The objective of the present study was to develop a deterministic simulation model for evaluating the reproductive performance of Thoroughbred mares. As an application, the model was used to estimate the herd level asymptotic foaling percentage (AFP) for evaluating the performance of mares in stabilized mare herds using the convergent method for estimating a steady-state distribution of mares. In this model, it was assumed that the mares were mated only during the breeding season. The effects of early pregnancy loss and fetal loss on reproductive performance were investigated. The sensitivities of AFP to changes in conception rate (CR), early pregnancy loss rate (EPLR), fetal loss rate (FLR), and postpartum mating interval (PMI) were examined. Furthermore, the AFPs were compared for two types of postpartum first mating schedules: mating during the first postpartum estrus cycle (foal heat) and first mating during the subsequent cycle. The results indicated that AFP was sensitive to changes in EPLR, FLR, and CR. The comparison of the two types of postpartum first mating schedules showed that AFP after foal heat mating was higher than after the subsequent estrus mating even with the decrease in reproductive parameters caused by foal heat mating in the model. In addition, even if mares did not mate during foal heat, AFP could be improved by shortening the estrus cycle with the administration of PGF 2 α ., Competing Interests: Declaration of competing interest None., (Copyright © 2020 Elsevier Inc. All rights reserved.)