Over the past few decades football has been criticised by many for being uncompetitive. The elite clubs in europe continue to increase their revenues year after year and thus can spend small fortunes on the best players of smaller clubs. This drastically reduces the competitveness of leagues which results in a predictable and less interesting spectacle for everyone.
In information theory entropy
is a measure of information and uncertainty. This analysis utilizes entropy as a measurement of the unpredictabiity of each league. Entropy is computed based on the probabilities of each outcome for every game based on the odds from bookmakers.
The idea here is that if there are consistently large favourites in games throughout the season then the league is considered predictable and therefore uncompetitive.
Entropy is at its minimum when one outcome has a probability of 1 (H = 0, totally predictable).
Entropy is at its maximum when each outcome has an equal probability (H = 1.098, totally unpredictable).
The data used in this visualization is taken from Kaggle’s European Soccer database, which spans across 8 years from 2008 to 2016. The data was processed and the entropy for each team and for each year was calculated in an ipython notebook.