Never Ne-ver Hang a Sli-der

You can feel it in the pit of your stomach. Symptoms of watching a slider hang in the zone from your favorite pitcher include but are not limited to jaw-clenching, toe-curling, fist balling, hot-flashes, “we’ll never make the playoffs” thoughts, and of course, a bubbling sensation in the lower abdomen region. Dylan Cease fans beware, the following GIF may conjure up some of the previously listed symptoms. Consult a part-time fantasy baseball professional before dropping Cease from your fantasy roster. Do not watch this video if you’ve experienced any slamming to the ground of nacho cheese, spit-taking of Diet Coke on the fans in front of you, or early departures from the ballpark only to turn on the radio on the car ride home in an attempt to justify your decision:

While everyone in Friar gear holds their breath, Dodger fans drool and, along with the batter, participate in the abnormal feeling where time stands still and a hidden source of energy surges to a full charge preparing to physically (batter) or emotionally (observer) unload on the hanger in question. Today, however, we’ll focus only on the gut sinking feeling of watching a breaking ball float in the heart of the zone from the pitcher’s, or fans thereof, perspective.

If you’re looking for a more scientific approach to defining a “bad pitch”, head over to PitcherList and read Kyle Bland’s model-driven definition. For this article, I’m taking the simple route and calling a slider in Statcast’s zone 5 a mistake, or a hanger. In 2024 pitchers threw sliders 21.0% of the time. That’s the second most utilized pitch behind the four-seam fastball at 31.1%. The more you decide to zig and zag across the freeway while putting hot sauce on your breakfast burrito during your morning commute, the more likely you are to be involved in a crash. The same goes for pitchers throwing sliders in the zone. For example, Dylan Cease threw the highest percentage of sliders in 2024 at 54.2%. As a result, he also recorded the most zone 5 sliders, 88.

Notice that of all those hanging sliders, only a portion ended up in play. Only a portion of those did damage. Two home runs and three doubles? Some of you eat that for breakfast. Dylan Cease may have tossed up 88 animal-style In-and-Out burgers, but they only encompassed 2.8% of all the pitches he threw in 2024. Drill down further and of all the sliders that Dylan Cease threw in the zone, only 14.7% of them landed in zone 5. He recorded a 34.9% CSW% on those zone 5 pitches.

The problem with simply running a search on BaseballSavant is that people always say, “What’d, you just run a search on BaseballSavant?” That’s why I pulled the search data into a friendly coding environment and aggregated the heck out of it. I divided zone 5 sliders by in-zone sliders to make a more meaningful comparison. Further sub-setting to only include pitchers who threw at least 1,000 total pitches in 2024 and at least 200 sliders in the zone (55 pitchers total) reveals the following fifteen most suspicious pitchers:

For reference, here are the league-wide slugging percentages on in-zone and zone5 sliders:

All pitches: .399

All sliders: .371

In-zone sliders: .474

Zone 5 sliders: .610

Five out of these 15 pitchers put their sliders in the nitro-zone more often than they should have, yet did not feel much consequence. That is, the slugging percentage of their in-zone slider was below the league average. Take Justin Steele for example, the pitcher with the second-best slider measured by Stuff+ in the list above. He threw the fourth-highest percentage of his in-zone sliders in zone 5. Those pitches were crushed for a .682 slugging percentage. However, all the other sliders he threw in the zone not in zone 5 were slugged for only .211. Using Steele as a case study might tell us that it’s ok for a pitcher to miss badly as long as his slider is good enough to perform well in the zone in aggregate. Yet this does not apply to Patrick Sandoval. His slider performed terribly in zone 5 (.850 slugging) which massively affected his overall in-zone slider slugging percentage of .522. All of his non-zone 5 in-zone sliders were slugged for .388, better than the average yet not enough to balance the scales back out for a better in-zone slider performance overall. This is not a one-size-fits-all situation.

So maybe it doesn’t matter. There is little to no relationship between the portion of in-zone sliders that land in zone 5 and opposing hitters’ slugging percentage on in-zone sliders:

The sample of data for the visual above is very small. Perhaps adding more data could create a stronger relationship. But, if we look at this differently, creating a ranking system for pitchers’ zone 5 slider portion to in-zone sliders, we do see some relationship in aggregate with the slugging percentage of in-zone sliders:

In aggregate, pitchers who throw their in-zone sliders in zone 5 less often (lower deciles) tend to have their in-zone sliders slugged less. That story was being crafted well until those pesky “Decile Rank” five and six pitchers came along:

Chris Sale adds a whole new list of questions. Why aren’t hitters getting to his in-zone sliders? Using the same qualifiers (1,000+ overall pitches, 200+ in-zone sliders) we see that Chris Sale had the second most Whiffs behind Dylan Cease on his in-zone sliders. Hitters aren’t slugging Sale’s in-zone sliders. They aren’t even making contact. Compare Sale and Logan Gilbert and now we really start scratching our heads. Gilbert has the same Zone5% as Sale with a better Stuff+ on both his slider and his fastball, yet his in-zone slider slug is much higher. Why isn’t he missing bats? Like the comparison between Steele and Sandoval, this has a lot to do with how those pitchers perform everywhere else in the zone. Sale get’s slugged at a .355 clip when in the zone and not in zone 5 compared to Gilbert’s .452.

The question remains, does it matter if a pitcher throws too many of his in-zone sliders in zone 5? What makes Steele and Sale’s sliders less likely to get smoked in the zone than Gilbert or Sandoval’s? To help answer this I conducted a simple OLS regression targeting the slugging percentage on a pitcher’s in-zone slider. I threw in some extra variables to see what might enter the model under a p-value (P>|t|) of 0.05:

In this small sample of in-zone slider performance, the ratio of zone 5 sliders to sliders thrown in the zone (zone5%), matters! It’s the only thing that matters. A pitcher’s overall slider and fastball Stuff+ come close to significance (a p-value less than 0.05), but don’t quite meet the mark. I’m fairly confident that with more data, we’d see a stronger relationship between in-zone slider slugging percentage and Stuff+.

As always, there are a lot of complicating factors in this analysis. Calling a zone 5 slider a mistake and every other pitch not a mistake is an oversimplification. Adding in more predictors, such as the performance of non-zone5 sliders may help paint a fuller picture, and measuring on a pitch level would help clear up some of the noise. Utilizing more data and analyzing by pitch level is exactly what Bland does with his “Mistake” pitch measurement. However, this small analysis tells us that the best way to drive down the opposing batter’s slugging percentage on a slider in the zone is to throw it in zone 5 less often. If you’re the type of baseball fan who makes an involuntary squeal every time your favorite pitcher leaves a slider in zone 5, you’re statistically correct, in aggregate. Squeal on.