Inflection Points and Pitch Design
--
Pitch design is essential at the higher levels of baseball to maximize a pitcher’s ability. Adding movement to pitches is great but not all added movement is equal. Knowing how to balance movement (both horizontal and vertical) and velocity is key to unlocking untapped potential. This led me to research inflection points in pitch performance to help understand what makes each pitch successful.
Methodology
I took the 2019 MLB Statcast Data and adjusted the horizontal movement to become the absolute horizontal movement so RHP and LHP data could be analyzed simultaneously. Doing this also helps neutralize the effect of the batter, since horizontal movement matters more for same handedness matchups (especially with breaking pitches). I then removed extreme outliers within each pitch to create a more representative model of how each individual pitch actually performed.
The two metrics that I decided to judge pitch performance on were Whiff Rate and wOBA Against. I focused on these two because strikeouts are the most valuable forms of outs for pitchers and wOBA Against is great at measuring the quality of contact.
Each pitch was grouped by horizontal movement, vertical movement, and velocity (all rounded to the nearest whole number) and then the average Whiff Rate and wOBA Against were found for each variable. Graphs were then generated for the three variables, including heatmaps for the movement numbers. Since the graphs only look at one variable at a time, it was important to include the heatmaps to show how each movement number can play off of the other.
These graphs and heatmaps were then scaled to find the performance against the average for each pitch. This is important to do because each pitch type is unique. Each pitch should be judged against itself, rather than the complete subset of pitches as each pitch type is unique. For example, sliders will naturally have a higher Whiff Rate than fastballs.
Results and Takeaways
While not every pitch has a clear inflection point, there are certainly good examples of this concept.
Cutters
The inflection point for cutters would be at the 4-inch mark for horizontal movement and the 5-inch mark for vertical movement. Every extra bit of movement past 4 inches horizontally or below 5 inches vertically makes for a big difference in performance compared to the average cutter. Whiff Rates go up quite a bit, along with a decreasing wOBA Against. Knowing these thresholds is important as it can provide a clear goal for the type of movement desired for maximum results, especially if there is not a clear relationship between velocity and pitch performance as shown below. If a cutter is within the 84–90 mph range it appears that the movement of pitch is much more significant in producing outs than the velocity at which it is thrown.
Curveballs
Curveballs are a different case as there is a clear rise in performance as velocity increases. Velocity should then play a significant role in the development of a curveball. Vertical movement should also be sought after. If a curveball can get past 17 inches of depth, the likelihood of swings and misses goes up, along with creating weaker contact. The horizontal movement of the pitch appears to play less of a role, which makes sense as sliders are known as the breaking pitch that has more horizontal movement while curveballs traditionally focus on getting depth.
A great example of the importance of velocity and vertical break on the curveball would be Tyler Glasnow. He averaged over 83 mph on the pitch with about 17 inches of vertical break over the course of 2019, both past the inflection points for curveballs. This helped him have a wOBA Against of .164 and a staggering 43.9% Whiff Rate with this pitch.
Usage in Pitch Design
These types of graphics can be utilized with Trackman data to see how a pitcher’s stuff lines up with how it performed in the Majors.
For example, player X’s fastball compares well, movement-wise, to other successful MLB fastballs, meaning that this player should focus on throwing this pitch harder (without losing movement) to further develop success with it.
This is a simplistic view of a complex process as I only focused on three different variables, but the main point still stands. Pitch design is all about creating a better pitch and knowing the thresholds to success can help pitchers set specific goals for how their pitches should move to maximize their success.
Future Research
- Integrate these ideas and charts with my MLB comp model
- Constrain the data into more subsections and see how that changes the output
- Look at resulting metrics other than Whiff Rate and wOBA Against to develop a more complete view of the performance of a pitch
Special thanks to @T_Zombro24 and @Steve_Nagy7 for help with these ideas
Please reach out to me on Twitter @k_camden for any further discussion or feedback
