In the coming weeks there will be no shortage of speculation about which players the Padres might add to further their hopes of surviving the second half of the season and earning a playoff berth. The team’s structural weaknesses are well articulated. But of equal intrigue is what they might expect of certain players already slated to play important roles down the stretch. One of the most nebulous performances of the first half was that of Dylan Cease. And thus far there haven’t been any clear explanations for why the pitcher that finished 4th in NL Cy Young voting in 2024 currently has the 6th worst ERA of any starter in baseball. Superficially, he seems to be the same pitcher in 2025 sporting almost identical strikeout and walk rates with no noticeable decline in pitch velocity. In fact, the average velocity of his four-seam fastball and slider are up year-over-year. But when you look a little deeper there are some noticeable differences in this year’s version of Dylan Cease, and it’s fair to wonder if these differences may have some explanatory power.
The Other TTO
One of the obvious struggles for Cease in 2025 has been his disastrous results the second time facing the opposing team’s order in the same game:
In 2013, Michael G. Lichtman (MGL) wrote ‘the Holy Grail of research articles’ on the idea of the times-through-the-order (TTO1) penalty for Baseball Prospectus. The TTO penalty is the observed phenomenon that pitchers tend to become less effective each successive time they cycle through the order in a game. This tendency was already well documented, but MGL made a few granular observations that got to the underlying physiology of this phenomenon. In particular, he noted that the TTOP didn’t seem to be a factor of fatigue, but rather familiarity, that of the hitter becoming successively more familiar with a pitcher’s offerings. One of the more powerful findings was that the number of pitches the hitter saw the first time he faced the pitcher had strong correlation with the degree of jump in wOBA the hitter achieved in successive times through the order, with a much higher bonus accruing to hitters who saw a high pitch count their first at bat:
From the article:
If a batter has seen more than four pitches in his first PA, he hits 25 points better the second time around…
So it’s interesting to note that in 2025 Cease is averaging 4.12 pitches per plate appearance in the 1st-3rd innings in which he is typically facing the opposing order the first time through. But this is almost identical to the 4.13 pitches per plate appearance he averaged in 2024. So what else could have changed?
Pitch Mix
Cease has always been a predominantly two pitch pitcher with the four-seam fastball and slider making up the bulk of his offerings, and the knuckle curve and changeup finishing a distant second and third in the hierarchy. And 2024, when he was quite effective, was no exception:
And at first glance it might seem that not much has changed in 2025:
He’s still a two pitch pitcher. But importantly, every other pitch has been relegated to well below 10% usage. In 2025 Cease has been throwing his Slider and Four-seam Fastball an astonishing 88.2% of the time. And it turns out that the extreme to which he’s shifted his pitch mix has crossed an important threshold.
MGL wrote a follow up to his blockbuster article looking at the effect a starting pitcher’s repertoire had on the TTO penalty he would face each successive turn through the order:
The penalty for successive times through the order is much harsher the fewer pitches you have in your arsenal2.
As the article notes:
It appears that the fewer pitches you have in your repertoire, the more that batters become quickly familiar with you.
In 2021 Fangraphs published an article from Carmen Ciardiello replicating the finding that relying more heavily on two-pitches correlates with more extreme TTO penalties, with the relationship becoming almost perfectly linear by the third time through the order:
So to summarize, in 2025 we can see that Cease has continued one trend that leads to harsher TTO penalties: Averaging a high number of pitches per hitter the first time through the order3. But in 2025 Cease added a second surge to his TTO penalty by falling into a very extreme two-pitch mix, relying almost exclusively on his slider and four-seam fastball. Combining a high pitch count the first time through the order with an extreme reliance on only two pitches almost certainly compounds the TTO penalty a pitcher is likely to incur.
