Effective robotic surgical training requires accurate, efficient performance assessment. It is widely recognised that the early stages of a surgeon’s learning curve are associated with poorer outcomes and higher complications. Current manual skills assessment methods by expert surgeons are time consuming and difficult to implement in clinical practice. Automated skills assessment has been shown to be feasible using kinematic data derived from surgical robotic systems (automated performance metrics (APM)). Models leveraging APM during robotic prostate cancer procedures can predicted surgical proficiency and also correlate intraoperative surgical performance with postoperative functional outcomes. However, these approaches have limitations. Current approaches to performance assessment are procedure-specific and overlook the progressive evolution of surgical skill over time. Assessment of surgeon hand motion is a novel approach to analyse surgical performance particularly in robotic surgery. Accurate prediction of surgical motion, and performance offer direct implications for training enhancement and human-robotic interactions.