Bet-hedging via polyandry (spreading the extinction risk of the female’s lineage over multiple males) may explain the evolution of female multiple mating, which is found in a wide range of animal and plant taxa. This hypothesis posits that females can increase their fitness via polyandrous mating when “unsuitable” males (i.e., males causing reproductive failure for various reasons) are frequent in the population and females cannot discriminate such unsuitable mates. Although recent theoretical studies have shown that polyandry can operate as a bet-hedging strategy, empirical tests are scarce. In the present study, we tested the bet-hedging polyandry hypothesis by using the red flour beetle Tribolium castaneum. We compared female reproductive success between monandry and polyandry treatments when females mated with males randomly collected from an experimental population, including 20% irradiated (infertile) males. In addition, we evaluated geometric mean fitness across multiple generations as the index of adaptability of bet-hedging traits. Polyandrous females showed a significantly higher egg hatching rate and higher geometric mean fitness than monandrous females. These results strongly support the bet-hedging polyandry hypothesis.