The decision to drink and drive may involve the subjective process of weighing the perceived risks of adverse consequences of the behaviour against perceived utility components. Deterrence theory proposes that an individual will refrain from drink-driving if the perceived chances of experiencing negative outcomes associated with the behaviour are high. Previous research has demonstrated that there may be a mechanism of judgmental bias which influences individuals' perceived probabilities of rarely occurring events. In general driver age and gender have been shown independently to be both indicative and non-indicative for the operation of the bias in terms of subjective perceptions of driver skill, safety and accident involvement. Little evidence has been presented to describe the nature of the bias in the specific domain of drink-driving. Responses from more than 1000 UK drivers were examined to establish whether a system of bias operated for judgments of the likelihood of experiencing several possible adverse consequences of drink-driving across males/females, age groups and offenders/non-offenders. In general drivers were found to perceive themselves as less likely than the average driver to be accident involved while impaired and non-impaired by alcohol. Drivers reporting previous drink-driving behaviour indicated the greatest bias. The age and gender of the driver, independently and in interaction, were not shown to be important in the operation of these perceptual inconsistencies. Findings are discussed for implications of deterrence based drink-driving counter-measures.