INTRODUCTION Successful completion of the MRCS examination is mandatory for surgical trainees in the UK. The MRCS is attempted early in surgical training, before applying for competitive entry to specialty training posts. The aim of this study was to establish whether MRCS performance can predict future surgical specialty destination and whether those who perform well at the exam enter more competitive specialties. METHODS UK candidates who attempted the MRCS examination between 2007 and 2020 were matched to career outcomes data using the General Medical Council database (n=2,910). Specialty competition ratios were calculated from national recruitment between 2017 and 2019. Chi-squared tests determined associations with first-attempt MRCS performance. Multinomial regression models were developed to establish whether MRCS results can predict surgical specialty destination. RESULTS There was a statistically significant association between MRCS Part A first-attempt pass rates and training specialty (p=0.001) but not for Part B (p=0.655). Candidates who passed Part A at the first attempt were more likely to enter neurosurgical training (odds ratio [OR]: 1.75, 95% confidence interval [CI]: 1.18–2.61), cardiothoracic training (OR: 2.25, 95% CI: 1.30–3.90), plastic surgery training (OR: 1.75, 95% CI: 1.20–2.56), paediatric surgery training (OR: 2.01, 95% CI: 1.10–3.67) and general surgical training (OR: 1.48, 95% CI: 1.18–1.86) than trauma and orthopaedic training. MRCS Part B performance was not an independent predictor of surgical specialty. CONCLUSIONS Trainees who perform well at MRCS Part A are more likely to enter more competitive specialties. Further work is required to investigate the relationships between this finding, specialty application and selection patterns, and what surgical trainees value in their career decision making.