Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 243 - Statistics in Sports and Beyond
Type: Contributed
Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
Sponsor: Section on Statistics in Sports
Abstract #319153
Title: Predicting the Baseball Winning Percentage Using Three Regression Models
Author(s): Austyn Hughes and Jacob Wheeler and Abdelmonaem Jornaz*
Companies: United States Department of Agriculture and University of Missouri - Kansas City and Northwest Missouri State University
Keywords: Baseball; winning percentage; Ordinary Least Square regression; Beta regression; fractional logit regression
Abstract:

Baseball is one of the most popular sports in the United States, and one of the most statistically driven. Predicting a team’s winning percentage can determine the direction a team takes in each season. This can help teams make financial decisions to win now or focus on the future. The historical data for Major League Baseball (MLB), the oldest major professional sports league in the United States and Canada, from 1985 to 2020 has been used as training data to predict the 2021 team winning parentage which is the test data. Since the winning percentage can be observed on the open interval (0, 1), three regression models have been applied to estimate the winning percentage for the subsequent year 2021, these models are Ordinary Least Square (OLS) regression, Beta regression, and fractional logit regression. The OLS model gave a better result comparing with the other two models.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program