Behavior modeling is an important area of research as it severs many applications and disciplines; it expands on our knowledge of human behavior as well as allows us to develop novel AI entities that adapts to its users or mimics them. This area of research is by definition interdisciplinary and has stimulated interest from multiple funding organizations. In the past few years, games have emerged as a new platform with large context-rich data that has the utility to advance behavior modeling. In particular, recently there has been much work investigating the use of MOBA games, such as League of Legends and Dota, and Massively Multiplayer games, such as Eve Online, for this purpose. Even though there are many exciting new results, the current techniques and methods for modeling player behaviors within these environments are still in their infancy. In this talk, I will discuss our current work towards developing a methodological platform to develop player models that leverages and builds on current theory and data-driven techniques, including machine learning and visualization methods.