The force-vector theory contends that horizontal exercises are more specific to horizontal sports skills. In this context, the focus is on horizontal force production relative to the global coordinate frame. However, according to the principle of dynamic correspondence, the direction of force relative to the athlete is more important, and thus the basis for the force-vector theory is flawed. The purpose of this study was therefore to test the force-vector theory. According to the force-vector theory, hip thrust is a horizontally loaded exercise, and so hip thrust training would be expected to create greater improvements in horizontal jump performance than vertical jump performance. Eleven collegiate female athletes aged 18–24 years completed a 14-week hip thrust training programme. Pre and post testing was used to measure the following: vertical squat jump, vertical countermovement jump, horizontal squat jump, horizontal countermovement jump and hip thrust 3 repetition maximum (3RM). Subjects improved their 3 repetition maximum hip thrust performance by 33.0% (d = 1.399, p < 0.001, η2 = 0.784) and their vertical and horizontal jump performance (improvements ranged from 5.4–7.7%; d = 0.371–0.477, p = 0.004, η2 = 0.585). However, there were no differences in the magnitude of the improvement between horizontal and vertical jumping (p = 0.561, η2 = 0.035). The results of this study are contrary to the predictions of the force-vector theory. Furthermore, this paper concludes with an analysis of the force-vector theory, presenting the mechanical inconsistencies in the theory. Coaches should use the well established principle of dynamic correspondence in order to assess the mechanical similarity of exercises to sports skills.