Change Piece: Finding the Ideal FB/CH Combination
There appear to be many ways to be successful with a changeup, and perhaps as important as the quality of the changeup itself is how it “interacts” with other pitches in the repertoire, especially the fastball. We have known for decades that the ideal changeup has a large velocity difference from the pitcher’s fastball. But, there are also successful changeups with significant movement, typically fading armside and with downward depth. In a perfect world, everybody’s changeup would be 10–15mph slower and with much less vertical rise and much more horizontal run than the fastball. But most players do not possess a Pedro Martinez-esque killer changeup, so how should they think about pairing their fastball and changeup to achieve the best results they can with what they have?
The way I see it, these are the possible combinations of a fastball and a changeup (very simplified):
If you want your pitches to move differently, then they are going to need to be thrown differently. This can be quantified by spin axis, which is made publicly available by BrooksBaseball.net. We will use spin axis to break up fastballs and changeups into subcategories.
Characteristics of a “Straight” Pitch
Typically, pitches that are “straight” have a spin axis of between 180 and 225 degrees (RHP) which corresponds to 12:00 to 1:30 on a clock. Here’s what that looks like.
Players can, and often do, throw their changeups with a similar axis. These are the “straight” changeups. However, you would be hard pressed to find one of these on a YouTube compilation of Nastiest Pitches That Will Have You Saying WOW 2019. They are straight and look like slow fastballs, which is not exactly pleasing to the eye.
Straight pitches almost all have a “vertical” axis, and the higher their spin rate, the more vertical rise they will have.
Characteristics of a “Running” Pitch
On the other hand, running pitches typically have a spin axis between about 225 and 270 degrees (RHP) or between 1:30 and 3:00 on a clock. Fastballs in this category are usually gripped like two-seamers and may be called sinkers. Take a look at their more tilted spin axis.
That pitch is spinning at almost 270 degrees (complete sidespin), which is very extreme. We see the movement is extreme as well. Pitches with this direction of spin are more commonly changeups, like this one:
Running pitches with a more horizontal spin axis and higher spin rate with move more horizontally than pitches with a horizontal spin axis and low spin.
Useful Combinations
So if we agree that our ideal changeup has a large movement difference in addition to a large velocity difference from the fastball, we are left with two FB/CH combinations: Straight FB/Running CH or Running FB/Straight CH.
I can say that I have personally never seen or heard of a pitcher with a running fastball and a straight changeup. Though a player in this category may exist, the vast majority of pitchers are in the other three categories. Here are a few examples:
We see that the bottom left box, the one of the “ideal” FB/CH combination, is not the only one with successful pitchers! There are many ways to be successful with a changeup. Let’s see what these three FB/CH interactions look like with some PitchingNinja overlays.
Straight FB/Running CH:
Here we see how having a different type of fastball and changeup can prove advantageous. The pitches are different in three dimensions: horizontally, vertically, and depth-wise as a result of the velocity difference.
Straight FB/ Straight CH:
These pitches have far less movement differentiation and rely almost entirely upon velocity difference. Conventional wisdom is that a hitter could simply sit fastball and adjust to the changeup more easily with this combination because both will end up in around the same location. On the other hand, pitchers with this FB/CH combination often induce weak contact with the changeup, catching the batter out in front.
Running FB/Running CH
This gif is a great illustration of how much this FB/CH combination relies on velocity difference to fool hitters, as the movement profiles of these pitches are very similar (though it may not look like it due to the pitches being thrown to different locations).
Categorizing Pitchers
To analyze the effectiveness of each possible combination, I put MLB pitchers from the last 5 seasons through a “flowchart” that placed each pitcher into one of 16 possible combinations. The flowchart necessitates only a few pieces of information:
What is your average fastball spin axis?
What is your average fastball adjusted spin rate (Bauer Units)?
What is your average changeup spin axis?
What is your average changeup adjusted spin rate (Bauer Units)?
Using this information, we can figure out which one of the 16 possible FB/CH combinations a pitcher has:
So now that we have the groups defined (and we can easily find which group a pitcher belongs to), let’s see which groups have had the best results by wOBA Allowed and Whiff Rate.
