This article considers Defoe's use of statistical data in his historical novel A Journal of the Plague Year, a device generally considered as a means of supplying a work of fiction with verisimilitude. Re-evaluating Defoe's attitude to the science of political arithmetic and the earliest proponents of statistics, it argues that Defoe validates a subjective and novelistic account of the plague over fallacious figures that purport to be hard facts. It therefore contextualizes the emergence of the novel within shifts in epistemology, as certainty was increasingly perceived as unattainable, and probability deemed the best standard for knowledge and action.