Keep Them Guessing
Last Thursday the Yankees once again showed they are amongst the best in the league at recognizing pitch signs/tells and relaying this to the batter. In an excruciating sequence Trent Grisham and Cody Bellinger relayed every slider Andres Munoz threw during a blown save:
Jomboy Media did an excellent breakdown. Once a hitter knows what pitch is coming he becomes orders of magnitude more effective with his next swing. It’s always wise to consider whether a pitcher with very good stuff, having a bizarrely bad string of outings, is the victim of relayed signs or a pre-pitch tell. Thus far there hasn’t been any consistent evidence that that opposing teams have a pre-pitch tell when facing Cease, at least none we can discern. But there are other ways a hitter can be tipped off. When a pitcher changes arm angle it can be a sign that a different pitch is coming. And it’s interesting to look at the year-over-year changes in Cease’s delivery:
He’s seen a dip in average arm angle to the lowest since he was a rookie. And we can see where this is coming from when we look deeper at each of his pitches including their arm angles and shapes:
There’s a lot here. He’s dropped his arm angle on the slider by 4 degrees and upped its usage significantly. And while he has also lowered his arm angle on the four-seam fastball, it’s a smaller 2 degree change. A change in arm angle alone doesn’t necessarily have a positive or negative valence. But if we look at year-to-year differences in arm angle between the slider and four-seamer, along with the year-to-year differences in pitch shape of his top two offerings, we start to see some concerning findings:
In 2025 he’s increased the average speed of his slider from 87.7 MPH to 89 MPH while keeping the four-seam fastball speed almost unchanged at ~97 MPH. This has the effect of reducing the difference in velocity between the two pitches from 9.2 MPH to 8 MPH. He’s also seen a decrease in the vertical drop of the slider from 33.6 inches to 31.3 inches meaning the difference in pitch shape in the vertical plane has narrowed from 22.8 inches in 2024 to 20.6 inches in 2025. He hasn’t seen much difference in the horizontal plane with the pitches being about 4.9 inches apart, but it’s notable that he shaved almost an inch off of the slider’s horizontal movement and it barely had any to begin with. It’s almost a one-plane break. And finally, in 2024 he threw the slider from a 52 degree arm angle, and the four-seamer from a 55 degree arm angle, a delta of only 3 degrees. In 2025 he’s dropped the slider arm angle all the way down to 48 degrees while the four-seamer arm angle only dropped to 53 degrees, a wider delta of 5 degrees. To summarize: In 2025 his slider and four-seamer have a smaller difference in speed, less separation on the vertical plane, and are being thrown with a greater difference in arm angle which may be more noticeable for hitters.
There are a few implications: For a hitter who is up there simply guessing which pitch is coming, the diminished difference between the two pitch shapes may incrementally diminish the penalty for guessing wrong; the more similar the two pitches are the less change in bat trajectory is required for a defensive swing from a hitter than was fooled. More importantly, for a hitter who is trying to recognize and react to the pitch, a wider difference in arm angle is likely to be detectable more often. A two-pitch guessing game gets that much easier if there is any discernable difference in delivery.
Compounding
In essence, from 2024 to 2025 Cease’s slider and four-seamer usage have ballooned to such extremes that he’s crossed a threshold that suggests that he is much more susceptible to the TTO penalty than he may have been even one year ago. And in moving to a two-pitch mix, he’s also seen a decrease in the vertical plane and velocity separation between the pitches, and a widening difference in the arm angles used to throw the two pitches which may make pitch recognition more easy for a hitter. It’s not a huge stretch to think these factors could compound on one another to play a part in Cease’s down season. Of course some of his down season likely is due to bad luck as well. But it’s increasingly doubtful that bad luck is the entire explanation.
These are changes that the team has undoubtedly noticed. And it will be interesting to see if across the second half of the season Cease moves towards a broader pitch mix, or works on tightening up the similarity in the arm action used for the slider and four-seamer.
Cease will open the second half of the season Friday on the road against the Washington Nationals. Last season Cease pitched a no-hitter in Washington as part of a very good second half in which he maintained a 2.66 ERA across his final 13 starts. It can’t be overstated what such a return to form would mean for the Padres’ second half survival.
Not to be confused with the 2023 jump-scare TTO: Three True Outcomes
Incidentally this is most pronounced the second time through the order
This has been his tendency throughout his career, and is the byproduct of being a very high strikeout pitcher. As such it is unlikely to change.
Banger of an article btw. If you could just CC Cease and Niebla on it that would be great.
Have to think he’s lost trust or command in the other pitches when the 2 pitch focus clearly looks to be a reason he’s been bad this year. Anything to suggest in the limited number of knuckle curves that he’s thrown that it’s a worse pitch than it’s been in prior years?