Social media is the primary platform for discussions and reactions during various social events. Studies in this space focus on the aggregate opinion and sentiment analysis but fail to analyze the micro-dynamics. In this work, we present a case study of the 2016 US Presidential Debates, analyzing the user opinion micro-dynamics across the timeline. We present an opinion variation analysis coupled with micro and macro level user analysis in order to explain opinion change. We also identify and characterize varied user-groups derived through this analyses. We discover that aggregate change in opinion is better explained by the differential influx of polarized population rather than the change in individual’s stance or opinion.