Student Work

Baseball Pitching Biomechanics and Injury Prevention


Downloadable Content

open in viewer

In the past decade, baseball pitchers have been incentivized to throw harder, and this has come with a dramatic increase in elbow-related injuries, specifically regarding the UCL. There has been a strong rise in Tommy-John surgeries, from 27 in 2000 to 166 in 2015 in the MLB plus MiLB. With this throw harder mentality spreading down to amateur and youth leagues, teams are becoming more invested in trying to prevent these injuries in the first place. The use of motion capture has been on the rise, and while it aids in helping pitchers throw harder, it does not have the thorough capability to measure injury risk and prevent injuries. The goal of the project is to develop a wearable sensor-based system that measures metrics of interest (i.e., linear acceleration, angular velocity) to estimate elbow injury risk that occurs as a result of baseball pitching, with respect to pitch types. The design incorporates inertial motion sensors attached to the forearm that have Bluetooth transmitting capability, for real-time feedback, that sends data to a MATLAB script that takes the measurements and estimates injury risk, based on the pitcher's physical attributes. Literature suggested that the UCL experienced an upper limit of 60 Nm of torque per pitch. The sensors were verified when attached to a wheel moving at a fixed velocity of 5 mph (2.23 m/s), where the angular velocity was expected to be 4.68 degrees per second, and the IMU calculated 4.63 degrees per second. Once it was verified that the sensors were collecting accurate data it then validated the data analysis system in MATLAB. Preliminary data was then collected from the team throwing pitches to detect any issues that could arise. These data were used to calculate the force and torque at the elbow when the pitch was thrown. Both calculations were the same and were within the expected range of force and torque. The sensors were then validated with a pitcher human subject who wore the sensors and threw several pitches to develop a healthy baseline of force and torque experienced. Then, the pitcher was fatigued and threw again in order to create a fatigued baseline. The force and torque collected were analyzed to determine what the expected “drop” was to be between each baseline. If there is more fatigue there would show a greater amount of torque, which could lead to a higher risk of injury. The average fastball torque increased from 42.53 Nm to 45.41 Nm with fatigue and the average curveball torque decreased from 49.29 Nm to 47.37 Nm with fatigue. While the fastball results were expected, the curveball results were not. While the results were not expected, the limitations can be addressed to optimize the system overall. The project was able to demonstrate a proof-of-concept that a real-time feedback capability is feasible and that different pitch types could be accounted for.

  • This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.
  • E-project-042622-192112
  • 63981
  • 2022
UN Sustainable Development Goals
Date created
  • 2022-04-26
Resource type
Rights statement


In Collection:



Permanent link to this page: