Traditional performance assessment techniques utilize a top-down approach focused on mission outcomes (i.e., “what” occurred). Unfortunately, these techniques often fail to uncover “why” the outcomes occurred, which is critical information for training and resourcing decision makers. An alternative approach that focuses on the factors that directly affect mission outcomes (i.e., individual skill execution coupled with systems reliability), has traditionally been deemed “impossible” due to the fidelity of data required. However, in 2015, a team of former TOPGUN instructors in support of NAVAIR and CNAF, overcame significant challenges (e.g., disparate data sources, historically subjective/self-reported assessments, etc.) to develop and apply a methodology that links the performance demonstrated by F/A-18 aircrew at the TTP level of execution to team/mission outcomes during live training events. Over the last five years, this new approach to data collection and analysis has been developed into an automated toolset and applied to hundreds of F/A-18 training events. The results provide unprecedented insight into the individual skill and aircraft systems deficiencies that have the greatest impact on F/A-18 mission outcomes. The data collected during these exercises also enables the application of groundbreaking AI/ML data analytics to derive standards for individual performance that directly link to desired combat outcomes. In addition to providing invaluable feedback to the F/A-18 aircrew, CNAF is currently relying on this effort to inform decisions ranging from syllabi improvements to resourcing investments, which proves the utility of this new approach. Although initially demonstrated in the context of live F/A-18 air warfare, the process has now been demonstrated across multiple platforms (e.g., E-2C/D, EA-18G, etc.), mission sets (e.g., STW, C2, AEA, etc.), and training environments (i.e., live, virtual/constructive, and LVC), and is scalable to other disciplines. This skills-based approach to performance assessment, once deemed impossible to achieve, represents the future of automated performance assessment.