Myeloproliferative neoplasms (MPNs) are clonal hematological diseases in which cells of the myelo-erythroid lineage are overproduced and patients are predisposed to leukemic transformation. Hematopoietic stem cells are the suspected disease-initiating cells, and these cells must acquire a clonal advantage relative to nonmutant hematopoietic stem cells to perpetuate disease. In 2005, several groups identified a single gain-of-function point mutation in JAK2 that associated with the majority of MPNs, and subsequent studies have led to a comprehensive understanding of the mutational landscape in MPNs. However, confusion still exists as to how a single genetic aberration can be associated with multiple distinct disease entities. Many explanations have been proposed, including JAK2V617F homozygosity, individual patient heterogeneity, and the differential regulation of downstream JAK2 signaling pathways. Several groups have made knock-in mouse models expressing JAK2V617F and have observed divergent phenotypes, each recapitulating some aspects of disease. Intriguingly, most of these models do not observe a strong hematopoietic stem cell self-renewal advantage compared with wild-type littermate controls, raising the question of how a clonal advantage is established in patients with MPNs. This review summarizes the current molecular understanding of MPNs and the diversity of disease phenotypes and proposes that the increased proliferation induced by JAK2V617F applies a selection pressure on the mutant clone that results in highly diverse clonal evolution in individuals.