The projected discovery and exclusion capabilities of particle physics and astrophysics/cosmology experiments are often quantified using the median expected p-value or its corresponding significance. We argue that this criterion leads to flawed results, which for example can counterintuitively project lessened sensitivities if the experiment takes more data or reduces its background. We advocate for standard use of the “exact Asimov significance” detailed in this talk. We then consider a modified frequentist approach, which for exclusion case is known as the CLs technique, for projecting sensitivities, and contrast it with the standard frequentist approach among others. And, as an application of some of these techniques, we study the statistical significance of proton decay at the current and future neutrino detectors. We first review some statistical methods that are employed in various proton decay sensitivity studies, and along the way we estimate the current lower limit on proton partial lifetimes at various confidence levels based on Super-Kamiokande’s data. We present our projections for proton partial lifetime in p → ν K+ and p → e+ π0 decay channels at DUNE, JUNO, and Hyper-Kamiokande.