Basketview: interactive visualization of NBA games

Big data analysis is revolutionizing professional basketball, influencing everything from game strategy to player conditioning. To help understand the underlying information, data visualization techniques are used to display data in a straight forward manner. However, most existing visualizations of basketball games are static and often targets only a certain dimension of the game. In this paper, we present a multifaceted web-based NBA data visualization system, BasketView that enables basketball professionals to analyze the game from three perspectives: season level, game level, and event level. Season level view adopts a novel visualization called BandView, where a combination of bar graph and flow graph is used to display the contribution value of different stat-lines. In game level view, two visualizations, PlayerStream and GameClock, are coupled together to reveal the changing contributions of all players during a basketball game, and the all events that take place. To complement the game level view, event level view uses an animated visualization called CourtChart to simulate the movement of on-court players and basketball. Case study and user feedback are provided to validate the effectiveness of BasketView. In sum, BasketView gives users an easy-to-use and comprehensive toolbox to analyze basketball games.