Abstract

Fantasy games, in which players compete to correctly predict realworld outcomes in sports, entertainment, and politics, have grown in popularity and now represent a significant portion of online gaming. Pick’em pools, also known as office pools, are a fantasy game specifically focused on tournament-style competitions such as the “March Madness” NCAA basketball championship. Pick’em pool players often spend significant time trying to understand the current state of competition and to anticipate future events that may significantly affect their performance within the pool. Unfortunately, the combinatorial nature of the outcome space makes these tasks extremely challenging, and intuition is often a highly inaccurate guide. In this paper we present iSee, a system that allows players to make these complex calculations and inferences. We describe a variety of interface options for the interactive presentation of tournament outcome visualizations. We also describe in detail the implementation of a set of algorithms for reliably projecting player performance and distilling the complex outcome space to a number of key scenarios. Finally, we report on a pilot study soliciting user feedback on the system.