At a high level, sport analytics systems can be broken into two components: sensory data collection and analytical models that turn sensory data into insights for users. In this talk, we will focus on the latter, and more specifically on the challenges we’ve encountered in adapting advanced analytics originally developed for professional leagues to create a new product for use in a new market—youth sports. These challenges arise due to the unfamiliarity of end users with sophisticated analytical metrics, incomplete and partially accurate underlying visual data and the inherent limitations of vision-based data collection systems. We will explain how we use a combination of vision systems, generative models and reinforcement learning techniques to develop compelling products for youth sports, and share what we learned in this process.