wiki:SummerSchools/2014/ProblemSets/OpenFootballData
Last modified 3 years ago Last modified on 07/14/14 10:23:33

Modified Mon Jul 14 10:23:33 2014 by Matt.Sottile.

Open Football Data

Introduction

This data set includes information about both league and international soccer matches over many years. One can play with this data to build simple models that only take into account the winner/loser outcome of each game, to more sophisticated models that consider score differentials, rosters (and roster changes over years, indicating trades), as well as overlapping competitions that could affect outcomes due to player participation (e.g., players out from league play due to international duties). Working with this data will be a good exercise in model building, as well as practice in data processing to distill from the raw data the necessary inputs for a model.

Data Sets

The data is available online in both textual and SQLite3 form.