Instrument-based biomechanical movement analysis is an effective injury screening method but relies on expensive equipment and time-consuming analysis. Screening methods that rely on visual inspection and perceptual skill for prognosticating injury risk provide an alternative approach that can significantly reduce cost and time. However, substantial individual differences exist in skill when estimating injury risk performance via observation. The underlying perceptual-cognitive mechanisms of injury risk identification were explored to better understand the nature of this skill and provide a foundation for improving performance. Quantitative structural and process modeling of risk estimation indicated that superior performance was largely mediated by specific strategies and skills (e.g., irrelevant information reduction), and independent of domain-general cognitive abilities (e.g., mental rotation, general decision skill). These cognitive models suggest that injury prediction expertise (i.e., ACL-IQ) is a trainable skill, and provide a foundation for future research and applications in training, decision support, and ultimately clinical screening investigations.