Home is Where the Points Are
Are Soccer leagues with high levels of parity between teams more affected by the concept of home field advantage? The world of team management and sports betting rely heavily on understanding how teams perform beyond player stats and win loss records. The concept of a home field advantage is not new to sports. An intrinsic value may exist between a team and their home crowd, enabling a team to play more effectively than statistics would predict. This paper intends to quantify the impact of home field advantage while controlling for variations of parity
Researchers have theorized what factors cause home field advantage. The Institute for the Study of Labor, a German independent research group, published a study written by Thomas Dohmen in 2003 focusing on how a crowd may impact referees and force them to be partial to one team over another. The study determined that crowd size and position relative to the field are directly related to the number of favorable calls to the home team. In a similar vein, the book “Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won” by Tobias Moskowitz and Jon Wertheim discusses how referee bias affects games in the NFL, NBA, MLB, MLS, and NHL. The pair discovered that referees assign penalties and make decisions that are more likely to benefit the home team. Both of these studies show that home field advantage exists in some form, and how referee bias may impact a match. This paper will explain how a team may over perform or underperform against a team, based on whether they are playing at home or away. Data was collected from four regional European soccer leagues; the English Premier League, the German Bundesliga, the Italian Serie A, and the Spanish La Liga, from 2007-2011. Taken from five seasons, this data includes the results from thousands of matches. Teams are ranked by tiers, separating the top, middle and bottom thirds, giving the study a set of variable that transcend a specific year or league. Because tiers will be compared rather than teams, a team’s variation in skill across seasons did not play a factor in the study. Due to the large number of matches analyzed, variables that influence a single match, such as inclement weather, stadium conditions, and teams’ style of play, are less likely to be statistically significant.
This study defines the parity as the variance of total points a team earns per game within a specific league during a specific year. The dependent variable will be the number of points earned by a team at year i in the league j. A team earns three points if it wins the game, 1 point if the two teams tied, and 0 points if the team loses. The independent variables will include the league a team plays in, the given year, and if the team plays at home. Controlled variables are the parity levels of the three tiers of teams.
II Model specification
The raw data came from ESPNFC which has a...