Plant breeders often practice recurrent selection and other forms of population improvement, inter-mating and selecting material to improve the overall fitness of the population to the environment of interest. This selection shifts favorable alleles, and linked loci, to higher frequency while lowering the frequency of, or losing, unfavorable alleles. Screening these pre- and post-selection populations with molecular markers have allowed investigators to putatively determine which regions of the genome are linked to, or under selection. However, the current tests for such temporal selection - selection over time have been underutilized, difficult to apply, require conflicting estimators of effective population size (Ne), and remain untested for their power to detect selection or false positive rate. Plant breeders often have information on the mating scheme, the number of individuals evaluated and selected for each generation (selection intensity), and remnant seed, which can be used for detecting allele frequencies (via molecular markers) from population improvement projects. This information is more complete and powerful then approximations of Ne and can be used to locate regions under selection that contain or are closely linked to loci of agronomic importance for both target trait improvement and local adaptation. In this study we demonstrate the power of available tests of temporal selection and suggest a new test that uses information commonly available in a breeding program.