When Zack Wheeler signed with the Phillies, he mentioned that the Phillies’ analytics department factored into his decision. He had seen what Gerrit Cole learned in Houston and wanted a club to do that for him. Well, there’s only one Gerrit Cole. No one expected Wheeler to set strikeout records. But Wheeler did change. In fact, he shifted in the opposite direction from Cole. Instead of becoming more like Nolan Ryan, he became more like Jamie Moyer. In an age of ever increasing true outcomes, switching bodies with Jamie Moyer seems like a way to demolish a career, although it might help one empathize across generational divisions and make a wonderful movie direct to Disney+. Yet Wheeler thrived on his new found Moyerism. How did he pull it off? What turned him into the hardest throwing contact pitcher this side of the Pacific Ocean?
We can find many differences between Wheelers past and present. But I think one difference is at the heart of them all: Wheeler changed how his pitches move. For this post I’ll focus on the four-seamer and comment on the changes to the rest of his arsenal at the end. While every pitch moved differently this year, the four-seamer is the most interesting of them.
Let’s begin with two pictures, courtesy of Baseball Savant. The first shows the average trajectory of each of his pitches to a left-handed hitter in 2019. The second shows the same in 2020.
As I describe the changes to Wheeler’s four-seamer, it will be helpful to refer back to these pictures.
In years past Wheeler threw a standard, above average four-seamer. Typically, four-seamers have both rise and run and the more of each the better (up to the limits of the pitcher’s control anyway). The rise is generated by backspin as the ball rolls out from under the pitchers index and middle finger. Of course, the ball does not actually move higher from the ground than it was at release. Instead, the backspin resists the effects of gravity enough to keep the ball higher than the batter expects, making the pitch appear to rise from the batter’s perspective. The run, i.e., arm-side break, is caused by sidespin generated by the angle at which the pitcher releases the ball. Together, the rise and run work to keep the batter from squaring up the pitch. Ideally, either he swings under it or it slides away from the barrel.
Wheeler’s four-seamer used to display both of these traits at an above average level, although neither were especially noteworthy. His rise registered at 0-10% above average, whereas the best rise registered in the 20s. His run showed better, coming in at 30-45% above average. But, then again, Chris Sale sits around 80%. Despite not having a standout four-seamer, Wheeler used it well. In 2019 his four-seamer drew whiffs as often as teammate Jacob DeGrom’s and put a higher rate of batters away. All told, Wheeler’s four-seamer before 2020 was a typical and useful, albeit not noteworthy, weapon.
In 2020 his four-seamer became weird. Indeed, he straightened it out. No longer did it resist gravity’s call but it heeded it. Obviously, the ball is still spinning similarly. In fact, his spin rate dropped barely at all. So, the ball is still moving some relative to a normal parabolic path. But it does so less than league average along both the vertical and horizontal dimensions.
Usually, a straighter four-seamer is a worse pitch. And as you might expect, the straighter fastball induced whiffs and put away hitters less often. Both rates dropped by about 7 points. Nevertheless, the pitch was just as valuable a part of his arsenal, as measured by Baseball Savant’s run value. (In fact, it was ever so slightly more valuable but I don’t know whether that difference is within the margin of measurement error.) That value is, I think, located in one big change in how batters hit his fastball: on the ground. In 2019, when batters put Wheeler fastball in play, they tended to hit the ball in the air. The average launch angle was 20 degrees. This season, the average Wheeler fastball off an opponent’s bat was a grounder. At 6 degrees, his launch angle suggests an increase in linedrives as well, which is reflected in the data. His fastball did become a bit easier to hit. But the increase in groundballs allowed Wheeler to erase those mishaps with double plays. And, most importantly, Wheeler surrendered just one homer on his fastball. Although his straighter fastball became vulnerable to contact, the contact it became vulnerable to was, on the whole, less damaging.
So, if the spin rate is basically the same, why has the pitch straightened out? And why did it remain effective? We can start to answer the first question with the pitch’s active spin. Active spin estimates the amount of spin that moves the ball off its normal parabolic path to the plate. Not all movement is caused by active spin, but it is a prominent factor. It is also a complex phenomenon that can be difficult to understand. The ball travels along one axis, which constantly changes from one instant to the next. At the same time it spins along another axis that can shift relative to the axis of travel as the ball moves but generally holds constant relative to the ground. How do we know what spin or how much of it bends the ball’s path? Here is the basic principle: only the component of spin perpendicular to the axis of travel contributes to movement.
You might, if you ponder it for a stretch, remember that word ‘component’ from a high school physics course, when the teacher wanted you to diagram a ball rolling down a hill and you were wondering, among other things, if you could get away with eating the granola bar in your backpack. Obviously, a ball only spins in one direction, along one axis. That axis can be perfectly aligned with the axis of travel, perfectly perpendicular, and anything in between. If it is perfectly aligned then it is spinning like a well thrown football or a rifled bullet. That spin doesn’t bend the ball’s path at all. If it is perfectly perpendicular, like a well drawn cueball, all of the spin bends the path. If it is somewhere in between, then some of the spin is inert, like the football spin, and some is active, like the drawn cue. Hence, active spin.
