Transforming Baseball Analytics with Artificial Intelligence

Home Transforming Baseball Analytics with Artificial Intelligence

“The Orioles approached us with a problem because they weren’t able to analyze pose positions and, subsequently, the biomechanics of their pitchers at games that may not have access to high-resolution cameras.”

Dr. John Zelek, Professor of Engineering, Co-Director, Vision and Image Processing (VIP) Lab, University of Waterloo

Transforming Baseball Analytics with Artificial Intelligence

With a goal to help baseball teams evaluate pitching talent with greater precision and accessibility, a team of researchers at the University of Waterloo launched PitcherNet, an AI-powered system that analyzes pitcher biomechanics using simple video, even from a smartphone.

Developed in collaboration with the Baltimore Orioles, the project extracts advanced metrics – such as release point, velocity, and arm extension – from basic video by combining 3D human modeling and machine learning.

The system maps a pitcher’s skeleton, then applies predictive algorithms to identify pitch types and evaluate key performance indicators. This approach enables quantitative analysis for any location, giving teams deeper insight into player potential and injury prevention.

Now, the researchers are refining the technology alongside Orioles management and exploring applications in other sports like hockey.

For more information, visit the University of Waterloo.