Visualisation of top teams in the Premier League, treated as a pool in an Elo scheme. Calculating the elo for each team over the time period was done in Ruby, outputted to a .csv file and visualisations are done in javascript with usage of d3.js.
Elo rating systems are used to calculate relative skill between players in competitive player vs player games, most notably in Chess and Video Games. All players start with a pre-chosen number of points, in my case 1500. Players that win gain points depending on the point difference between them and the loser - the lower relative ranking the loser, the less points they gain. The loser loses the same number of points. The points to take is calculated with the following formula : where
. a represents the team to calculate for and b its opposition. Sa is 0 for a loss, 0.5 for a draw and 1 for a win. K is some constant that determines how much effect one game has.
Data was obtained from here. The code used to format the data for the elo calculating script is here.
Just messing around with d3.js.