Projects

Predicting Left-Handed Batting Performance with Modified Shift Rules

By: Max Brown, Noah Costa, Aritra Acharjee, Jack Casey

This project evaluates the effect of MLB’s rule change that prevents a defensive shift in the infield. Are we able to identify left-handed hitters that will increase performance based on prior hitting data? How will hitters alter their approach due to this rule change?

Predicting Sports Bets within the FIFA 2022 World Cup

By: Sean Li, Aaditya Warrier, Christina Yoh, Brian Janger

This project aims to build a predictive model for the FIFA 2022 World Cup. The model predicts goal line bets and money line bets and tries to find margins against the sportsbooks, comparing odds to sportsbooks to determine which bets are more likely to be profitable.

Does Handedness Still Matter In The MLB?

By: Max Brown, Evan Ginsburg, Luke Lorentzatos, Raymond Xiong

This project evaluates the overarching effect of handedness on pitcher/hitter matchups across different time periods in the MLB. Should teams still be utilizing the assumption that same sided matchups are advantageous for the defense?

Measuring the Impact of NCAA Football Head Coach Changes

By: Jack Miller, Alex Jackson, Bryce Grove

This project aims to quantify the impact of first-year NCAA football coaches. The scope of this project is NCAA Division 1 programs who have made head coaching changes in the last 10 seasons. The project compares metrics like win percentage to contruct linear regression models to ultimately quantify and predict the impact of first-year coaches.

Predicting NFL Upsets Using Multilevel Logistic Regression

By: Jack Miller

This goal of this project is to predict upsets in NFL games using multilevel logistic regression. The project utilizes NFL play-by-play and schedule data to create a model to discover which variables are the best predictors of an upset.

NBA Fouls Survivorship Model

By: Bryce Grove

This paper utilizes 2020-2021 NBA Play-by-Play data to help answer the “pull your player” problem. By using historical data to estimate foul rates by position and game time, the paper provides an accurate probability a given player fouls out for each additional minute played.

Tutorial Post

By: Jack Lichtenstein

The goal of this post is to familiarize R programming beginners with some public sports data sources. Specifically, we will look at scraping sports-reference.com ’s college basketball data and then making some visualizations. Let’s get started!