Predicting the Success of Games on Steam Platform

Main Article Content

Duangruthai Preedanon
Jumpon Kluaymai-ngarm

Abstract

This study aims to identify key factors influencing the success of video games on the Steam platform and to develop a statistical model for predicting a game’s likelihood of success. It serves as a tool to support entrepreneurs in making comprehensive business decisions, enhancing their competitive advantage. Data covering 47,557 video games released between 2018 and 2023 were collected from VGInsights and SteamDB. A logistic regression model was employed to examine the relationships between game success and independent variables: genre, subgenre, price, release month, and publisher quality. The findings revealed that Casual and Sports genres demonstrated a significantly positive impact on success across multiple years. Subgenres such as 2D Platformer, Survival Horror, Shoot ’Em Ups, VR, and FPS also exhibited consistently positive influences. Game pricing showed a significantly positive effect on success annually, with exceptionally high odds ratios ranging from 38 to 196. Conversely, release months such as February, March, August, September, and October displayed a negative impact across multiple years. Furthermore, games categorized as Indie demonstrated a significantly positive effect during 2018–2020, reflecting strong market attention toward indie games during that period before the market became highly competitive. These results provide valuable insights and practical predictive tools for developers and publishers to enhance strategic decision-making in the game industry.

Article Details

Section
Research Article

References

Allison, P. D. (2014, March 23–26). Measures of fit for logistic regression [Paper presentation]. SAS Global Forum 2014, Washington, DC, United States. https://statisticalhorizons.com/wp-content/uploads/GOFForLogisticRegression-Paper.pdf

Arora, K. (2023, November 17). The gaming industry: A behemoth with unprecedented global reach. Forbes. https://www.forbes.com/councils/forbesagencycouncil/2023/11/17/the-gaming-industry-a-behemoth-with-unprecedented-global-reach/

Arsenault, D. (2009). Video game genre, evolution and innovation. Eludamos, 3(2), 149–176.

Cassell, A. (2013, August 23). Premium vs. freemium: How mobile games make money. Full Sail University. https://hub.fullsail.edu/articles/premium-vs-freemium-how-mobile-games-make-money

Chandler, G. N., & Hanks, S. H. (1994). Market attractiveness, resource-based capabilities, venture strategies, and venture performance. Journal of Business Venturing, 9(4), 331–349.

Costa, R. (2019, April 29). Steam reaches one billion accounts. TechRapter. https://techraptor.net/gaming/news/steam-reaches-one-billion-accounts

Esposito, N. (2005). A short and simple definition of what a videogame is. Proceedings of DiGRA 2005 Conference: Changing Views: Worlds in Play, 1–6. https://doi.org/10.26503/dl.v2005i1.177

Fang, F., Sun, Y, & Leu, J. (2015). Profit analysis for video game releasing strategies: Single-player vs. multiplayer games. Journal of Supply Chain and Operations Management, 13(1), 58–77.

Friedman, M. (2007). The social responsibility of business is to increase its profits. In W. C. Zimmerli, M. Holzinger, & K. Richter (Eds.), Corporate Ethics and Corporate Governance (pp. 173–178). Springer. https://doi.org/10.1007/978-3-540-70818-6_14

Gangwani, D., & Zhu, X. (2024). Modeling and prediction of business success: A survey. Artificial Intelligence Review, 57, Article 44. https://doi.org/10.1007/s10462-023-10664-4

Hani, E. H. (2021). What is business success? Sea – Practical Application of Science, 9(27), 191–195.

Hu, W., Wang, Y., & Xia, R. (2024). Machine learning-based Steam platform game's popularity analysis. Proceedings of the 1st International Conference on Data Science and Engineering (ICDSE), 1, 582–588. https://www.scitepress.org/Papers/2024/128538/128538.pdf

Indarti, N., & Langenberg, M. (2004). Factors affecting business success among SMEs: Empirical evidences from Indonesia. Proceedings of the Second Bi-Annual European Summer University, 1–14.

Jiang, Z. (2021). Predicting the popularity of independent video games on the Steam platform [Master’s thesis, University of North Carolina at Chapel Hill]. Carolina Digital Repository. https://cdr.lib.unc.edu/concern/masters_papers/s1784v376

Koskela, S. (2021). Pricing economics of video games: a panel data study on the effects of versioning on revenue [Master’s thesis, University of Oulu]. OuluREPO. https://oulurepo.oulu.fi/bitstream/handle/10024/20548/nbnfioulu-202201191079.pdf

Lee, S. (2025, March 19). Exploring 7 logistic regression case studies for business success. Number Analytics. https://www.numberanalytics.com/blog/logistic-regression-business-case-studies?/

Luisa, A. D., Hartman, J., Nabergoj, D., Pahor, S., Rus, M., Stevanoski, B., Demsar, J., & Strumbelj, E. (2021). Predicting the popularity of games on Steam. Elektrotehniški Vestnik, 88(4), 151–162.

Rogers, S. (2014). Level up!: The guide to great video game design. John Wiley & Sons.

Soule, P. B. (2017). Predicting student success: A logistic regression analysis of data from multiple SIU-C courses. [Master’s thesis, Southern Illinois University Carbondale]. OpenSIUC. https://opensiuc.lib.siu.edu/cgi/viewcontent.cgi?article=2107&context=gs_rp

SteamDB. (n.d.). All Steam game tags. https://steamdb.info/tags/

VGInsights. (2024, July 3). Games database. https://vginsights.com/games-database?priceRange=5&releaseRange=2017-01-01,2023-12-31&adult=1

Zhang, J., Zhao, H., Chen, Z., & Song, Z. (2020). Prediction of the most popular game tags on Steam under the influence of COVID-19 based on machine learning and natural language processing. The Frontiers of Society, Science and Technology, 2(12), 70–80.

Zukowski, C. (2022, August 23). 4 Tips to help you price your indie game. How to Market a Game. https://howtomarketagame.com/2022/08/23/4-tips-to-help-you-price-your-indie-game/

Zukowski, C. (2023, April 25). What if Valve didn’t take 30 percent? How To Market A Game. https://howtomarketagame.com/2023/04/25/what-if-valve-didnt-take-30-percent/