The purpose of this study was to determine the discrepancy rates of radiology residents interpreting emergent neck and Circle of Willis magnetic resonance angiography (MRA) studies and to detect any adverse clinical outcomes. Three hundred seventeen MRA studies given preliminary reading by radiology residents were retrospectively reviewed over a 2-year period. Discrepancies were classified as either false negatives (failure to diagnose abnormalities) or false positives (misinterpreting normal scans as abnormal). The overall discrepancy rate was 12.1% for Circle of Willis MRA and 7.9% for neck MRA. Fourth-year residents had the lowest discrepancy rates (7.7%), but this was not statistically significant. The most common misses were stenosis greater than 70% (n = 9) and aneurysm (n = 12). No adverse clinical outcome was detected mainly due to rapid turnaround time for final reporting.