The determinants of attendance at neutral site college football games

This paper develops a predictive model of attendance at neutral site Division 1-A college football games. From 2004–2012 seasons, 427 games were identified and split into training and holdout data sets. A Tobit model is developed using matchup-specific, game-specific, location-specific, and universi...

Full description

Saved in:
Bibliographic Details
Published in:Managerial and decision economics 2015-04, Vol.36 (3), p.191-204
Main Author: Mirabile, McDonald P
Format: Article
Language:English
Subjects:
Athletics
Attendance
Bowl Games
College football
Colleges
Decision
Decision making models
Decision theory
Determinants
Division
Economics
Erfolgsfaktor
Football
Fußball
Games
Hochschule
Mathematical models
Modell
NCAA
Neutral Site
Panel data
Sports
Sports training
Studies
USA
Venue
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper develops a predictive model of attendance at neutral site Division 1-A college football games. From 2004–2012 seasons, 427 games were identified and split into training and holdout data sets. A Tobit model is developed using matchup-specific, game-specific, location-specific, and university-specific determinants. Results are generally consistent with the existing literature on attendance modeling, showing that these prior home team-specific models are adaptable to neutral site locations. Results are useful as they reveal how the selection of game location and matchup of opponents – decisions made by athletic directors or bowl officials – ultimately affect game-day attendance.
ISSN:0143-6570
1099-1468
1099-1468
DOI:10.1002/mde.2670