A Taxonomy of Collectible Card Games: A Game-Playing AI Perspective

Collectible Card Games represent a complex domain for artificial intelligence research, characterized by a dual-phase structure consisting of strategic deck-building and tactical battling. By analyzing ten representative titles, a formal taxonomy emerges that categorizes these games based on their core mechanical foundations, including resource management systems, turn structures, and win conditions. This framework distinguishes between various design approaches, such as the implementation of hard versus soft class systems and the divergence between combat-centric and conquest-based objectives.

The primary objective of this classification is to provide a standardized foundation for developing game-agnostic AI agents capable of navigating the high-dimensional decision spaces inherent in the genre. The analysis reveals that while individual titles exhibit significant heterogeneity, they share fundamental structural commonalities that allow for a unified analytical approach. Computational complexity within these games is directly correlated with the density of distinct mechanical elements, such as specific card abilities and drafting formats, which dictate the difficulty of the decision-making problems faced by both human players and autonomous systems.

This research establishes a comprehensive baseline for understanding the technical requirements of CCG gameplay across the modern industry. By formalizing the rules and mechanics that define these interactions, the study highlights the specific challenges that must be addressed to advance machine learning in strategy-based environments. Future efforts in this field will focus on mapping these identified mechanical variables to the specific computational problems encountered during active play, ultimately facilitating the creation of more robust and adaptable AI architectures.