Presenter(s)

Zachary Andrew Lentz

Download

Download Project (145 KB)

Description

This project will use a logistic regression analysis to determine the effect of home court on the probability of winning an NBA game. The data collected is on all thirty NBA teams from the 2018-19 regular season, since this was the last normal season before Covid-19 happened. Each data point analyzed is a particular game with multiple variables for that game. The dependent variable is win, which is a binomial variable, 0 if that game is a loss and 1 if it is a win. The independent variable home is a binomial variable, 0 if the team was a away and 1 if the team was home for the game. The other independent variables are continuous: number of All-Stars for the 2018-19 season that were on the team, number of All-Stars for the 2017-18 season that were on the team, number of wins the team had in the 2017-18 season, and a variable representing head coaching experience. The head coaching experience variable is the number of wins the head coach had divided by the number of years of head coaching experience entering the 2018-19 season. Logistic regression is used to evaluate the impact of the independent variables on the probability of winning the game. The ultimate goal is to determine if there is an advantage to have the home court and crowd in an NBA regular season game.

Publication Date

4-22-2021

Project Designation

Capstone Project

Primary Advisor

Peter W. Hovey

Primary Advisor's Department

Mathematics

Keywords

Stander Symposium project, College of Arts and Sciences

Home Court Advantage In The NBA

Share

COinS