It has recently been argued that virtue ethics cannot accommodate the possibility of supererogation. In response, Rebecca Stangl proposes a neo-Aristotelian account of supererogation that, she argues, generates plausible verdicts, while also being compatible with the doctrine of the mean. I argue that Stangl’s response is unsuccessful. First, I demonstrate that the proposal in its current form is problematically indeterminate, meaning that we cannot know what verdicts would be produced in response to classic examples. Second, I argue that anyone attempting to develop the account faces a dilemma, and that both options for responding to this dilemma generate problematic results.