Image credit: 2021/2022 Lega Serie A Teams, Tweet July 14, 2021 @SerieA_EN.

Introduction

Serie A is the premier Italian football league, and one of the strongest leagues in the world, home to many famous clubs such as AC Milan, Internazionale, and Juventus. Since 2004, 20 clubs have vied for the Serie A championship title, with the winners displaying the coveted scudetto on their uniforms the following year. Note that the three lowest-placing teams are demoted to Serie B the following season (and similarly three teams from Serie B are promoted to Serie A), and so the teams comprising Serie A change across seasons.

The tournament itself is a round-robin with 38 total games per season. During the first half, each club plays each of the others exactly once; in the second half, the clubs play once again but with the home/visiting team for each match being reversed. This is intended to address the home-field advantage, a perceived benefit that the home team (i.e., the team which hosts the match) has over the visiting team (i.e., the team that must travel to the stadium to play).

Project objectives

As a preliminary analysis, investigate and quantify the magnitude of any potential home-field advantage in Serie A football. Is there evidence that some teams appear to have more home-field advantage than others, while accounting for the relative strength of their opponents? Be sure to suitably define “home-field advantage” using the data at hand.

Of particular interest is the following primary analysis question. Home-field advantage is often attributed to factors such as crowd involvement (indeed, the “12th man” is often said to be a crucial factor in home wins). The 2019-2020 and 2020-2021 Serie A seasons were impacted by the global coronavirus pandemic, with all matches after the March 1, 2020 game between AS Roma and Cagliari being played in completely empty stadiums.

Is there evidence that any potential home-field advantage was attenuated, or even eliminated, during the COVID-19 pandemic compared to prior years? If so, estimate the degree to which it was reduced, and provide a suitable quantification of variability in your estimate. You must account for potential confounding factors in your analysis (e.g., relative skill disparity between teams, potential size of audience missing, etc.).

Clearly write any model(s) using correct mathematical notation. Care should be made to use readily-interpretable models, with conclusions and interpretations able to be understood by allied researchers and the knowledgeable public.

Detailed instructions, the data, and data descriptions are available in the course GitHub repository.

Learning objectives

  • Solidify skills in reproducible research and programming, including version-control and collaboration via GitHub
  • Critically think about reasonable analysis approaches in the context of real-world data
  • Express statistical models clearly and correctly
  • Develop scientific writing skills by providing clear, concise, data-driven conclusions suitable for allied researchers

Project timeline

  • Group: Report and reproducible code
    • Due Friday, January 21
  • Group: Revised report and response to reviewers
    • Due Friday, February 4

Note: each team’s GitHub report repository and commit history will also be evaluated by the instructor. The GitHub repository must contain the reproducible R Markdown document corresponding to the submitted reports, and will be checked throughout the course of the case study to ensure all team members are making meaningful contributions to the project.

References

[1] Lega Serie A. https://www.legaseriea.it/en, Accessed January 1, 2022.

[2] Eurosport. “Serie A Table.” https://www.eurosport.com/football/serie-a/standing.shtml, Accessed January 1, 2022.

[3] ESPN. “Italian Serie A Table.” https://global.espn.com/soccer/standings/_/league/ita.1, Accessed January 1, 2022.