Individuals responsible for decision-making during critical incidents must wrestle with uncertainty, complexity, time pressure, and accountability. Critical incidents are defined as rare events where demand outstrips resources and where there are high stakes, uncertainty, and dynamic and ever-shifting elements that frustrate clear predictions. This paper argues that critical-incident decision-making is highly complex because many critical incidents have no such analogue, and thus there is no prior experience to draw upon. Further, while prescriptive models argue for a selection of a “best” outcome, rarely in critical incidents is there a “best” outcome and, instead, more likely a “least-worst” one. Most options are high risk, most will carry negative consequences, and many will be immutable and irreversible once committed to. This paper analyzes data collected from critical decision method interviews with members of the United States Armed Forces to explore the psychological processes of making (or not making) least-worst decisions in high-consequence situations. Specifically, and based on thematic analysis of interviews with those who have made least-worst decisions while serving as part of the Armed Forces, we identify a host of exogenous (external to the incident such as resources, political agendas) and endogenous factors (features of the incident itself—size, scale, duration) that affect the decision-making process. These factors have, to date, not been factored into theoretical models of how high-stakes decisions are made under conditions of uncertainty.