History shows that presidential candidates often win their home states by large margins. This phenomenon, which is sometimes called localism, has attracted the attention of scholars for decades. In this study, we improve upon previous localism research in several significant ways. Most important, we challenge the traditional conceptualization of the home state effect, arguing that a candidate's national showing is best thought of as a determinant of the effect instead of part of it, as has often been the case. We also devise three other new hypotheses to account for variation in the home state effect. All of these hypotheses find support in our empirical analyses, providing the most complete understanding of the home state effect to date.