Purpose <p>Hip arthroscopy is technically demanding, with complications concentrated among inexperienced surgeons and limited structured opportunities for pre-clinical skill acquisition. This pilot study quantified changes in objective simulator performance after a two-day non-immersive virtual reality training programme.</p> Methods <p>15 male orthopaedic residents (PGY1–PGY3) at a single academic centre, none with prior hip arthroscopy or simulator experience, completed a 12-min baseline assessment on the simulator, two days of training (one hour Fundamentals of Arthroscopic Surgical Training plus one hour hip-specific) using different cases, and an identical post-training assessment. Five simulator-derived metrics were analysed: targets attained, time per target, camera–tissue collisions, femoral head scratches, and probe steadiness. Wilcoxon signed-rank tests with Bonferroni correction (α = 0.01) and effect sizes (r) were applied.</p> Results <p>All 15 participants completed training. Significant improvement occurred across all five metrics with very large effect sizes (r = 0.88–0.89) surviving Bonferroni correction. Median targets attained increased from 1.0 to 4.0 (of 12; p &lt; 0.001); time per target decreased 50% (n = 11; p = 0.003); collisions decreased 42% (p &lt; 0.001); scratches decreased 50% (p &lt; 0.001); probe steadiness increased from 9 to 37% (p &lt; 0.001). Every participant improved on every paired metric.</p> Conclusion <p>A two-day non-immersive virtual reality training programme produced consistent improvement in simulator performance across efficiency, precision, and safety-related domains, with every participant improving. These findings support further evaluation of brief simulator-based training as an adjunct to early orthopaedic training, pending assessment of retention and operative transfer.</p> Graphical Abstract <p></p>