Has the connection between polyploidy and diversification actually been tested?

•Most studies of polyploidization and diversification use indirect methods as evidence for polyploidy, and often cannot capture a reticulate history.•Approaches that use nuclear gene trees to place the polyploidy events are underutilized.•The point of coalescence of ancestral genomes is not necessarily the date of the polyploidization event.•The connection between polyploidy and diversification has not been tested rigorously. Many major clades of angiosperms have several whole genome duplications (polyploidization events) in their distant past, suggesting that polyploidy drives or at least permits diversification. However, data on recently diverged groups are more equivocal, finding little evidence of elevated diversification following polyploidy. The discrepancy may be attributable at least in part to methodology. Many studies use indirect methods, such as chromosome numbers, genome size, and Ks plots, to test polyploidy, although these approaches can be misleading, and often lack sufficient resolution. A direct test of diversification following polyploidy requires a sequence-based approach that traces the history of nuclear genomes rather than species. These methods identify the point of coalescence of ancestral genomes, but may be misleading about the time and thus the extent of diversification. Limitations of existing methods mean that the connection between polyploidy and diversification has not been rigorously tested and remains unknown.