Skip to content

A machine learning project focused on analyzing optical tracking data from college baseball games, with visualizations and classification of pitches as balls or strikes.

Notifications You must be signed in to change notification settings

indexjoseph/BaseballOpticsAnalysis

Repository files navigation

BaseballOpticsAnalysis

Project Overview

BaseballOpticsAnalysis is a machine learning project analyzing pitch-level optical tracking data from college baseball games. Using a dataset of 3,344 observations from TrackMan V3, this project visualizes pitches within the strike zone and classifies them as "balls" or "strikes." The objective is to leverage machine learning techniques to gain insights into pitch placement and classification accuracy.

Dataset

The Track_Combo.csv dataset includes pitch tracking data from 21 college baseball scrimmage games. Each observation provides detailed information on pitch type, count, and trajectory, recorded by TrackMan V3. Player and team IDs have been anonymized. For further details, see the TrackMan glossary.

Team Members

  • Andrew Lang
  • Joseph Oladeji
  • Richard Prange

Course Information

Final project for MATH-475, Statistical Machine Learning, under Professor Soheyl Anbouhi

Running the plot cell

This cell will take a while as it has to manually create and graph 2200 plots. Will probably take about 5 minutes. The create images from the execution of this cell will be used for the classification. s

About

A machine learning project focused on analyzing optical tracking data from college baseball games, with visualizations and classification of pitches as balls or strikes.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •