Below are a variety of data analyses I have performed across several leagues using R and Python:
What Variables Determine Success in the Modern NHL?Stanley Cup Winner Predictive Modeling Leveraging 16 seasons worth of NHL data from Natural Stat Trick, I designed multiple predictive models using Python to determine at what accuracy rate I could: 1) Predict the Stanley Cup Winner using regular season data (using 15 seasons worth of training data) 2) Predict the Stanley Cup Winner of the following season using the trained data
The analysis required extensive variable development, outlier and noise consideration given the size of the dataset, such as movement of players post-season and trade deadline acquisitions. Examples include returning goalies/Goalie Stability (high indicator of success), top-6 forwards, and weighting variables based on impact.
Sabermetrics Analysis – OBP vs SLG: I performed a high-level Sabermetrics analysis using R on the relationship between On-Base Percentage (OBP) vs Slugging Percentage (SLG) on Red Sox hitters from the 2024 – 2025 season. Although these 2 features are extremely closely related, there are insights that can be gleaned from potentially lower OBP but higher SLG hitters (as was analyzed).
Despite a player potentially having a lower OBP than other teammates, their quality as a hitter as per their SLG could determine they are actually more valuable in certain scenarios.
Sabermetrics Analysis – Player Offensive Value Statistics This analysis performs a high-level sabermetric evaluation of Boston Red Sox hitters using 2024–2025 data from Baseball Reference, focusing on key offensive metrics and player value. Using R, it examines the relationship between On-Base Percentage (OBP), Slugging Percentage (SLG), Home Runs (HR), and Wins Above Replacement (WAR) through correlation and regression techniques.
Sabermetrics Analysis – Player Position and Offensive Statistics This analysis conducts a high-level sabermetric examination of MLB hitters using 2023–2024 season data from Baseball Reference, with a focus on how player position and offensive statistics relate to overall hitting performance.
Using R, the study cleans and restructures the data to analyze variables such as Batting Average (BA), Home Runs (HR), Runs Batted In (RBI), At-Bats (AB), and Extra-Base Hits, while also categorizing players by position and position type (infielder, outfielder, DH).
Through frequency distributions, cross-tabulations, and correlation plots, the analysis shows that hitters with higher batting averages tend to perform better across key power metrics, including HRs and RBIs. While not always leading in batting average, significantly outperform other position groups in home run production, highlighting how different offensive profiles contribute to player value.
