A model-based method is developed to predict probabilistic margins to safety limits for passively safe reactors where the same physical mechanisms that control reactor behavior at power also control off-normal response. Model parameter values are estimated using the maximum likelihood method from the plant response to perturbations of flow, temperature, and rod reactivity applied during normal operation. The resulting model can be used to predict plant response to upsets and provide a probabilistic measure of how closely safety limits would be approached. The method is applied to the Integral Fast Reactor.