Let’s not bury the lede - George Kirby had a very excellent rookie year. I will not be making any Kirby-related puns - I am not going to write that Kirby had a Dreamland first season, for example, nor will I make a joke that though he may not have any plans to be a Popstar, his numbers absolutely will make up a star pitcher, etc. It should be clear why now, as you can see that the pun well dries up real fast.
I wrote about Kirby earlier this offseason, last November, speaking very broadly about his season and the success he had. If you’re looking for a lighter, more narrative piece, I recommend checking that out here!
This time around, I want to dig in with a more analytical lens, especially when looking at last year, and play with some fun numbers. There were so, so, so many interesting things I wanted to write about Kirby as a pitcher, so you’re welcome in advance for cutting out most of them (for this piece, at least). To make this easier to digest, I’m going to break this up into two categories: Things From Last Year That Were Pretty Cool, and the I’m Curious To See How These Play Out This Season category.
Things From Last Year That Were Pretty Cool
In the article above, I wrote about the huge returns that Kirby saw from implementing a two-seamer, as well as changes he made to his slider to turn it into less of a heavy cutter and more of the favored sweeper. For those who don’t want to re-read, here’s a crucial chart comparing his numbers pre-adjustments (which coincided with the midseason break) and after the changes he made.
Kirby first half vs. Kirby second half
|Pre All-Star Break||3.78||107||19.3%||0.313|
|Post All-Star Break||3.02||48||21.6%||0.249|
This is huge stuff, and my favorite part is how well these two pitches work together.
The two-seamer, like most of Kirby’s offerings, has a pretty ordinary shape. Compared to similar two-seamers (defined by Baseball Savant as within +/- 2 MPH and from within +/- 0.5 feet of extension and release, to avoid bias from slower pitches getting more time for gravity to affect their total movement), it has exactly average rise, and only 8% more horizontal break, with a slightly higher spin rate than normal.
The slider has a slightly more interesting shape - it gets an extra 4 inches of horizontal break compared to similar sliders, or 78% more break. It also has 6 inches more of break than the slider he started the season with, which was so firm that Savant categorizes it as a cutter (Kirby described it as a hard slider, though, so that’s how we will refer to it here). He also took a few ticks off the speed, dropping from 88-89 to 85-86. This all gets it approaching sweeper territory. Overall, though, it’s not an earth-shattering shape on its own.
However, the way this two-seamer/slider combo plays together is what makes them outstanding.
Looking at their spin axes, they have nearly the 6:00 separation that you seek in a combo for tunneling/deception purposes. The orange indicates his two-seamer (Savant calls it a sinker, but again, we’ll go by what the pitcher describes it as) spins from the 2:30 slot, while the slider spins around 8:45.
This is what we call spin mirroring. The idea is that if you have two pitches with perfectly opposite spin direction, it becomes very difficult for batters to distinguish between the two. The human eye and brain have limitations, and it seems that while batters can use spin axes to distinguish between, say, a four-seamer and a gyro slider, figuring out the direction of spin on the same axis is difficult.
This also leads to some really deceptive movement on the two pitches. I wasn’t able to (spend the large amount of time required to) find any examples of back-to-back two-seam/slider sequences from last year that had Pitching Ninja-level sexiness and GIF-appeal, but I did find a few that highlighted how these pitches can work in tandem, especially if it’s hard to tell which is which out of the hand.
Warning in advance: I am not a video editor, I’m a Dumb Guy trying my best, so these are not the most beautiful, well-done edits. Kirby releases his pitches from slightly different slots, so again, none of these have perfect tunneling but here’s a few from his best game of the year, his August 24 start against the Washington Nationals.
This first one is my favorite. These two pitches remain in about the same location most of the way to the plate, highlighted by the freeze-frame about 20 feet out from the plate, before breaking out to different parts of the zone.
George Kirby two-seamer/slider combo vs Luke Voit, freeze frame.— NickVitalisLL (@NickVitalisLL) February 14, 2023
August 24, 2022. pic.twitter.com/V6RkgpwjBH
This one in particular how late his slider breaks, and how drastically different these two pitches that look similar to the batter end up moving.
George Kirby two-seamer/slider combo vs Nelson Cruz, August 24, 2022. pic.twitter.com/hCSTNVF8AR— NickVitalisLL (@NickVitalisLL) February 14, 2023
This last one highlights the late break of the two-seamer and the overall trickiness of this combination of pitches.
