The orientation of the ground reaction force (GRF) vector is a key determinant of human sprint acceleration performance and has been described using ratio of forces (RF) which quantifies the ratio of the antero-posterior component to the resultant GRF. Different methods have previously been used to calculate step-averaged RF, and this study therefore aimed to compare the effects of three calculation methods on two key “technical” ability measures: decline in ratio of forces (DRF) and theoretical maximal RF at null velocity (RF0). Twenty-four male sprinters completed maximal effort 60 m sprints from block and standing starts on a fully instrumented track (force platforms in series). RF-horizontal velocity profiles were determined from the measured GRFs over the entire acceleration phase using three different calculation methods for obtaining an RF value for each step: A) the mean of instantaneous RF during stance, B) the step-averaged antero-posterior component divided by the step-averaged resultant GRF, C) the step-averaged antero-posterior component divided by the resultant of the step-averaged antero-posterior and vertical components. Method A led to significantly greater RF0 and shallower DRF slopes than Methods B and C. These differences were very large (Effect size Cohen’s d = 2.06 – 4.04) and varied between individuals due to differences in the GRF profiles, particularly during late stance as the acceleration phase progressed. Method B provides RF values which most closely approximate the mechanical reality of step averaged accelerations progressively approaching zero and it is recommended for future analyses although it should be considered a ratio of impulses.