Sports Analytics Case Study
Applying Python-based data analysis to professional baseball for the Houston Astros.
This was a two-part sports analytics case study of the Houston Astros, a Major League Baseball organization. It involved applying statistical modeling and exploratory data analysis to professional baseball.
Bayesian Batting Performance: Applied a Beta-Binomial framework to evaluate a breakout hitter's performance, incorporating prior expectations and observed batting results to estimate the probability of reaching a season-long hit target.
Baseball Aerodynamics: Analyzed five seasons of MLB batted-ball data to investigate whether changes in baseball drag could explain increasing home run rates. Examined relationships between exit velocity, hit distance, and season using statistical summaries and visualizations.
Key Finding
Exploratory analysis showed that batted-ball distances generally increased alongside exit velocities, while the distance-to-exit-speed ratio appeared relatively stable across seasons. These results did not provide clear evidence of changing aerodynamic drag, although additional controls would be needed to distinguish ball properties from changes in hitting performance.
Skills Demonstrated
Bayesian inference, Beta-Binomial modeling, probability distributions, exploratory data analysis (EDA), statistical reasoning, trend analysis, data visualization, hypothesis evaluation
Tools & Libraries
Python, Jupyter Notebook, NumPy, matplotlib, SciPy, seaborn, Git, GitHub, statistical modeling, data visualization