clock menu more-arrow no yes mobile

Filed under:

What’s going on with Kevin Gausman?

Gausman was one of the best pitchers in the NL during the first half of the season, but has struggled during the second half.

MLB: San Francisco Giants at Oakland Athletics Darren Yamashita-USA TODAY Sports

Less than two months ago, Kevin Gausman was receiving legitimate NL Cy Young consideration, going so far as to cause the NYPost to write a slightly panicked article that read “One pitcher could steal NL Cy Young from Jacob deGrom.” In addition, Trevor Bauer (currently undergoing an MLB investigation) somewhat sarcastically awarded Gausman the NL Mickey Mouse Cy Young, a reference to the best pitcher over the first 60 games of the season (as Bauer won the Cy Young in the pandemic-shortened 2020 season).

There were few candidates more deserving. Before the All-Star break, Gausman pitched 114.1 innings with a 1.73 ERA, 0.820 WHIP, and 10.3 K/9; the only qualified starting pitcher with a lower ERA in all of baseball was Jacob deGrom, who pitched three less games and 22.2 less innings. For further comparison, in deGrom’s Cy Young-winning 2019 season, he had a 2.43 ERA, 0.971 WHIP, and 11.3 K/9 in 204.0 innings. Obviously, deGrom’s ability to sustain that over roughly a hundred extra innings was what made his 2019 season so impressive, but to note that Kevin Gausman was outperforming Jacob deGrom’s Cy season in certain categories is to give a sense of just how ridiculous his first half was. There are more stats to drive the point home, as well: Gausman had a .213 wOBA against (weighted on-base average: scaled the same as OBP but accounts for the value of particular hits as well) during the first half. No single MLB batter had a wOBA that low (minimum 200 PA’s) in the entire first half (the lowest was Kevin Newman, at .232). Basically, Gausman made opposing hitters look like they weren’t even MLB-caliber.

Had the San Francisco Giants finally found their ace of the future? It certainly seemed like it. Think pieces were popping up left and right to explain Gausman’s success. The secret, according to many, was a mesmerizing splitter.

What is a splitter, also known as a “split-finger fastball?” In its most base form, a splitter is a pitch that looks like a fastball to a hitter but dives to the ground, making it a pitch that is extraordinarily easy to swing over. Nowadays, a splitter is a rarely used pitch in major league baseball because, in part, of its difficulty to throw. Many pitchers who used to rely on the splitter complained of elbow and shoulder problems, and teams began to discourage pitchers from throwing splitters. Gausman clearly has had no interest in listening: he has averaged a 22.7% split-finger usage across his career, with those rates jumping to 28.5% in 2020 and 36.6% in 2021 with the Giants. And it’s been devastating in its effectiveness, worth 35.1 runs above average for his career and 16.5 runs above average for 2021. Gausman has paired his splitter with a quality fastball (worth 15.0 runs above average for his career) and rarely used slider and changeup (both the slider and changeup are under 10% usage for his career).

But in the second half of the 2021 season, Gausman has struggled to find his footing, pitching 31.1 innings to the tune of a 5.17 ERA, 1.69 WHIP, and .360 wOBA against. He’s basically turned every hitter against him into Nelson Cruz (.361 wOBA) or Avisaíl Garcia (.360 wOBA). So what in the world is going on with Kevin Gausman?

Remember how we discussed the devastating effectiveness of Gausman’s splitter? In the first half, his splitter was worth 20.0 runs above average, or 3.12 runs above average per 100 splitters thrown. That is phenomenal. Ohtani, often reported to have the best splitter in the game (and Ohtani’s splitter is occasionally cited to be the best pitch in the game), had his splitter worth 7.1 runs above average (3.56 runs above average per 100 splitters thrown) in the first half — so Gausman’s splitter was as nearly as unhittable as Ohtani’s in many more pitches. See below for a few examples of Ohtani’s mesmerizing splitter.

