The network
behind the game.

Clustering coefficient measures how often a player's teammates also pass to each other. High clustering means ball movement is spread across the whole roster — no isolated stars.

Shannon entropy measures how evenly distributed a player's passes are. High entropy means passes spread across many receivers — low entropy means one or two dominant targets.

Each player is classified across three dimensions using GMM clustering on graph and usage metrics — how they score, how they move the ball, and how they're used. Every player gets a label in each dimension.