Data Science and Analytics for Esports

Abstract

The use of analytics in professional sports is widespread and rapidly increasing. Similarly, there is a need for analytics in the emerging area of esports, or professional video gaming. Counter-Strike: Global Offensive, widely known as CS: GO, is one of the most popular esports with over forty million copies sold, yet it has lacking analytics. This impedes simple and efficient evaluation of competitive CS: GO matches, player performance, and team performance, which is critical to teams, bettors, media, and fans. The data for each CS: GO match, which includes both player actions and non-player events, is stored in a demofile. A demofile is a recording of the match generated by CS: GO that stores the data as a text of sequential sets of events with no contextualization of information. In order to perform analytics on the stored data, it must be modeled into organized data structures. The data parser developed by Dr. Claudio Silva and Peter Xenopoulos parses the data into Pandas DataFrames, which are spreadsheet-like data structures with rows and columns. Using the data parser, we introduce an analytics package consisting of (1) generalized functions to allow for the efficient filtering and aggregation of CS: GO data; and (2) specialized functions to allow for the efficient calculation of CS: GO statistics. The analytics package has been incorporated into a public software library and commercialized, with professionals and the worldwide CS: GO community currently using it.