Rituals shape many aspects of our lives, and they are no less common in scientific research than elsewhere. One that figures prominently in hypothesis testing is the null ritual, the pitting of hypotheses against chance. Although known to be problematic, this practice is still widely used. One way to resist the lure of the null ritual is to increase the precision of theories by casting them as formal models. These can be tested against each other, instead of against chance, which in turn enables a researcher to decide between competing theories based on quantitative measures. This article gives an overview of the advantages of modeling, describes research that is based on it, outlines the difficulties associated with model testing, and summarizes some of the solutions for dealing with these difficulties. Pointers to resources for teaching modeling in university classes are provided.