The European Soccer Database contains data on more than 25.000 national football matches from the best European leagues. The aim of this exercise is to present interesting relationships in R using explorative data analysis and visualization.
First you need to access some tables in the database. Note: You can use the RSQLite::dbConnect() function to do this. To access a particular
database table and convert it to a data.frame , you can use the tbl_df(dbGetQuery(connection, 'SELECT * FROM table_xyz')) command.
The first leagues of Spain, England, Germany and Italy are considered the four most attractive football leagues in Europe.
-
In which of the four leagues do on average score the most or the fewest goals per game?
-
Compare the average, median, standard deviation, variance, range and interquartile distance of goals scored per match between the four most attractive European leagues and the remaining leagues.
-
Is there really a home advantage? Use a box plot to show the number of goals scored by home and away teams.
-
“All soccer players are fair-weather players!” Check the assertion with a line chart: Do on average more goals fall per game in the summer months than in the rest of the year?
- Display the average goals scored per game for the top 4 leagues per year from 2008 to 2016.
- Use an estimated density function curve AND a QQ-Plots to check whether the home_team_possession variable is (approximately) normally distributed.
- Use a box plot to show whether there is a correlation between ball ownership (
home_team_possession) and the number of goals (home_team_goals) scored per game for home teams. Create four categories of ball ownership shares: very low (<=25%), low (25%<x<=50%), high (50%<x<=75%) and very high (x>75%).
