Predicting Ulnar Collateral Ligament Injury in Rookie Major League Baseball Pitchers
Open Access
- Author:
- Rendar, Sean
- Area of Honors:
- Data Sciences
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Fenglong Ma, Thesis Supervisor
John Yen, Thesis Honors Advisor - Keywords:
- Machine Learning
Deep Learning
Injury Prediction
Ulnar Collateral Ligament
Major League Baseball - Abstract:
- The world of data-driven analytics and insight has become one of the most powerful aspects of the corporate realm over the past 20 years. The healthcare industry has followed in suit in using data science for a multitude of purposes. In addition, within the past few years, data science has expanded to the world of professional sports. Most teams, whether it be in the National Football League, National Hockey League, National Basketball Association, or Major League Baseball now employ data scientists or analysts in some capacity. Major League Baseball’s use of data analytics is often the most discussed topic. This being the use of sabermetrics, which is the analysis of baseball statistics to answer specific problems. One of the problems still to be answered is an intersection of healthcare and professional sports using machine learning for the purpose of injury prediction. That is, being able to alter athletes’ training or workload by discerning injury risks before the injuries occur. Within baseball, pitchers are regarded as the most important players and one of the most common injuries among them is the tearing of the ulnar collateral ligament. Afflicted pitchers are sidelined for the entirety or remainder of the 162-game season. By way of machine learning, modeling has been done using commonly recorded pitching statistics to predict ulnar collateral ligament tears. Injury prediction is a very complex topic in machine learning, like that of fraud detection in the sense of classification bias. The modeling conducted does show promising results, but the findings of this work serve more toward creating a precedence for further research.