The life cycle of land plants alternates between multicellular gametophyte (1N) and sporophyte (2N) life stages. Land plants vary widely in the degree to which one life stage is dominant over the other. In homosporous ferns, both stages are free-living (i.e. the lifecycle is biphasic) and therefore both stages are subjected to environmental conditions that exert selection pressure on the genome. Transcriptome data suggests a high degree of overlap in the identity of genes expressed in haploid and diploid life stages, which introduces many gene interactions between life stages. In theory, there is a high potential for gene conflict, as beneficial mutations in one life stage are likely to be deleterious in the other. Such conflicts can alter predicted patterns of genome evolution, but little is known about how variation in life history traits among biphasic stages affects evolutionary phenomena, such as the fixation rates of beneficial alleles. Emerging data suggests that haploid selection in primarily diploid organisms is more prevalent than previously thought, even in organisms which have a reduced or ephemeral haploid stage (e.g. animals, seed plants). Studying the dynamics of genome evolution in organisms with more balanced biphasic life cycles (e.g. ferns), where gene conflict should be more extreme, may provide insights that are applicable to many sexually reproducing organisms. Here, we use the Python program shadie to simulate genome evolution in the presence of gene conflict under models of fern-like life cycles and contrast our findings with traditional theoretical expectations. We characterize the effects of gene conflict on rates of fixation of beneficial alleles, purging of deleterious alleles, and the effects of linked selection on genome diversity. We also examine how differences in demographic parameters (e.g., population size, cloning, selfing) among diploid and haploid life stages affects these patterns. Finally, we search for signatures in simulated genomes characteristic of genetic conflict which may be used to identify conflicts in empirical data.