Most performance statistics using aggregate level data about courts in India show delays. There is limited analysis of the actual duration and trajectories of cases. In this paper, we create a de novo data-set using publicly available data on cases at the Indian Income Tax Appellate Tribunal (ITAT). We apply statistical techniques of hazard models to address questions around case duration at the Income Tax Appellate Tribunal (ITAT). We describe patterns in case life-span, compare these patterns among groups, and build statistical models of the risk of case completion over time. We find differences in the probability of case completion between the ITAT benches in Mumbai and Delhi. We also find that probability of case completion differs by case type. Our results point to the need to study case trajectories to better understand the causes of delays in order to design appropriate policy solutions to improve the performance of courts and tribunals.