This article presents conclusions from a study of 1,400 criminal trials in cases involving anti-caste and tribe violence, filed in the trial court of Aurangabad District (Sambhajinagar) in western India. Employing computational analysis upon the metadata and text of decisions filed under the Scheduled Castes and Scheduled Tribes (Prevention of Atrocities) (POA) Act 1989 during a 20-year period (2002–2023), I offer an exploratory socio-legal analysis of hate-crime adjudication in India. This article sheds light on hitherto unexplored questions about criminal adjudication in India such as: Which cases are most likely to result in acquittals and convictions? What kind of evidence does the court consider and in what manner is evidence read? How frequently are cases dismissed due to witnesses turning hostile? What kinds of punishments do Indian courts hand out? How long do criminal trials typically last? What kind of charges are most commonly brought against the accused in these cases? How easy is it to get bail and what conditions do courts typically impose? How long do POA-accused spend in jail before they get bail?. In doing so, this article offers a first-of-its kind phenomenology of ‘hate-crime’ adjudication and, more broadly, of criminal adjudication in India.