Abstract Background First-phase ejection fraction (EF1; the ejection fraction measured during active systole up to the time of maximal aortic flow) measured by transthoracic echocardiography (TTE) is a powerful predictor of outcomes in patients with aortic stenosis. We aimed to assess whether cardiovascular magnetic resonance (CMR) might provide more precise measurements of EF1 than TTE and to examine the correlation of CMR EF1 with measures of fibrosis. Methods In 141 patients with at least mild aortic stenosis, we measured CMR EF1 from a short-axis 3D stack and compared its variability with TTE EF1, and its associations with myocardial fibrosis and clinical outcome (aortic valve replacement (AVR) or death). Results Intra- and inter-observer variation of CMR EF1 (standard deviations of differences within and between observers of 2.3% and 2.5% units respectively) was approximately 50% that of TTE EF1. CMR EF1 was strongly predictive of AVR or death. On multivariable Cox proportional hazards analysis, the hazard ratio for CMR EF1 was 0.93 (95% confidence interval 0.89–0.97, p = 0.001) per % change in EF1 and, apart from aortic valve gradient, CMR EF1 was the only imaging or biochemical measure independently predictive of outcome. Indexed extracellular volume was associated with AVR or death, but not after adjusting for EF1. Conclusions EF1 is a simple robust marker of early left ventricular impairment that can be precisely measured by CMR and predicts outcome in aortic stenosis. Its measurement by CMR is more reproducible than that by TTE and may facilitate left ventricular structure–function analysis.