DETERMINING RELATIONSHIPS BETWEEN KINEMATIC SEQUENCING AND BASEBALL PITCH VELOCITY USING MARKERLESS MOTION CAPTURE

Authors

Ryan W. G. Bench, Brock UniversityFollow
Colin D. McKinnon, 3MotionAI
Anthony Brady, Driveline Baseball
Casey Mulholland, KineticPro Baseball
Michael Sonne, Brock UniversityFollow
Mike Holmes, Brock UniversityFollow

Category

Athletics

Document Type

Paper

Abstract

The purpose of this study was to determine how the timings and magnitudes of peak pelvis rotational velocity, peak trunk rotational velocity, peak elbow extension velocity, and peak shoulder internal rotation velocity affect pitch velocity. Eighty pitchers (187.2 ± 8.2cm, 89.3 ± 13.0kg, 20.1 ± 3.3yrs) had a minimum of 3 fastballs recorded and video was processed using pitchAI^TM. Average pitch velocity was 38.1 ± 2.5 m/s. A multilinear regression generated a significant prediction for pitch velocity (R² = 0.368 and p < 0.01). Pitcher weight (β = 0.535, p < 0.001), peak pelvis rotational velocity timing (β = -0.157, p = 0.001), peak elbow extension timing (β = 0.122, p = 0.006), and peak shoulder internal rotation timing (β = -0.113, p = 0.018), were significant contributors to the multilinear model. In conclusion, player weight and their kinematic sequence metrics from pitchAI^TM can be significant predictors of pitch velocity.

Recommended Citation

Bench, Ryan W. G.; McKinnon, Colin D.; Brady, Anthony; Mulholland, Casey; Sonne, Michael; and Holmes, Mike (2023) "DETERMINING RELATIONSHIPS BETWEEN KINEMATIC SEQUENCING AND BASEBALL PITCH VELOCITY USING MARKERLESS MOTION CAPTURE," ISBS Proceedings Archive: Vol. 41: Iss. 1, Article 10.
Available at: https://commons.nmu.edu/isbs/vol41/iss1/10