Desktop viewing is recommended. Viewing this on mobile? Click here to view the interactive visualization full-screen
Network analysis and visualization can help us understand the rich web of connections existing between players, clubs and countries. In the interactive network above, players are connected to their countries and clubs. This is an example of a so-called tripartite or three-mode network, since we’re visualizing three node types in one network. Club nodes are sized according to the number of players they represent.
Clicking a node highlights its direct and indirect connections. For example, clicking England shows you that all its players originate from domestic clubs:
Clicking a player highlights his country and club, but also immediately shows you which other players they are connected to via their countries and clubs. You can also read up on some more information on the player in the panel on the right-hand side. Here we see how Lionel Messi connects to other players via his club Barcelona and his home country of Argentina:
Focusing on a club highlights the players connected to it, and also visualizes the countries to which this club supplied players. Manchester City has most players represented, a total of 16 to 8 countries:
Other clubs with over 10 players represented are Real Madrid (15), Barcelona (14), Paris Saint-Germain and Tottenham Hotspur (12 each) and Juventus, Bayern Munich, Chelsea and Manchester United (11 each).
You can also use the group filters on the left-hand side to focus on a (combination of) group(s) of interest to you. Take a look at Group H, which has quite a few connections between countries based on players’ shared clubs:
Or if you want to find out where a specific player, club or country is located in the network, use the search function:
A social network of players
Instead of focusing on players, clubs and countries at the same time, we can also visualize the social network of players connected to other players based on shared clubs and countries. This is called a monopartite or one-mode projection of the network above, and makes it easier to spot players with key positions. It also allows us to calculate some measures of importance for each of the players, such as their degree centrality (the number of other players they are connected to) or their betweenness centrality (how often a player appears on a shortest path between other players).
Let’s start by taking a look at the players with the largest number of connections (the darker and larger a node, the more connections it has):
The players with most connections are clustered together based on clubs which have a large number of players represented in the tournament. David Silva (Spain), Bernardo Silva (Portugal), Benjamin Mendy (France) and İlkay Gündoğan (Germany) are the top scorers with 37 connections and play for Manchester City (see above). Most of the other highly connected players in the image above play for Manchester City, Real Madrid or Barcelona.
The picture looks quite different when looking at betweenness centrality:
The players with the highest betweenness scores appear most often on the shortest paths between other players in the network, and are important in keeping the network together because of their hub positions. They all reside outside of the cluster of highly connected players above. In a sense, they are the main hubs through which their country’s team mates can reach the rest of the network, because they play for a club which supplies players to various other country teams as well.
Brian Idowu (Nigeria), Saeid Ezatolahi (Iran), Yohan Benalouane (Tunisia), Son Heung-min (South Korea) and Jefferson Farfán (Peru) have the highest scores. As an example, Tunisia’s Benalouane (Leicester City) connects to players from Portugal, Denmark, England, Japan and Nigeria, and holds a key position in terms of connecting his fellow Tunisian players to the rest of the network.
Comments are welcome! Drop me a line on Twitter: @Andre_Vermeij
Enjoy the World Cup!
