Experimental studies have shown that significant carbon exchange occurs between bone-apatite and the pyre atmosphere during cremation, which can cause a calendar date offset between the radiocarbon (14C) event and the date of cremation. There are limited empirical data available to assess the magnitude of such wood-age offsets, but the aim of this paper is to test if they can be modeled statistically. We present new 14C dates on modern bone cremated in realistic open-air experiments and on archaeological samples of cremated bone and associated organic material. Experimental results demonstrate a wide range of carbon exchange with a mean of 58.6 ± 14.8%. Archaeological results indicate that the wood-age offsets have an approximately exponential distribution. We test whether the default Charcoal Outlier_Model in OxCal v4.3, developed to reduce the impact of wood-age offsets in dates of charcoal, is appropriate for cremated bone, but find that it slightly underestimates apparent offsets. To counter the intrinsic age of both pyre fuel and unburned bio-apatite, we instead propose a bespoke Cremation Outlier_Model, which combines an exponential distribution of calendar age offsets with a minimum offset, and provides better estimates of the actual dates of cremations.