Contrary to classical sexual selection theories, females of many taxa mate with multiple males during one reproductive cycle. In this study, we conducted an experiment on the "trade-up hypothesis", which proposes that females remate if a subsequently encountered male is potentially superior to previous mates to maximize the genetic quality of their offspring. We presented bank vole females (Clethrionomys glareolus) sequentially with two males of known dominance rank in different orders, i.e., either first subordinate and second dominant, first dominant and second subordinate, or two males that were equal in dominance (high ranking) and observed their mating behavior. We found that 92% of the females mated multiply and did not base their remating decision on male social status. Therefore, polyandry cannot be explained by the "trade-up hypothesis" based on dominance rank in this species. However, we found that dominant males sired significantly more offspring than subordinate males. This varied according to mating order: dominant males sired more offspring when they were second than when they were first. Moreover, litter sizes were significantly smaller when the dominant male was first (smallest relative success of dominant males) compared to litter sizes when mating order was reversed or both males equal in status. Our results suggest that even though multimale mating includes males that are of poorer quality and thus potentially decreases the fitness of offspring, most of a female's offspring are sired by dominant males. Whether this is due to cryptic female choice, sperm competition, or a combination of both, remains to be tested.