Sprinting is an integral component of sporting performance, and therefore a large body of experimental biomechanics studies have been carried out to identify the kinematics- and kinetics-based variables linked with improved performance. However, to date, very few studies have been performed to assess the influence of technique modifications on sprinting performance, potentially due to the difficulties in doing so using experimental approaches. Furthermore, sprinting, as with many sporting skills, has coaching frameworks that coaches follow when developing athletes, although the recommendations proposed by such frameworks have received minimal support from a scientific perspective. The purpose of this thesis was therefore to develop a computational modelling and simulation framework for sprinting to explore potential performance-enhancing modifications to technique and to assess how they compare to the recommendations proposed by a prevalent coaching framework (front-side mechanics). The first investigation of this thesis featured the development of a computational modelling and simulation framework for sprinting and evaluating its ability to reproduce experimental data for different sprinting phases. The evaluation step was carried out by performing a series of data-tracking calibration and validation simulations. The data-tracking simulations also enabled dynamically consistent simulated outputs to be obtained and footground contact model parameters to be identified. The simulated outputs from the validation simulation were found to be in good agreement with the experimental data, with average root mean squared differences (RMSDs) less than 1.0° and 0.3 cm for the rotational and translational kinematics, respectively. The anterior-posterior ground reaction force component had the largest percentage RMSD (11.4%). The second study of this thesis explored how hypothetical modifications to technique affect accelerative sprinting performance by performing a series of (data-tracking and predictive) simulations using the framework developed in the first study. Technique modifications were explored through enabling either individual or combinations of the net lower-limb flexor-extensor joint moments (ankle, knee, and hip) to freely vary within the predictive simulations, whilst the remaining net joint moments were tracked (established from performing a data-tracking simulation). The 'kneefree' simulations led to the greatest improvements to overall performance (22.0%; 1401.2 vs. 1148.7 W) due to modifying the timing and magnitude of the net knee flexor-extensor moments. The kinematics aspects of the front-side mechanics coaching framework were not found to emerge from the performance-enhancing predictive simulations.