George Kirby two-seamer/slider combo vs César Hernández, August 24, 2022. pic.twitter.com/v7Q3C6yYOb— NickVitalisLL (@NickVitalisLL) February 14, 2023
This combo not only is fearsome on its own, it also helps to take the heat off of what was Kirby’s most successful pitch last season. Speaking of...
Four-Seam Fastball: Success and RV/100
First, let’s quickly break down Run Value and RV/100 as a statistic. Run Value is a way of assigning a “score”, to each at-bat, or even each pitch in regards to its effect on probability of a run scoring. Tom Tango (the father of RV and FIP, and currently the Senior Database Architect of Stats for MLB Advanced Media) says “the idea is that every base-out situation has a run potential. And after the event, the new base-out provides a new run potential. The CHANGE in those run potential is what we attribute to the event.”
Simply, Run Value is an attempt to calculate the change in run potential for the offense, which we can break down to looking only at a pitch (ie. George Kirby’s four-seamer) - RV/100, on the other hand, or Run Value per 100 pitches, serves to turn the cumulative stat of RV into something that you can use to evaluate overall pitch quality.
Looking at RV/100, George Kirby’s fastball last year was one of the best in the league.
There were 121 pitchers who threw more than 600 four-seamers last year in the league, a cohort that consists mostly of starters and the odd relievers who throw a ton of fastballs. George Kirby’s fastball had the third-lowest (lower is better!) RV/100, behind only Nestor Cortés and teammate Luis Castillo. Alek Manoah and Justin Verlander fill out the top five. That’s good company.
It’s important not to extrapolate too much from Run Value - it’s not a predictive statistic at all, and wasn’t designed to be - it is purely descriptive of the outcomes. We can’t use RV/100 to say that George Kirby’s fastball will be this good again. We can, though, use it to see that his fastball was one of the best in the league last season at preventing runs from scoring, which has its own value.
The Top 11 Pitches from the Seattle Mariners in the 2022 Season.— Gravel (@Gravel_sense) November 1, 2022
Day 6 - #6 - George Kirby Fastball
After being recalled for the second half, opponents hit .184 with no home runs and a .246 SLG against his fastball, and this was his primary offering!
Here's a montage of it. pic.twitter.com/Bx6TzSeKvc
His fastball also ranked highly in both wOBA (6th) and xwOBA (4th) for this group, and is only a little further behind in K% (14th) and Put-Away% (20th).
Yet, for all of this success with his fastball, as I alluded to earlier, it lacks a strong shape or profile that makes it jump off the page. It’s fast, but not overly so - it’s not Luis Castillo’s turbo-ball, nor does it feature Cortés’ rise. So, what unlocked such positive results?
His Pitch Placement Was To Die For
It is impossible to talk about Kirby and his strengths without talking about his greatest weapon, his command.
In general, there are established best-practices for pitch sequencing and locations that work for certain pitch profiles. Most of these are intuitive, and utilize the movement of the pitch. For example, we like four-seamers to be up in the zone, as the ride can be deceptive and can lead to pop-ups and whiffs, whereas we like curveballs down in the zone as they start in the zone and look hittable before they dive down and out of reach, etc.
Compare Kirby’s zones to a pitcher without great command, Justin Dunn (love you, miss you):
When looking at Kirby’s pitch locations, you see tight hot-zones, and for his best pitches, they are just about where you want them. His four-seamer is up in the zone, his slider starts in the zone and dives out, while his two-seamer starts just out of the zone and breaks back into the zone. In particular, his two-seamer/slider combo both ride the black, making swing decisions very difficult. As Ben Clemens of FanGraphs wrote in his article “More Data About Sliders” (the article title of my dreams), “The most important single characteristic [in a slider] is hitting the corners of the strike zone.”
Obviously, his changeup and curveball are a little bit less comfortable for him - you can see this in the wider spread of the bender and the tendency for the change to sit in the heart of the plate. His four-seamer could sit a little higher in the zone to avoid meatballs (which he surprisingly allows at a league-average rate), too, but overall this profile works spectacularly.
For Kirby, despite the upgrades to his arsenal, the Maddux-esque command is still the glue that holds his burgeoning ace status together. None of his pitches are absolute stunners or physical specimens - we don’t look at any of his Statcast pitch data and go full Tex Avery Awwooga, Awwoooga, Boi-oi-oi-oi-oi-oing, Hummina Hummina Hummina before rolling our tongues back into our mouths. Kirby’s success, this year and going forward, is a testament to a concept that runs somewhat counter to today’s stuff-first tendency: good pitches in the perfect spot can be better than the gnarliest 22” breaking balls or the 2800 RPM, ride-iest fastball if you don’t know where it’s going.
