Purpose - Medial patellofemoral ligament (MPFL) reconstruction for recurrent patellar instability has gained popularity, and anatomical and biomechanical studies have recently altered our operative techniques. The aim of this study was to report the clinical outcome of this new anatomical MPFL reconstructive technique and investigate whether correlating factors could be identified. Methods - Between 2009 and 2012, a total of 31 consecutive patients underwent MPFL reconstruction using an autologous gracilis graft and anatomical tunnel placement. Pre- and post-operative data were collected as a part of routine clinical practice. The preoperative assessment included a rotational profile CT scan of the lower extremity according to the Lyon protocol with TT–TG distance measurement. Outcomes were evaluated with the Kujala and Norwich patella instability (NPI) scores preoperatively and at follow-up (1.5–5.1 years). Results - A significant improvement in both the Kujala (p <0.001) and NPI (p = 0.012) scores was recorded. A medium and large negative correlations were found between TT–TG distance and Kujala score improvement (ρ = −0.48, p = 0.020) and NPI score improvement (ρ = −0.83, p = 0.042), respectively. Multiple regression analysis identified TT–TG distance, Beighton score and BMI as factors explaining the variance of Kujala score improvement. Conclusion - Anatomical MPFL reconstruction with the gracilis autograft for patellar instability resulted in good outcome. This underlines the importance of anatomical tunnel placement in MPFL reconstruction. With a precise preoperative work-up, factors can be identified that may guide selecting the optimal operative strategy and improve counselling of the patient.