AILGMLNov 2, 2020

Valuing Player Actions in Counter-Strike: Global Offensive

arXiv:2011.01324v232 citationsHas Code
AI Analysis

This addresses a problem for CSGO teams, media, bettors, and fans by providing a reproducible analytical framework, though it is incremental as it builds on existing sports analytics concepts.

The paper tackled the lack of quantitative evaluation tools for Counter-Strike: Global Offensive players by introducing a data model, graph distance measure, and context-aware framework to value player actions based on changes in win probability, using over 70 million in-game events to demonstrate consistency and independence from existing methods.

Esports, despite its expanding interest, lacks fundamental sports analytics resources such as accessible data or proven and reproducible analytical frameworks. Even Counter-Strike: Global Offensive (CSGO), the second most popular esport, suffers from these problems. Thus, quantitative evaluation of CSGO players, a task important to teams, media, bettors and fans, is difficult. To address this, we introduce (1) a data model for CSGO with an open-source implementation; (2) a graph distance measure for defining distances in CSGO; and (3) a context-aware framework to value players' actions based on changes in their team's chances of winning. Using over 70 million in-game CSGO events, we demonstrate our framework's consistency and independence compared to existing valuation frameworks. We also provide use cases demonstrating high-impact play identification and uncertainty estimation.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes