Several hypotheses have been put forward to explain the evolution of extreme sexual size dimorphism (SSD). Among them, the gravity hypothesis (GH) explains that extreme SSD has evolved in spiders because smaller males have a mating or survival advantage by climbing faster. However, few studies have supported this hypothesis thus far. Using a wide span of spider body sizes, we show that there is an optimal body size (7.4 mm) for climbing and that extreme SSD evolves only in spiders that: (1) live in high-habitat patches and (2) in which females are larger than the optimal size. We report that the evidence for the GH across studies depends on whether the body size of individuals expands beyond the optimal climbing size. We also present an ad hoc biomechanical model that shows how the higher stride frequency of small animals predicts an optimal body size for climbing.