Hibernation has received considerable attention from physiologists and natural historians, but theoretical and ecological treatments of hibernation are rarer. There is ample recent evidence that costs associated with hibernation affect the degree to which hibernation is expressed in nature, but we currently lack a quantitative framework under which to make predictions about how the costs and benefits of hibernation interact under various environmental conditions. Here, we attempt the first steps towards building an optimal hibernation theory for making specific predictions about the expression of hibernation (i.e. the depth and duration of torpor bouts), metabolic functioning, and the total period of hibernation in mammals and birds. Our current understanding of the costs associated with hibernation do not allow for parameterisation of optimal hibernation theory, but we hope this work provides a roadmap for physiologists and ecologists to collect the necessary data in the future. Hibernation was historically caricatured as little more than an energy conservation strategy during winter. This view drove the assumption that all hibernating mammals and birds can maximize fitness by maximising the use of hibernation. However, we now know that relative use of hibernation varies widely among individuals, populations, and species. Here, we provide a theoretical framework to predict, at least in relative terms, what factors should drive the degree to which an individual mammal or bird should express hibernation using classical optimality theory. We then use cave‐hibernating bats as a case study to demonstrate the principles of ‘optimal hibernation theory’.