STUFFpL: About
What makes some pitches better than others? We all know that more Speed is great, but is more Movement always better, more spin? How do we make a pitch better? The answers to these questions aren’t usually simple, but there are real, evidence-based answers to them. pitchLogic’s STUFFpL system makes those answers available to coaches and players at every level.
The pitchLogic STUFFpL system provides players with a STUFFpL score, which rates a pitch’s performance relative to their level of play. It also provides a SKILLpL score, which removes Speed from the equation, and can be used to track progress even when throwing at less than full effort. For both scores 100 represents a pitch that has an average performance against batters, and higher is better. Location heatmaps show where to locate a pitch for maximum effect against both right and left-handed batters. Finally, STUFFpL provides personalized recommendations to improve each pitch.
If you want to get started, dive on in! That’s all you really need to know. If you want to learn more about how STUFFpL works, keep reading.
The pitchLogic STUFFpL system is based on the concept of a pitch model. Pitch models evaluate pitches on any number of measurements and predict how they would perform on average in a game. Professional baseball organizations have been using pitch models for years to scout talent and improve their existing pitchers. STUFFpL makes this capability available to every coach and player.
STUFFpL uses a cutting-edge machine learning model which is trained on hundreds of thousands of real in-game pitches. This model can recognize complex patterns and identify combinations of qualities that make pitches successful in game. Rather than trying to fit every pitcher to a common template, STUFFpL assesses each pitcher’s unique stuff.
STUFFpL evaluates pitches primarily on Speed, Horizontal Movement, Vertical Movement, and Arm Slot. Pitches other than a pitcher’s primary fastball are also evaluated based on how well they pair with the primary fastball.
Speed and Movement describe the shape of a pitch, and they directly affect how difficult it is for a batter to hit. Arm Slot affects the way a batter perceives a pitch. For example, a batter will expect a low Arm Slot fastball to have more arm-side run. Similarly, a pitcher’s fastball affects the way a batter perceives their other pitches.
Rather than having ‘good’ or ‘bad’ values for each of these metrics, STUFFpL looks at the combination of metrics as a whole.
Although STUFFpL doesn’t consider metrics such as Spin Rate, Spin Efficiency, or seam orientation directly, these factors are extremely important in creating Horizontal and Vertical Movement. The STUFFpL system can even be used to predict how changing any of these factors would change a pitch’s performance.
While there are many ways to define a good pitch, STUFFpL is trained to recognize the statistical impact that a pitch has on the opposing team’s run scoring ability.
pitchLogic provides two different scores, called STUFFpL and SKILLpL.
STUFFpL is an overall evaluation of a pitch’s performance. The pitcher’s level of play must be selected to set an appropriate standard. A higher level of play can be selected for pitchers who wish to see how they would perform at the next level.
SKILLpL evaluates a pitch while completely ignoring speed. It can be applied to a pitch without needing to know the pitcher’s level of play and while throwing at any effort level. SKILLpL is presented in real time while throwing, as long as a pitch type is selected. Use SKILLpL to track improvement while working on pitch shape from the mound or in flat-ground sessions. If a pitcher remains consistent with their delivery and release, they should be able to maintain their SKILLpL score when working at full effort.
For a pitch with average Speed (for its type and level of play), the SKILLpL and STUFFpL scores will be equal. If STUFFpL is lower than SKILLpL, Speed is probably low for the level of play, but there are exceptions. For example, a changeup may suffer from being thrown too fast compared to the pitcher’s fastball.
For both scores, 100 represents a pitch that has an average impact on run scoring when thrown with typical command. Higher scores are better, and most pitches will fall between 50 and 150, though higher and lower scores are possible. In general, some pitch types tend to outperform others. An average slider tends to score close to 107, while the average fastball is around 92. Although most fastballs are statistically less effective than most sliders, they are generally an important aspect of pitching and help to set up other pitch types. They may also be easier for many pitchers to command.
STUFFpL reports also include location heat maps for each pitch type against right and left-handed batters. The orange locations in the heat map describe which locations are most beneficial to the pitcher, while blue locations are less effective. Heat maps are always shown from the pitcher’s point of view.
STUFFpL heat maps are created by the STUFFpL model and are personalized to your unique pitch characteristics.
Many pitches have multiple locations where they perform well. For example, a right-handed slider may be effective down and away from a right-handed batter, but it might also work up and inside (sometimes called a ‘backdoor’ slider). It can also be useful to see how much margin for error a particular location has. A typical sider may have much more room to perform well down and away, but it might require hitting the corner of the zone more precisely from the ‘backdoor’ location.
The heat map extends beyond the limits of the strike zone on all sides. The yellow box in the heat map represents the strike zone. Locations outside of the yellow box would be balls according to the rule book (although they may not always be called balls in game). In some cases, these locations may be effective if batters are likely to swing and miss.
STUFFpL also provides up to three recommendations to improve each pitch type that you throw. These recommendations are listed in order starting with the one most likely to have the largest impact on your pitch.
Almost any change you can make to a pitch will end up affecting multiple aspects of the pitch. For example, a small grip change is likely to impact your Spin Direction, Spin Efficiency, Vertical Movement, Horizontal Movement, Speed, and Spin Rate to varying degrees. Using data from millions of pitches, the STUFFpL system is designed to anticipate these interactions. The recommendations it provides are designed to have a positive overall effect after all these interactions are considered.
We recommend picking one goal at a time to focus on. Every pitcher is different, and some changes may feel more natural to you than others. That’s why we give you options. Once you’ve made a change to a pitch, run a new STUFFpL session. Your other recommendations are likely to change.
For additional tips, be sure to click on the recommendations in the app, which link to more information on each recommendation. This will help you deepen your understanding of the metrics and how to improve the performance of your pitches. Be sure to check back frequently. We are always adding new information and videos from our network of pros, coaches, and instructors.