Cheaters in a gaming social network

Abstract

The popularity of online gaming has raised signi cantinterest in understanding the technological needs for sup-porting gaming platforms. Consequently, various studiescharacterized network trac due to gaming, resource pro-visioning, work load prediction, and player churn in onlinegames [1]. Other studies have focused on the psycholog-ical and social properties of gamers and gaming commu-nities [5]. Gamers form strong communities around theiractivity, with recent work suggesting that explicitly de nedrelationships of gamers tend to be supported by real worldrelationships [6].One major problem with gaming is cheating. For somecheaters, the motivation is monetary. Virtual goods areworth real world money on eBay, and online game econo-mies provide a lucrative opportunity for cyber criminals [2].For other cheaters, a competitive advantage and the desireto win is motivation enough [3].In this work, we analyze how cheaters are embedded in theSteam Community, a large online social network for gam-ing with millions of active users. Cheaters are identi ed byan automated mechanism operated by the Steam gamingservice and their pro les are permanently agged in a pub-licly visible way. We analyze di erences in gaming-speci cproperties of cheaters and non-cheaters. We also examinewhether cheaters occupy a particular position in the socialnetwork when compared to non-cheaters, and characterizethe relationships they have among themselves, and with non-cheaters.Our study shows that cheaters are well embedded in thesocial network, having similar connectivity to non-cheaters;their social position is largely undistinguishable from thatof fair players; and their geographic distribution does notcorrespond to real world population nor is it determinedby the popularity of the Steam Community at a given realworld location.