For publications currently in preparation or in review, information and updates will be published on the main blog and on my twitter feed (@andersdrachen). See the “publications” tab for a complete list of publications and reports. Selected reports are listed below.


Intro to User Analytics 

The science of game analytics has gained a tremendous amount of attention in recent years. Introducing analytics into the game development cycle was driven by a need for better knowledge about the players, which benefits many divisions of a game company, including business, design, etc. Game analytics is, therefore, becoming an increasingly important area of business intelligence for the industry. (read more) (pre-print)

darkwing_duck_vector_by_sailorcardknight-d4vnmrsGame Analytics: The Basics

This is an early version of a document that later became an introduction chapter in the book “Game Analytics – Maximizing the Value of Player Data”. It is intended as a basic introduction to what Game Analytics is for the non-expert and forms a good place to start. It contains some of the material as the text above but has more depth (and is longer) but written earlier, so some new ideas are included in the Game Developer Magazine feature above which are not in this article (read more) (pre-print)






Guns, Swords and Data: Clustering of Player Behavior in Computer Games in the Wild

Behavioral data from computer games can be exceptionally high-dimensional, of massive scale and cover a temporal segment reaching years of real-time and a varying population of users. Clustering of user behavior provides a way to discover behavioral patterns that are actionable for game developers. (read more) (pre-print)



tomb_raider_underworld_wallpaper_10-widePlayer Modeling using Self-Organization in Tomb Raider: Underworld

We present a study focused on constructing models of players for the major commercial title Tomb Raider: Underworld (TRU). Emergent self-organizing maps are trained on high-level playing behavior data obtained from 1365 players that completed the TRU game. The unsupervised learning approach utilized reveals four types of players which are analyzed within the context of the game. (read more) (pre-print)



Just_Cause_2How Players Lose Interest in Playing a Game: An Empirical Study Based on Distributions of Total Playing Times

An important problem in game analytics is to characterize how player engagement evolves over time. Reliable models allow for assessing the long-term success of game products and can provide estimates of how long players may be expected to keep actively playing a game. We introduce methods from random process theory into game data mining in order to draw inferences about player engagement. We find that the Weibull distribution gives a good account of the statistics of total playing times. This implies that an average player’s interest in playing one of the games considered evolves according to a non-homogeneous Poisson process. Therefore, given data on the initial playtime behavior of the players of a game, it becomes possible to predict when they stop playing. (read more) (pre-print)





fig_4_leftEvaluating Motion. Spatial User Behavior in Virtual Environments – PDF (pre-print)

Games are in the vast majority of cases based on spatial operations. Therefore, spatial behaviour analyses are useful to game research and design. In this paper, spatial analysis methods are introduced and arguments posed for their use in user-behaviour analysis. Case studies involving data from thousands of players are used to exemplify the application of instrumentation data to the analysis of spatial patterns of user behaviour. (read more) (pre-print)





imagesdddPredicting Player Behavior in Tomb Raider: Underworld 

This paper presents the results of an explorative study on predicting aspects of playing behavior for the Tomb Raider: Underworld (TRU). Various supervised learning algorithms are trained on a large-scale set of in-game player behavior data, to predict when a player will stop playing and, if the player completes the game, how long will it take to do so. Results reveal that linear regression models and other non-linear classification techniques perform well on the tasks and that decision tree learning induces small yet well-performing and informative trees. (read more) (pre-print)



fig_2_bottomAnalyzing Spatial user Behavior in Computer Games using Geographic Information Systems 

In this paper from 2009, existing work on gameplay metrics is reviewed, and spatial
analysis of gameplay metrics introduced as a new approach in the toolbox of user-experience testing and –research. Furthermore, Geographic Information Systems (GIS) are introduced as a tool for performing spatial analysis. A case study is presented with Tomb Raider: Underworld, showcasing the merger of GIS with gameplay metrics analysis and its application to game testing and –design. (read more) (pre-print) [Best Paper Award at MindTrek 2009].







Fun Fact about World of Warcraft Character Names

In this presentation and the associated report, the first large-scale analysis of virtual identities in the massively multi-player online game World of Warcraft is presented based on a dataset of nearly eight million avatar names and associated information. The results presented highlight the inventiveness of the names the users of World of Warcraft express in terms of naming their virtual identities, and the varied nature of the sources of inspiration for these names, which range from mythology, literature, popular culture, real-world names and others. The results also indicate that how players name their characters is influenced by the aesthetics and game function of the characters. The server type also appears to have an impact on player naming strategies, with role-playing servers forming a distinct cluster from Player-vs-Player type servers. Additionally, the analysis presented reveals that character name frequencies follow, similar to real-world names, a power law distribution. (read more) (pre-print).           



Player Progression Analysis

Is there an identifiable playstyle which people adopt throughout a game? Does this style change as the game progresses? Recent research has investigated the patterns of play and progression in Tomb Raider: Underworld, and found that most players vary their playstyles as they progress through the game. (read more) (pre-print)




Dynamic Behavioral Profiling and Visualization

In this brief report we present an interactive visualization, based on Sankey flow diagrams, which shows the proportion of different behavioral clusters from the browser-based MMORPG Glitch, as a function of time, and the flow of players between profiles. The diagram allows evaluation of migration of players between clusters as a function of time, as well as churn analysis. (read more) (pre-print)






Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s