This paper shows how a system of ordinary differential equations describing the evolution of the anaerobic energy, the oxygen uptake, the propulsive force and the velocity of a runner accurately describes pacing strategy. We find a protocol to identify the physiological parameters needed in the model using numerical simulations and time splits measurements for an 80 m and a 1600 m race. The velocity curve of the simulations is very close to the experimental one. This model could allow to study the influence of training and improving some specific parameters for the pacing strategy.