Scientific research is performed by fallible humans. Degrees of freedom in the construction and selection of evidence and hypotheses grant scientists considerable latitude to obtain study outcomes that align more with their preferences than is warranted. This creates a risk of bias and can lead to scientists fooling themselves and fooling others. Preregistration involves archiving study information (e.g., hypotheses, methods, and analyses) in a public registry before data are inspected. This offers two potential benefits: (1) reduce bias by ensuring that research decisions are made independently of study outcomes; and (2) calibrate confidence in research by transparently communicating information about a study’s risk of bias. In this article, we briefly review the historical evolution of preregistration in medicine, psychology, and other domains, clarify its pragmatic functions, discuss relevant meta-research, and provide recommendations for scientists and journal editors.