider that his in-zone slider called strikes plus whiffs (CSW) is 38.7%, while is out-of-zone CSW is 34.5%. I would argue that the difference between the two is negligible, especially when you consider that his in-zone slider xwOBAcon is .306, while his out-of-zone xwOBAcon is just .099. Both are good, but when you have the choice between the two, it’s obvious that burying his slider engenders almost as many strikes, but it also offers significantly more pitcher-friendly contact.

Now, Kikuchi may have thrown his slider more on Saturday, but he’s still filling up the zone with them. The next change he needs to make is getting them below the plate. Especially because his fastball characteristics appear to be declining.

Like most pitchers, Kikuchi has seen a dip in spin rate ever since the foreign substance memo. Theoretically, less raw spin on his fastball should translate to less ride — and that’s what’s happened. His fastball ride has been reduced significantly over the course of the year:

Despite better velocity than in 2019, Kikuchi’s fastball ride has dropped all the way down to 2019 levels. In part due to his velocity, he’s been able to get away with a worsened fastball that’s hemorrhaged an inch of fastball ride. Despite all of this, I don’t find raw spin to be the most compelling factor here. Raw fastball spin is one of the indicators that will dictate if a fastball has a lot of ride, or carry. But what also matters is active spin (i.e., spin efficiency). And Kikuchi hasn’t been spinning the ball as efficiently.

Kikuchi’s fastball’s active spin rate, by game, courtesy of Max Bay:

For full transparency, these numbers are derived from an estimation formula. Hawkeye directly measures transverse and gyro spin, whereas these are inferred numbers. The specific numbers don’t matter so much here. I would find it hard to believe that Kikuchi’s active spin percentage has dipped into the mid-to-low 70s. Given that the estimations are consistent, what matters to me are the trends. And this trend shows that Kikuchi started the year with a fastball predominantly made up of transverse spin (i.e., active spin), before dipping considerably. Adding more gyro spin (i.e., inactive spin) often leads to lessened fastball ride and added arm-side movement. In other words, more sink, and more arm-side movement. That’s not good for a fastball like Kikuchi’s. But luckily, I think the culprit is mechanical.

Take note of Kikuchi’s vertical release point, plotted by game, which you may notice closely corresponds with his decreased fastball ride and active spin:

Although I went in with the presumption that dropping his arm slot has added more gyro spin to his fastball, I was shocked to find how closely they correlated. Even though it’s intuitive, I’ve never seen anything like it. At least in my baseball research. That makes for a clear takeaway: Kikuchi needs to raise his arm slot again if he wants to regain his fastball ride and active spin.

If he can’t get his release point back, Kikuchi should still be able to throw his fastball with frequency. His fastball command can be erratic, but that’s a fair trade-off so long as he continues to get solid results on the pitch anytime it’s in or even close to the zone. At this point, it’s abundantly clear that his cutter isn’t up to snuff as a primary offering. In fact, it’s his worst pitch. I’ll admit, I once thought that he should go slider-cutter heavy, but after demonstrating that his cutter neither limits hard contact nor racks up strikes, there’s little utility for it if he’s going to use it in the form that it’s currently used.

Yusei Kikuchi’s fastball is on the fritz, and his cutter has been one of the worst cutters in the league by xwOBA. That means that his slider is as important as ever, and he has the opportunity to weaponize like he’s never done before. Hopefully that’s something he took with him after his outing on Saturday. Either way, this will surely be something I revisit before long.