Hedrick et al (2016) reported on ’negative-assortative mating for color in wolves’ from Yellowstone National Park, the ‘first documented case of significant negative-assortative mating in mammals.’ Here I report a logical inconsistency in their population genetic model that effectively imposes selection against some assortatively mating genotype. After pointing out this inconsistency, I derive new expressions for the frequency of different matings and the equilibrium allele frequencies. Compared to Hedrick et al. (2016), the system rapidly approaches an equilibrium based on observed levels of negative-assortative mating in this population. Hence, the revised model may strengthen the case that assortative mating can maintain polymorphism and provides logically consistent results to inform the study of nonrandom mating in other organisms.