Positions in the table have been classified by using a clustering algorithm which groups similar players based on their previous ten games, where they have played less games they are unclassified. Classifications change over time as players switch roles however it will take multiple games for this to happen.

As the classifications are generated from an algorithm there will always be a small percentage of players that are classified incorrectly (e.g. swing men may come up as wings as stats are gathered at both ends of the field over multiple games), over time these tend to correct themselves so please don’t @ us on Twitter to let us know of an obvious misclassification ????.