Sport science can be thought of as a scientific process used to guide the practice of sport with the ultimate aim of improving sporting performance. However, despite this goal, the general consensus is that the translation of sport-science research to practice is poor. Furthermore, researchers have been criticised for failing to study problems relevant to practitioners and for disseminating findings that are difficult to implement within a practical setting. This paper proposes that the situation may be improved by the adoption of a model that guides the direction of research required to build our evidence base about how to improve performance. Central to the Applied Research Model for the Sport Sciences (ARMSS) described in this report is the idea that only research leading to practices that can and will be adopted can improve sporting performance. The eight stages of the proposed model are (i) defining the problem; (ii) descriptive research; (iii) predictors of performance; (iv) experimental testing of predictors; (v) determinants of key performance predictors; (vi) efficacy studies; (vii) examination of barriers to uptake; and (viii) implementation studies in a real sporting setting. It is suggested that, from the very inception, researchers need to consider how their research findings might ultimately be adapted to the intended population, in the actual sporting setting, delivered by persons with diverse training and skills, and using the available resources. It is further argued in the model that a greater understanding of the literature and more mechanistic studies are essential to inform subsequent research conducted in real sporting settings. The proposed ARMSS model therefore calls for a fundamental change in the way in which many sport scientists think about the research process. While there is no guarantee that application of this proposed research model will improve actual sports performance, anecdotal evidence suggests that sport-science research is not currently informing sport-science practice as we would hope and that sport-science researchers need to consider a new approach.