Evaluation of safety benefits is an essential part during design and development of pedestrian protection systems. Translating physics into human benefits for large-scale simulations of the benefit requires reliable and validated injury and fatality models. To this end, multivariate predictive models for pedestrian fatalities and different injury severities by means of the ISS as well as the MAIS scale are developed using the US Pedestrian Crash Data Study (PCDS). In addition to impact speed, significant multivariate predictors include physiological and vehicle characteristics. The in-sample as well as out-of-sample predictive quality is remarkably high. An approach to define a metric suitable for comparing active and passive safety is presented. As active safety is capable of influencing impact speed directly, the benefits of a reduction of impact speed regarding injury probability / mortality while controlling other influencing factors is computed. The relative reduction of injury probability / mortality with respect to exposure seems to be an appropriate metric to evaluate and compare safety benefits of measures considering both active and passive safety.