Understanding and Predicting Player Retention in Online Gaming


User retention is a critical concern for online gaming platforms. This talk focuses on understanding and predicting player retention in online poker games by characterizing user behaviors and developing predictive models. The study examines two types of games available on the platform: cash games and tournaments with entry fees. Through exploratory data analysis, the user journey and behavior patterns are explored using historical online poker playing data. Key actions, such as deposits, withdrawals, and game participation, are analyzed to gain insights into user attitudes and preferences.

The analysis reveals that the level of deposits and buy-ins varies among users, reflecting their risk-taking propensity and investment in the game. The monthly retention rate fluctuates around 50%, indicating that approximately half of the players from the previous month leave the platform. Moreover, users with longer tenure exhibit higher retention rates, eventually converging at around 87%. Noteworthy correlations between retention and user actions are identified, such as the positive relationship between retention and deposit intensity and frequency, as well as the increased likelihood of retention among users who engage in both cash games and tournaments.

Based on the exploratory analysis, relevant features for building the retention model are selected, including game types, tournaments played, cash buy-ins, deposits and withdrawals, and guaranteed prize pools won. Logistic regression is employed as the modeling technique due to its interpretability and simplicity to prevent overfitting. The model results demonstrate that deposit amount, the number of cash games and tournaments played, and the guaranteed prize amount positively impact player retention. Notably, in tournament games, players who pay entry fees are more likely to be retained compared to free players. Conversely, for cash games, the buy-in level does not correlate with higher retention, as players with varying buy-ins exhibit similar retention rates.

The talk concludes by offering recommendations for player retention based on the data analysis and modeling outcomes, as well as presenting ideas for future research. Although the study employs data from an online poker platform, the analytical methods and models can be applied to other types of games as well.


Veena Mendiratta is currently an Adjunct Professor in the Machine Learning and Data Science (MLDS) Program at Northwestern University, and an accomplished technology advisor, mentor, and speaker. With over 35 years of experience in the telecommunications industry, she recently retired from her position as the research lead (Director Level) for network reliability and analytics at Nokia Bell Labs. Veena holds a B.Tech. in Engineering from the Indian Institute of Technology (IIT), New Delhi, India and a Ph.D. in Operations Research from Northwestern University, USA. Veena is widely recognized as a researcher in the field of network reliability and analytics, having published over 65 papers in conferences and journals of the IEEE, ACM, and INFORMS. She has also presented tutorials on reliability modeling and analysis, and on data analytics at ISSRE, DSN, and KDD conferences. Her work has focused on the reliability and performance analysis for telecommunications systems products, networks, and services to guide system architecture solutions. Her research interests include system and network dependability analysis, software reliability engineering, network resiliency, and telecom data analytics. Veena was a Fulbright Specialist Scholar for 5 years and has visited universities in India, Norway, and New Zealand.

Open Data Science




Open Data Science
One Broadway
Cambridge, MA 02142

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google