Do Justices telegraph their preferences during oral arguments? We demonstrate that Justices implicitly reveal their leanings during oral arguments, even before arguments and deliberations have concluded. Specifically, we extract the emotional content of over 3,000 hours of audio recordings spanning 30 years of oral arguments before the Court. Using only the level of emotional arousal in each of the Justices’ voices during these arguments, as measured by their vocal pitch, we are able to accurately predict many of their eventual votes, while using none of the text or substantive content. These predictions are statistically and practically significant and robust to including a range of controls. Our findings suggest that mannerisms that may be subconscious, such as vocal pitch, carry information that basic legal, political, and textual information do not, and can be used to predict the decisions of even elite political actors.