But First, Sample Sizes:
Remember when I said I couldn’t recall a single pitcher with a running fastball and a straight changeup? That’s because there aren’t any. This combination (represented in the upper right quadrant) has not been utilized by a single pitcher in 2019. The combination of an upright, low-spin fastball paired with a running low-spin changeup is by far the most popular combination among MLB pitchers in 2019.
wOBA Against
It should not come as much of a surprise that upright, high-spin fastballs are the most effective type of fastball, regardless of what type of changeup it is paired with. This is because changeups typically “play off of the fastball” more than fastballs “play off of the changeup.” In other words, changeup type should not influence fastball success as much as the other way around.
Here we see that fastball type does have a major influence on changeup type (because the wOBA Against for a given changeup type varies a lot by which type of fastball it is paired with). The most effective FB/CH combination for allowing a low wOBA on changeups has been an upright, low-spin fastball and a running, high-spin changeup.
We also see from this table that the combinations on the diagonal (those where the fastball and changeup are of the same type) are the worst for allowing a low wOBA. This finding backs up the long-held wisdom that a fastball and a changeup should have some “separation” which for us means differentiation of either spin axis or spin rate. So much for “tunneling”?
Whiff Rate
The FB/CH combinations that have induced the highest whiff rate on fastballs have been, no surprise, the upright high-spin fastballs. In particular, the upright high-spin fastball paired with an upright, low-spin changeup has been the most effective. This fits with our preconceived notion that upright fastballs (Four-Seams) are better at getting whiffs than running fastballs (Two-Seams) that are better at inducing ground balls. Notice how just about any running fastball is likely to coax an unspectacular amount of whiffs.
In looking at whiff rates on changeups, we see a bit of a trade-off. The changeup paired with a high-whiff fastball type is not very likely to get whiffs itself. That is with the exception of the straight high-spin fastball and the running low-spin changeup, which was pretty good at inducing whiffs both from the fastball and the changeup.
The best way to get whiffs with a changeup would be to pair a straight low-spin fastball with a running high-spin changeup. This was also the best combination for allowing a low wOBA. Although maybe a bit counterintuitive, we may have found our winner for Best FB/CH Combination, at least according to these two measures of pitch quality.
Players in this group include Luis Castillo, Mike Foltynewicz, Carlson Martinez, Trevor Richards, Jharel Cotton, Zack Wheeler, Chase Anderson
Switching Groups
Having all of this knowledge is certainly helpful, and in years past would have been enough for teams and analysts. But in 2019, we must acknowledge that a pitcher whose FB/CH combination is suboptimal is not doomed. Through informed pitch design and concentrated effort, I see no reason why a pitcher would not be able to alter his pitches so as to switch to a more optimal FB/CH combination.
Remember that the two attributes of the pitches used in this analysis are spin axis and adjusted spin rate. The bad news is that nobody currently knows how to (legally) improve a pitcher’s adjusted spin rate. There are ways to decrease adjusted spin rate like altering the pitch grip, but even this has not been studied much in the public sphere. However, the good news is that a pitch’s spin axis is very pliable.
The spin axis of a fastball or changeup is a function of hand and wrist position at release, which is certainly in control of the pitcher. To go from a “straight” pitch to a “running” pitch, a pitcher can tilt his hand to be closer to parallel with the ground at release, pronate his wrist more at release, switch from a two-seam to a four-seam grip, and/or “drop down” his arm slot to get more sidespin on the ball. To go from a “running” pitch to a “straight” pitch, a pitcher can make his wrist position for vertical at release, pronate his wrist less, switch from a four-seam to a two-seam grip, or throw from a more over the top arm slot.
Experimentation with any of these changes could help a pitcher eventually find himself in a more optimal FB/CH combination group.
Conclusions
There are few pitch interactions that are as fundamental to pitching as the fastball/changeup dynamic. Even the name of the offspeed pitch is a direct reference to its relationship with the fastball. It is widely accepted that a changeup should have a large velocity difference from the fastball. But when it comes to the pitches’ other characteristics, we can now say that a pitcher should aim to throw their fastball and changeup with a different spin axis, different adjusted spin rate, or both. This increases the likelihood that a hitter will be fooled in more ways than one, creating a greater chance of success for the pitcher.
In using this method to analyze a pitcher and the effectiveness of his FB/CH interaction, we should find out which combination group he is in. The wOBA Allowed and Whiff Rate tables should be used to see how MLB pitchers with the same FB/CH combination have done in 2019. Depending on the goals of the pitcher, steps should be taken to adjust his fastball or changeup profile to put him in a more desirable combination group.
Members of Each Group!
