Traditional race categories are not scientific hypotheses, and geneticists should, therefore, proceed with caution when discussing the relationship between racial categorizations and patterns of genetic variability. Nonetheless, we might ask if traditional racial categorizations are useful abstractions in genetic research. I argue that there are at least two reasons why they are not. First, decades of research have shown that human genetic variation is clinal in nature so discrete categories are not particularly useful for describing this variation. More importantly, if humans are forced into discrete categories using genetic data, the categories that emerge, when using representative panels, do not coincide with the traditional race categories. I warn against redefining race so that it may better fit observed patterns of genetic variability, because of the potential to mislead the public discourse on race and racialization.