This study attempted to clarify the factors responsible for the change in hand propulsive force when stroke frequency was changed in the crawl. Eight male swimmers performed five 20-m front crawls. The first trial involved maximal effort, and then the stroke frequency was controlled during the next 4 trials. The instructed stroke frequencies were 70%, 80%, 90%, and 100% of the stroke frequency in the maximal effort trial. To calculate the hand resultant force, the pressure acting on the palm and the dorsum of the hand were measured, and the product of the pressure difference and the hand plane area was defined as the hand resultant force. Since the hand propulsive force is the propulsion component of the hand resultant force, its value was calculated from the normal vector of the hand plane and its unit vector by three-dimensional motion analysis. Additionally, the propulsion ratio was calculated to show how effectively the hand resultant force acted as the hand propulsive force. These variables were compared with the average for 1 stroke cycle and each movement phase, respectively (glide, pull, and push). The results indicated that as the stroke frequency increased, the dorsum pressure value increased significantly to negative (η2 = 0.82, p <0.001). Also, an increase in the pressure difference between the palm and the dorsum resulted in an increase in the hand resultant force. In the push phase, the increase in the hand resultant force was directly related to the increase in the hand propulsive force. In the glide phase, not only the fluid force, but also the propulsion ratio increased. This was assumed to be because the ratio of the dorsal plane to the propulsion direction was increased while the hand was moving in a vertical direction and a lateral direction in addition to a forward direction. This was considered to allow the fluid force to act more effectively in the propulsive direction, leading to an increase in the hand propulsive force.