I realize all of that was dense preamble. But I needed to get it on board so that we can ask: why did Zack Wheeler’s active spin on his four-seamer decrease from 91% in 2019 down to 78% in 2020? Given what we know about active spin, it must be that he altered the ball’s axis of spin relative to the ball’s axis of travel. Without interviewing him or knowledgable Phillies staff, we can’t know exactly how he changed his mechanics. We can only reason backwards so far.
But let’s try! Imagine a fastball as it leaves your hand. Unless you are a pitching machine, the ball spins around a titled axis, relative to your upright torso. If you’re like me, you’re probably imagining the spin axis at a right angle to the axis of travel. Ideally, it would be, but the human body is not like Mr. Fantastic’s. It is a jumble of flesh and bone that can get in each other’s way. In order to get that angle just right a pitcher has to line up his legs, hips, shoulders, arm, and palm so that they are all sending the ball on one line at the moment of release. That’s extremely difficult. Again, we can’t all be Gerrit Cole.
For the mechanics of active spin, the most important variable is the palm. It orients the ball’s spin to the direction of release. When you imagine rotating the axis of spin away from 90 degrees, you’re imagining a slight turn of the palm at the moment of release. That turn could be created by a rotation of the forearm or a tilt of the wrist. Most likely, one of these two adjustments is at the root of Wheeler’s diminished active spin and, thereby, straighter four-seamer.
But why would he make this adjustment? I can think of a couple reasons, both of which have some evidence behind them. First, he might have found the straighter four-seamer easier to command. Here are strike zone heat maps for his 2019 and 2020 four-seamers:
In 2019 Wheeler focussed his four-seamer in a stout box over the heart of the zone. That box leaked inside to right-handed hitters, as we would expect for a pitch with tail. But he did not go inside to lefties often. In 2020, on the other hand, the box spreads and separates a bit. Wheeler attacked down and in to righties and up and in to lefties. He also threw his four-seamer about 40% more often. To me this suggests that Wheeler had better command of his four-seamer and knew it. That better command turned a pitch typically hit in the air into one that reliably produced groundballs. So, perhaps, he made the mechanical adjustments in order to better locate his fastball. After all, a pitch that moves less is easier to target.
I think that explanation is less likely, however, than this second one. Check out the vertical release point for Wheeler’s pitches over his career.
While Wheeler’s horizontal release is consistent for all his pitches, for most of his career, the vertical release point for his four-seamer stands out from his other pitches. And in general his release points on all his pitches move around a lot relative to each other. In 2020, however, his vertical release points are clustered together. They do not overlap completely. And his vertical release moved down as the season progressed. But with every alteration the pitches move in lockstep.
This synchronization of his release would allow Wheeler to tunnel his pitches better. Tunneling pitches keeps them on the same path for as long as possible on their trip towards home plate so that the hitter has less time to react to which pitch is coming.
Here are the same pitch trajectories as shown earlier but from a new perspective. Also, I’ve added grey dots, which represent the point at which the hitter can recognize the pitch. Using them, I think we can see improvement in tunneling. In 2019 four of his pitches appear to tunnel better. But the curveball stands far apart from them, starting with its high and rearward release point. Moreover, all the release points in 2019 were spread further out. So, even though they reach the recognition point clustered, there was an immediate clue to the hitter which pitch Wheeler was throwing. In 2020, as compared to 2019, the beginnings of those paths to the recognition points are slightly closer together and synchronized on release. Even the curveball comes out of the same place. Thus, it takes longer for the hitter to distinguish the curve in particular and all of Wheeler’s arsenal generally.
I’m also not convinced that Wheeler actually tunneled those four non-curve pitches better in 2019. Wheeler tended to locate those four pitches in similar parts of the zone. That will tend to keep the average trajectories close together. In 2020, Wheeler moved his pitches around the zone more and, so, of course, the average trajectories don’t adhere as much.
Hopefully, I’ve convinced you that Wheeler improved his tunneling in 2020. Why is that good? Well, limiting hitter reaction time increases the likelihood of whiffs and weak contact. Hitters swing at more pitcher’s pitches and fewer of the pitches they’re hunting. And even though, as I mentioned in a previous article, Wheeler faced much better offenses than the average pitcher did this year, he set career records in chase rate, chase contact, and barrel rate, while hewing very close to his career average for whiff rate. Wheeler did not dominate batters the way we expect a pitcher with a 97-mph fastball to do. But he did a better job rendering them harmless.
Unsurprisingly, Wheeler’s four-seamer is not the only pitch whose mechanics he changed. Both his slider and curve showed significantly more movement, including adding considerable active spin to the curve. Combining his improved tunneling with this additional movement, Wheeler managed to disguise his pitches on the way in and spread them out upon arrival. Not only was it harder to tell what he was throwing, it was also harder to adjust if you guessed wrong. As a result, the slider and curve became better secondary weapons than he’s ever had before. Even though he began the season barely striking anyone out, as the season came to a close the Ks had started to mount.
In 2020, Zack Wheeler was not necessarily a better pitcher than he’d been in previous years, but he was a different pitcher. And by being at least as good yet different, he gives us some hope. Whereas the Zack Wheeler of 2020 channeled Jamie Moyer, perhaps the Zack Wheeler of 2021 will be able to fuse Moyer and Nolan Ryan into one body. That would be terrific.