Data science is moving quickly, and new statistical methods regularly appear in the literature. This month’s feature takes a look at how to tell when an analysis is falling short.