According to the authors of this paper, in the past three years there has been an increased interest in Massive Multiplayer Online Games (MMOGs) in the academic community, specifically with the game World of Warcraft (WoW). Several areas of research have analyzed individual player attributes such as player motivations (Yee, 2005) and player characteristics (Bartle, 1996). An increasing amount of research has been dedicated to the social aspects of the game. Research in this area has looked at various types of social interactions among players, guild formation, and even the spread of viruses across a server.
This paper attempts to describe the structure of arena teams in relation to the overall structure of the social network on a particular server. By looking at guild relationships and arena team composition, the authors hope to shed light on which arena teams have the highest efficacy and are most successful in "ladder" competition. Currently, arena team players are given a rating based upon how many battles they won. By comparing this to arena team composition via by treating the rating of each arena team as a vector, this study aims at determining which factors are most important in developing an accomplished team via network visualization. In particular, the authors wanted to find out how an arena team's overall centrality in the network affects their overall success, and in addition, if there is a contrast in how arena teams are formed on large versus small servers (either through guild or non-guild relationships).
The first image represents a look into the Spinebreaker (PvP) server - 3 mode Alliance side: 16,534 players, 4065 guilds, and 5758 arena teams. The second is a visualization of the top 50 Horde teams in the the Windrunner (PvE) server.