Although the majority of animals and plants, including humans, are dominated by the diploid phase of their life cycle, extensive diversity in ploidy level exists among eukaryotes, with some groups being primarily haploid whereas others alternate between haploid and diploid phases. Previous theory has illuminated conditions that favor the evolution of increased or decreased ploidy but has shed little light on which species should be primarily haploid and which primarily diploid. Here, we report a discovery that emerged from host–parasite models in which ploidy levels were allowed to evolve: selection is more likely to favor diploidy in host species and haploidy in parasite species. Essentially, when parasites must evade a host's immune system or defense response, selection favors parasitic individuals that express a narrow array of antigens and elicitors, thus favoring haploid parasites over diploid parasites. Conversely, when hosts must recognize a parasite before mounting a defensive response, selection favors hosts with a broader arsenal of recognition molecules, thus favoring diploid hosts over haploid hosts. These results are consistent with the predominance of haploidy among parasitic protists.