Predation and/or parasitism often limits the evolution of conspicuous male traits and female preferences because conspicuous traits can attract predators or parasites and it is costly for females to associate with males that attract predators or parasites. As a result, males and females in high-risk populations are expected to evolve safer mating behaviors compared to individuals from low-risk populations. We tested this antagonistic selection hypothesis in the field cricket Gryllus lineaticeps. Males produce chirped songs, and both female crickets and the eavesdropping parasitoid fly Ormia ochracea prefer faster chirp rates. The flies attack the field crickets late in the breeding season and parasitized crickets die. We used a common garden rearing design to test for evolved differences in songs and preferences between high- and low-risk populations. In contrast to predictions of the antagonistic selection hypothesis, males from high-risk populations produced faster (riskier) chirp rates and females preferred faster chirps. We suggest that late-season parasitism selects for increased investment in reproductive traits to maximize reproduction before the advent of parasitoid activity (“late-season parasitism hypothesis”), which would at least explain riskier female preferences and potentially riskier male songs in the high-risk populations. Predation and parasitism may thus have diverse and unexpected effects on the evolution of reproductive behavior, depending upon the temporal pattern of predator- or parasite-induced mortality. Mating signals are typically conspicuous and not only attract partners but also predators and parasites. Even the silent mating partner may experience predation or parasitism by associating with the signaler. Under these circumstances, it is commonly assumed that natural and sexual selection act in opposite directions, effectively limiting the evolution of conspicuous signals and preferences. We demonstrate that an eavesdropping parasitic fly caused the evolution of preferences, and potentially songs, in a field cricket in the opposite, more conspicuous, direction than predicted by antagonistic selection. We argue that the temporal pattern of parasitism in relation to the reproductive season likely causes this unexpected evolutionary pattern. We propose the late-season parasitism hypothesis as an alternative to the antagonistic selection hypothesis, which might better explain more conspicuous mating trait values in other species that experience seasonal predation or parasitism.