If you’re still with me, good news! That was the end of the heavy, meaty stuff. We can move on to a little series of digestifs, if you will (you shouldn’t, don’t indulge me).
I’m Curious To See How These Play Out This Season
wOBA vs xWOBA on the Two-Seamer and Slider
If you’re familiar with wOBA and xwOBA, you can skip these two paragraphs. As a reminder, though, wOBA, or Weighted On-base Average, per MLB.com, “is a version of on-base percentage that accounts for how a player reached base — instead of simply considering whether a player reached base. The value for each method of reaching base is determined by how much that event is worth in relation to projected runs scored.” For a pitcher, we’re looking at how the ways that he specifically allowed batters on base impacted his likelihood of giving up runs.
xwOBA (Expected Weighted On-base Average) takes that same concept, but uses the Statcast expected outcomes based on batted-ball data (exit velocity, launch angle, and hopefully one day spray angle!) as opposed to what actually happened. The reason we look at this statistic in concert with wOBA is that pitchers and hitters have control over exit velocity and launch angle, but once the ball is hit, they no longer have any control over what happens (defenses, park factors, weather, etc.). By controlling for only what the pitcher controls, we can get a peek at what they “deserved” to have happen.
Disparities between the two statistics can be very interesting, and point to some small-sample size tomfoolery. Here’s Kirby’s pitches by wOBA and xWOBA in 2022:
George Kirby wOBA vs xwOBA
Most of his pitches are within a pretty reasonable margin, but his two-seamer and especially his slider are actually underperforming, based on Statcast data! Those pitches had pretty mediocre RV numbers despite, as I said above, these two pitches having the makings of legitimate weapons. That could potentially be explained by Kirby having some characteristics that means he will typically underperform his xwOBA, or batters having above average luck on those two pitches (an outcome I see as somewhat more likely given some other metrics on those pitches).
I’ll have an eye on how his two-seamer and slider perform individually on a wOBA, xWOBA and RV value this year. If Kirby pitched this well while actually underperforming on two of his highest-use pitches, the league should be scared.
Overall Slider Development
Kirby’s slider already went through some seismic shifts this season. As mentioned above, his slider started out as a hard and tight slider with more drop than sweep before getting retooled into a sweeper. It already has above average break compared to its peers, but Kirby isn’t done improving this pitch.
When he unveiled his new slider, it was typically sitting around 84-85 mph, but late-September velocity blues aside, he had it sitting around 86-87 by the end of the season. Kirby said in an interview with Eno Sarris of The Athletic that once comfortable with the pitch, he plans to get it up to the 90 mph range.
George Kirby, Nasty 86mph Slider...and Sword. ⚔️ pic.twitter.com/5lryKP085H— Rob Friedman (@PitchingNinja) August 13, 2022
If Kirby can throw a 90 mph sweeper and dot it on the corners in addition to everything else he’s doing, again, fear him.
After a 2019 season where he threw about 110 innings between college and a short stint in Everett, a lost 2020 season and a 2021 season where they eased him back with just 67 innings, Kirby was proud of his 156 innings (130 MLB, 26 MILB), though he did start to feel some fatigue through the year.
General wisdom says that it is reasonable to add around 30 innings a year to a young pitcher’s count per year, which would put him around 186 innings, more than enough to not be on any sort of innings count. I’m excited to see what Kirby can do with a full year to run near the top of the Seattle Mariners’ rotation, and what that extra time to accumulate stats will do for his ability to compete for hardware in the fall.
After debuting the new slider and two-seamer after the All-Star Break, Kirby racked up the second most fWAR among pitchers, behind only Justin Verlander. In that timeframe, Kirby had 13 starts for 2.5 fWAR, which projects out to 5.8 fWAR over a 30-start season. For his full-season totals (including his pre-ASG and pre-tweaks starts), he would project out to 3.6 fWAR for 30 starts.
Using that as a reasonable range for this upcoming season (and adding in time to perfect his arsenal and adjust to life in the big league), if he can shoulder a full workload, it’s dangerously easy to project Kirby as a top-5 pitcher in the league.
Thank you for reading a piece that is somehow both long and yet somehow barely even scratches the surface of what there is to write about with Kirby. I could easily have written another 2500 words and still feel like I’m leaving too many things out. I’ll let that in and of itself be the succinct and ringing endorsement for what Kirby has to offer this club and how important he will be this season to the Mariners’ hopes - someone who can inspire that many words on a page is likely to do a lot of work between the lines.