But in the second half, Gausman’s splitter has been worth -3.8 runs above average (or -1.92 runs above average per 100 splitters thrown). Ohtani’s, for comparison’s sake, has been worth +3.0 runs above average (5.3 runs above average per 100 splitters thrown). The issue with a splitter is this: when it doesn’t split correctly, it ends up looking like a low velocity fastball, easy enough for a major-league hitter to tee up. See below for an example of a Tanaka splitter that didn’t split correctly (and see where it ended up):

So, have hitters been teeing off on Gausman’s splitter? That seems to be part of the answer. Recall briefly that xwOBA is the “expected” version of wOBA; that is, how a hitter should be actually performing with regards to the underlying metrics like exit velocity, launch angle, etc. It’s a way to give a sense of how quality the at-bats are, regardless of the outcome. If we look at the xwOBA by pitch type against Gausman over the months of the season, we start to see some trends emerging:

Gausman’s fastball shows a decrease in weighted on base average between July and August of this year.
Gausman’s wOBA against fastball, by month (2021)
Gausman’s wOBA against splitter, by month (2021)

As can be seen, the fastball has not been a cause for concern in the months after the All-Star break: Gausman has actually lowered the xwOBA on his fastball. But the major increase in xwOBA against the splitter provides a clue that something is amiss with how he’s utilizing that pitch. There is unlikely to be a single reason for this uptick, but there are some graphs that may shed light on the situation:

Gausman splitter spin rate by month (2021)
Gausman splitter vertical rate break by month (2021)

These two things are somewhat correlated: a greater spin rate can indicate greater vertical break on pitches. Vertical break is defined as the difference between a pitch’s release point and where it ends up (Statcast takes into account how much gravity should affect the pitch).

But what these graphs tell us is that Gausman’s splitter is moving less than it usually does. It’s not spinning as hard, so it breaks less, which means that a hitter has a better idea of where the ball is going to end up.

Gausman splitter swing and miss rate by month (2021)

That means they swing and miss less on the pitch, which allows them to either foul off the pitch and wait on a fastball, or make enough contact to drive the pitch through a gap. That’s why Gausman’s splitter has been worth less this second half; hitters are able to make contact with the pitch in a way they couldn’t before.

In short: in the first half, Gausman utilized his splitter to such effectiveness that overall, opponent batters looked like they weren’t even MLB-caliber (with a wOBA of .213 against him). From April through June, opponents batted just 0.095 against his splitter, a batting average so low even Mendoza wouldn’t touch it. But through July and into August (the All-Star break is just a convenient stopping point) something appeared to change mechanically with his splitter. The spin rate and vertical break are both down, meaning that it’s easier for opponent batters to see the pitch and make contact with it, leading to a diminished swing-and-miss rate. In July and August, opponents batted .160 against his splitter; certainly still a ridiculously good set of outcomes, but not the exact outcome you want on a kill pitch. While there are likely many other factors affecting these outcomes, because Gausman is a majority two-pitch guy (fastball and splitter), any diminished effectiveness in one of those pitches will have severe ramifications (in a way it wouldn’t, perhaps, for a three- or four-pitch pitcher, who could simply rely more on their third or fourth pitches).

The splitter alone doesn’t tell the whole story, of course. Gausman also has 16 walks in 31.1 post All-Star innings, a 4.5 BB/9, which is significantly higher than the 30 walks in 114.2 innings (2.4 BB/9) from pre All-Star break. Part of that may have to do with the declining swing and miss percentages on his splitter, but it could also point to a general lack of command.

It’s also important to note that in other metrics, Gausman looks great: his strikeout rate has actually increased, from 10.4 K/9 in the first half to 11.9 K/9 in the second. His xFIP, a metric that normalizes home run rates in order not to penalize pitchers for bad luck, has only increased by half a run, from 3.31 in the first half to 3.73 in the second — not anywhere nearly as dramatic as his ERA increase. In addition, Gausman has been suffering from a highly over-inflated BABIP (batting average on balls in play): .212 in the first half, .379 in the second (league average: .292). So while there are concerns with his splitter, as evidenced from the above (and Gausman himself citing a “mechanical issue” as the reason for his struggles), it’s also true that Gausman has simply been the victim of bad luck; luck, that fickle mistress, that loves to gallivant and dissipate without warning. It’s also the kind of thing that should normalize given more starts.

Gausman has also been through grueling personal trials recently – enough to make anyone struggle, let alone the best pitcher in the National League. But in happy news, his baby girl was born healthy, which is the most important thing of all. Here’s to hoping Gausman’s luck turns and he figures out his splitter as the Giants play deep into October.