[Paper Review] Cheaters in the Steam Community Gaming Social Network

Jeremy Blackburn, Ramanuja Simha+6 more|arXiv (Cornell University)|Dec 21, 2011

Digital Games and Media27 references3 citations

TL;DR

This study analyzes cheaters in the Steam Community gaming social network using data from over 12 million users and in-game interactions from 10,000 players. It finds that cheaters are well-integrated into the social network, cheating behavior spreads socially through friend influence, and cheaters face social penalties such as friendship loss and increased privacy settings, despite remaining active in the community.

ABSTRACT

Online gaming is a multi-billion dollar industry that entertains a large, global population. One unfortunate phenomenon, however, poisons the competition and the fun: cheating. The costs of cheating span from industry-supported expenditures to detect and limit cheating, to victims' monetary losses due to cyber crime. This paper studies cheaters in the Steam Community, an online social network built on top of the world's dominant digital game delivery platform. We collected information about more than 12 million gamers connected in a global social network, of which more than 700 thousand have their profiles flagged as cheaters. We also collected in-game interaction data of over 10 thousand players from a popular multiplayer gaming server. We show that cheaters are well embedded in the social and interaction networks: their network position is largely undistinguishable from that of fair players. We observe that the cheating behavior appears to spread through a social mechanism: the presence and the number of cheater friends of a fair player is correlated with the likelihood of her becoming a cheater in the future. Also, we observe that there is a social penalty involved with being labeled as a cheater: cheaters are likely to switch to more restrictive privacy settings once they are tagged and they lose more friends than fair players. Finally, we observe that the number of cheaters is not correlated with the geographical, real-world population density, or with the local popularity of the Steam Community. This analysis can ultimately inform the design of mechanisms to deal with anti-social behavior (e.g., spamming, automated collection of data) in generic online social networks.

Motivation & Objective

• To understand the social network position of cheaters in the Steam Community gaming platform.
• To investigate whether cheating behavior spreads through social influence among players.
• To analyze the social consequences of being publicly flagged as a cheater, including changes in privacy and friendship dynamics.
• To examine whether geographical or population density factors correlate with cheating prevalence.
• To inform anti-cheat mechanisms and broader anti-social behavior detection in online social networks.

Proposed method

• Collected data from over 12 million Steam users via web crawling, including social connections and VAC-ban status.
• Gathered in-game interaction logs from a popular multiplayer server for over 10,000 players to analyze behavioral patterns.
• Used MapReduce on Amazon Elastic MapReduce to process large-scale network metrics, including degree centrality and betweenness centrality.
• Applied geo-social metrics to assess spatial clustering and regional patterns of cheating behavior.
• Trained machine learning models (logistic regression, naive Bayes, decision trees) to classify players as cheaters based on profile and network features.
• Conducted temporal analysis to assess the correlation between a player’s cheater friends and their future likelihood of becoming a cheater.

Experimental results

Research questions

• RQ1 How are cheaters positioned within the social and interaction networks of the Steam Community?
• RQ2 Does the presence and number of cheater friends predict a fair player’s future likelihood of cheating?
• RQ3 What social consequences do players face after being publicly flagged as a cheater?
• RQ4 Is the distribution of cheaters correlated with real-world geographical population density or local Steam popularity?
• RQ5 Can network-based features improve the detection of cheaters in online gaming environments?

Key findings

• Cheaters are well-integrated into the social network, with network positions largely indistinguishable from fair players in terms of centrality metrics.
• The presence and number of cheater friends significantly correlate with a fair player’s future likelihood of becoming a cheater, indicating social spread of cheating behavior.
• Cheaters are more likely to switch to private profile settings after being flagged, suggesting social embarrassment or discomfort.
• Cheaters experience greater friendship loss over time compared to fair players, indicating a measurable social penalty.
• Cheating prevalence does not correlate with geographical population density or local Steam popularity, suggesting cultural or community-level factors may drive cheating.
• Machine learning models using profile and network features can classify cheaters with accuracy between 65% and 74%.