FEATURE ENGINEERING FOR ROLE ASSESSMENT IN COUNTER-STRIKE 2

Yurii Kuzhii, Yuriy Furgala

Abstract


Background. The classification of in-game roles in team-based shooters, particularly Counter-Strike 2 (CS2), is an essential component of esports performance analytics. Existing approaches primarily rely on aggregate ratings or empirical assessments, which do not adequately capture the multidimensional structure of player behavior. As a result, there is a need to construct a behavioral feature space capable of reflecting role-specific differences and enabling reliable automated classification.

Materials and Methods. To construct the feature space, publicly available statistics and, when necessary, .dem files containing detailed logs of in-game events were utilized. The foundation consists of seven HLTV behavioral attributes, supplemented with metrics specific to the Terrorist (T) and Counter-Terrorist (CT) sides, as well as map-dependent indicators. The data were pre-cleaned, normalized, and structured at the player–map level. For the analysis, Principal Component Analysis (PCA) was applied, along with Analysis of Variance (ANOVA) to identify map-dependent features, and correlation analysis to examine relationships among behavioral metrics.

Results and Discussion. The results demonstrated that typical roles (entry-fragger, lurker, support, AWPer, anchor, and IGL) form distinct regions within the multidimensional feature space that cannot be reduced to a single numerical index. A set of features most influential for differentiating roles was identified, along with metrics that exhibit stable behavior regardless of map or side. The analysis based on grouping players revealed the absence of a universal player profile: strong performance in some metrics is accompanied by lower values in others, reflecting natural role specialization.

Conclusion. The proposed approach provides an informative representation of behavioral features and enables automated identification of player roles in CS2 without relying on aggregate rating systems. The constructed feature space has practical value for scouting, roster optimization, and match analysis, and can also be adapted for detecting smurfing or other forms of anomalous activity. The methodology demonstrates interdisciplinary potential and is promising for broader applications in behavioral analytics within online services.

Keywords: Counter-Strike 2; player role classification; behavioral features; HLTV attributes; ANOVA; correlation analysis.


Full Text:

PDF

References


[1] Serafico, L. (2025, May 12). Counter-Strike roles explained (and who does them best). Esports Insider. https://esportsinsider.com/counter-strike-roles-explained

[2] The utility of behavioral biometrics in user authentication and demographic characteristic detection: a scoping review. Syst Rev 13, 61. https://doi.org/10.1186/s13643-024-02451-1

[3] Seif El-Nasr, M., Drachen, A., & Canossa, A. (Eds.). (2013). Game Analytics: Maximizing the Value of Player Data. Springer. https://doi.org/10.1007/978-1-4471-4769-5

[4] Zimmer, F. (2025). Player behavior analysis for predicting player identity within pairs in esports tournaments: A case study of Counter-Strike using binary random forest classifier. Proceedings of the 58th Hawaii International Conference on System Sciences (HICSS 2025). University of Hawaiʻi at Mānoa. https://doi.org/10.24251/HICSS.2025.513

[5] Eggert, C., Herrlich, M., Smeddinck, J., Malaka, R. (2015). Classification of Player Roles in the Team-Based Multi-player Game Dota 2. In: Chorianopoulos, K., Divitini, M., Baalsrud Hauge, J., Jaccheri, L., Malaka, R. (eds) Entertainment Computing - ICEC 2015. ICEC 2015. Lecture Notes in Computer Science(), vol 9353. Springer, Cham. https://doi.org/10.1007/978-3-319-24589-8_9

[6] Demediuk, S., York, P., Drachen, A., Walker, J. A., & Block, F. (2019). Role Identification for Accurate Analysis in Dota 2. Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, 15(1), 130-138. https://doi.org/10.1609/aiide.v15i1.5235

[7] Drachen, A., Sifa, R., Bauckhage, C., & Thurau, C. (2012). Guns, swords and data: Clustering of player behavior in computer games in the wild. Proceedings of the IEEE Conference on Computational Intelligence and Games (CIG 2012) (pp. 163–170). IEEE. http://doi.org/10.1109/CIG.2012.6374152

[8] C. Bauckhage, A. Drachen and R. Sifa, "Clustering Game Behavior Data," in IEEE Transactions on Computational Intelligence and AI in Games, vol. 7, no. 3, pp. 266-278, Sept. 2015, doi: https://doi.org/10.1109/TCIAIG.2014.2376982

[9] Ferguson, M., Devlin, S., Kudenko, D., & Walker, J. (2020). Player style clustering without game variables. In Proceedings of the IEEE Conference on Games (CoG 2020) (pp. 556–563). IEEE. https://doi.org/10.1145/3402942.3402960

[10] Ding, Y.-J., Yap, W.-S., & Khor, K.-C. (2024). Profiling and identifying smurfs or boosters on Dota 2 using K-means and IQR. IEEE Transactions on Games. https://doi.org/10.1109/TG.2023.3317053

[11] Kuzhii, Y., & Furgala, Y. (2024). The anticheat development in the Counter-Strike main series. Electronics and Information Technologies, 26, 88–97. https://doi.org/10.30970/eli.26.8

[12] HLTV.org. (2024). Introducing HLTV Attributes. Retrieved from https://www.hltv.org

[13] Skybox Technologies ApS. (n.d.). CS2 stats – Track your stats like the pros | Skybox EDGE. Skybox EDGE. https://skybox.gg/edge/

[14] Qian, X., Sifa, R., Liu, X., Ganguly, S., et al. (2022). Anomaly detection in player performances in MOBA games. In Proceedings of the Australasian Computer Science Week (ACSW) 2022. ACM. https://doi.org/10.1145/3511616.3513095

[15] Pedraza-Ramirez, I., Sharpe, B. T., Behnke, M., Toth, A. J., & Poulus, D. R. (2025). The psychology of esports: Trends, challenges, and future directions. Psychology of Sport & Exercise, 81, 102967. https://doi.org/10.1016/j.psychsport.2025.102967




DOI: http://dx.doi.org/10.30970/eli.32.8

Refbacks

  • There are currently no refbacks.