'Alcohol-fuelled violence' and its prevention has been the subject of recent intense policy debate in Australia, with the content of this debate informed by a surprisingly narrow range of research resources. In particular, given the well-established relationship between masculinities and violence, the meagre attention paid to the role of gender in alcohol research and policy recommendations stands out as a critical issue. In this article, which draws on recent work in feminist science studies and science and technology studies, we focus on the treatment of gender, alcohol and violence in Australian research on 'alcohol-related presentations' to emergency departments (EDs), analysing this type of research because of its prominence in policy debates. We focus on four types of 'gendering practice' through which research genders 'alcohol-related presentations' to EDs: omitting gender from consideration, ignoring clearly gendered data when making gender-neutral policy recommendations, methodologically designing out gender and addressing gender in terms of risk and vulnerability. We argue that ED research practices and their policy recommendations reproduce normative understandings of alcohol's effects and of the operations of gender in social arrangements, thereby contributing to the 'evidence base' supporting unfair policy responses.