Jakob Junis and the Disappearing Fastball

This is Kyle’s first piece as a FanGraphs contributor. Kyle is a lifelong baseball fan who has always been enamored with the numbers and analytics behind the game. He has written for PitcherList on the pitcher GIFs team and for his own personal blog, covering topics from player analysis to the draft, mostly focused on the Angels. Kyle is a senior at the University of California, living in the Bay Area and studying education and math. As an aspiring teacher, he wants to think and write about the game of baseball from the perspective of an educator.

From 2017-20, Jakob Junis was a below-average starter for the Royals, posting a 4.78 ERA and 4.77 FIP. After a poor 2021 that saw him pitch out of the bullpen, he was non-tendered by Kansas City and signed with the Giants. Fresh off a 107-win season that was bolstered by career resurgences from hurlers like Johnny Cueto, Anthony DeSclafani, and Alex Wood, San Francisco clearly saw a path to improvement for Junis. Mostly serving as a member of the starting rotation, Junis has done just that – his FIP has improved by about a full run (3.83). While he hasn’t thrown harder, landed more strikes, or added movement to his pitches (his sweeping slider has actually lost about an inch of break), he’s made one important change, and it’s one that is emblematic of pitching philosophy in the modern age.

Junis has replaced a lot of fastballs with sliders and is on pace to be the first starting pitcher (knuckleballers excluded) since Jorge Sosa in 2007 to throw an individual secondary pitch over half the time. Others have come close – Lance McCullers Jr. threw curveballs about 47% of the time from 2017-18, and Jhoulys Chacín came tantalizingly close in 2019 when sliders comprised 49.6% of his pitches. But if Junis keeps pace in his last few starts, he will likely be the second starter ever to throw a single breaking or offspeed pitch the majority of the time.

One thing to note: PitchInfo actually has Yu Darvish having thrown sliders over half the time in 2021, but they were classifying nearly all of his cutters as sliders. Darvish himself has talked about manipulating the shape and speed of his cutter, and sources like Baseball Savant also consider those pitches to be cutters.

For most of baseball history, fastballs (and sinkers — I’ll be referring to four-seamers and sinkers collectively as fastballs here) have been the bread and butter of every pitcher’s arsenal. The idea of a “fastball count” has been treated as baseball gospel for over a century. Indeed, from 2008-10, the first three years of the pitch tracking era, over two-thirds of pitches thrown in hitter-friendly counts were fastballs. However, having prescribed situations to throw fastballs has hurt pitchers far more than it has helped, and fastballs are the only pitch type that has registered a negative run value in every season we have data for.

Teams have figured this out, of course, and the percentage of fastballs thrown has declined precipitously, especially since the introduction of Statcast in 2015. New tracking technology has told us a lot – fastballs generate lower swinging strike rates than other pitches, get hitters to chase out of the zone less than other pitches, and get hit the hardest when they’re put in play. Over the past decade, league-wide fastball usage has dropped by about 10%, and with 48.6% of pitches in 2022 being fastballs and sinkers, for the first time they’ve been relegated to a plurality of all pitches thrown rather than a majority:

The concept of throwing a secondary pitch over half the time has always been familiar to relief pitchers. While no one from yesteryear ever matched Matt Wisler’s 91% slider rate in 2021, plenty of relievers have leaned heavily on their breaking stuff. Kiko Calero had four consecutive seasons where sliders comprised over 50% of his pitches, while Scott Sauerbeck had a 2.30 ERA in 2002 while throwing 71% curveballs, a rate that even beats out Pierce Johnson, the most curveball-happy reliever of today. Before Devin Williams’ airbender became one of the best pitches in baseball, another Milwaukee reliever, Matt Wise, led with his changeup in 2005 and ’07. For decades, relievers have been allowed to ditch their fastballs because they’re almost never required to face the same batter multiple times. Now it seems that starters are going in the same direction.

In addition to the fact that fastballs as a whole are less effective than breaking and offspeed pitches, the current era of baseball has blurred the line between starter and reliever more than ever. The average start in 2022 is 10 pitches and two-thirds of an inning shorter than it was a decade ago; teams now consciously manage around the Times Through the Order Penalty, and the “five and dive” guy, which entered the baseball lexicon as a critical term, is now a solid descriptor for most starting pitchers. Junis, while starting 17 of his 21 games so far this year, has occasionally piggybacked with left-handed openers like Sam Long and Scott Alexander, sparing Junis from having to face opposing teams’ top bats three times in a start. The result of this shift in thinking by teams has been apparent in the way starters’ pitch mixes have changed, even across just the past five years:

Since 2017, the percentage of starting pitchers who have thrown something other than a fastball or sinker most of the time has more than doubled, and in 2022, a record 61.9% of starters are doing so. (I chose to treat cutters as non-fastballs here – while there are a few cutter-majority pitchers, like Corbin Burnes and Graham Ashcraft, the vast majority of starters who throw cutters use them in conjunction with another fastball, and most of them use their cutter to play off their primary fastball, not the other way around.) While many up-and-coming pitchers enter the league with a secondary-focused arsenal, looking at the pitchers who have evolved during their time in the majors is particularly illuminating. Junis’ fastballs returned a -43.7 pitch value during his time in Kansas City, likely due to the pitch’s subpar velocity and shape. Many others like him, including Shohei Ohtani and Chad Kuhl, have proven that the sweeper is not only a very effective pitch, but also relatively easy to learn and develop.

It’s not just the sweeper, though. Charlie Morton has leveraged his natural talent for spin into elite seasons in his late 30s, throwing more curveballs than four-seamers, while converted reliever Jeffrey Springs has kept his steady diet of changeups from his bullpen days. This shift in pitch usage is likely to continue – starters like Sandy Alcantara, Trevor Rogers, Jordan Montgomery, and many others have slowly reduced their fastball usage over time and as of right now are barely over the 50% threshold set for this metric.

Of course, this isn’t to say that an elite fastball isn’t a huge asset for a pitcher. Left-handed batters are sporting an OPS north of .800 against Junis, whose sinker/sweeper-focused arsenal makes him vulnerable to opposite-handed hitters. The Royals’ current group of young pitchers — guys like Kris Bubic, Jonathan Heasley and Jackson Kowar, who were known in the prospect world as hurlers with middling fastballs and elite changeups — has yet to find big league success despite limiting the use of fastballs. On the other hand, the likes of Justin Verlander, Carlos Rodón, and Jacob deGrom have carried their fastballs to pure dominance, finding themselves atop our pitch value leaderboards year after year. As with many things, all-around excellence is needed to make a a pitcher truly great.

As many fans and analysts alike lament modern pitching trends, from openers to starter load management to the countless number of faceless middle relievers on the Triple-A shuttle, the shift in arsenal usage among starters has been one of the most interesting developments to watch. Teams that have improved in this department have dramatically increased the effectiveness of countless pitchers whose ceilings were previously thought to be capped serving as swingmen or in long-relief duty. And while Jakob Junis will likely be the first starter in over a decade to use a secondary offering as his primary pitch, you’ll soon see plenty more just like him.

All statistics are as of September 18.