The footballers search relations visualization is an interactive network graph showing the connections of former and current football players and managers based on searches performed on Google. The graph consists of 1106 nodes and 2880 edges.
As a starting point I took the player names from the Guardian's Top 100 footballers list as seed queries and crawled Google's search results 5 levels deep. To be included in the graph the search query's result page needs to show knowledge graph information. If a result contains related searches up to 5 of them were also performed until the maximum depth level was reached.
To limit the graph to persons who relate to the football topic their knowledge graph description had to contain words such as footballer, football player, football manager, striker, soccer player, soccer coach... you get the idea.
The resulting graph is less surprising than the programmer search network I created before. The most connected node is Lionel Messi with a degree of 20, somehow playing in his own league here too with Cristiano Ronaldo and Neymar following up each with a degree of 16. The average degree in this network is 5.208 and the modularity 0.86 with 37 recognized communities.
Looking at the different communities one can see that they are often made up of players from the same team, e. g. an Arsenal London community around Bacary Sagna, an Internazionale community around Diego Milito, a Bayern Munich community around Bastian Schweinsteiger, and an FC Barcelona community around Andrés Iniesta.
This network graph shows only a small fraction of the football players known to Google. Freebase alone, which is one of the sources for Google's Knowledge Graph information, currently lists 104,710 football player topics.
Determined by the initial list of current top players, former football stars are less prominent in this network. If I had started with searches for Pelé, Franz Beckenbauer, Diego Maradona and the likes, I'd expect a different network, but still with many active footballers, who I suppose are searched more frequently.