Key points Severe aortic valve diseases are common cardiac abnormalities that are associated with poor long‐term survival. Before any reduction in left ventricular (LV) function, the left ventricle undergoes structural remodelling under the influence of changing haemodynamic conditions. In this study, we combined temporal changes in LV structure (volume) with alterations in LV functional characteristics (strain, ԑ) into a ԑ–volume loop, in order to provide novel insight into the haemodynamic cardiac consequences of aortic valve diseases in those with preserved LV ejection fraction. We showed that our novel ԑ–volume loop and the specific loop characteristics provide additional insight into the functional and mechanical haemodynamic consequences of severe aortic valve diseases (with preserved LV ejection fraction). Finally, we showed that the ԑ–volume loop characteristics provide discriminative capacity compared with conventional measures of LV function. Abstract The purpose of this study was to examine left ventricular (LV) strain (ԑ)–volume loops to provide novel insight into the haemodynamic cardiac consequences of aortic valve stenosis (AS) and aortic valve regurgitation (AR). Twenty‐seven participants were retrospectively recruited: AR (n = 7), AS (n = 10) and control subjects (n = 10). Standard transthoracic echocardiography was used to obtain apical four‐chamber images to construct ԑ–volume relationships, which were assessed using the following parameters: early systolic ԑ (ԑ_ES); slope of ԑ–volume relationship during systole (Sslope); end‐systolic peak ԑ (peak ԑ); and diastolic uncoupling (systolic ԑ–diastolic ԑ at same volume) during early diastole (UNCOUP_ED) and late diastole (UNCOUP_LD). Receiver operating characteristic curves were used to determine the ability to detect impaired LV function. Although LV ejection fraction was comparable between groups, longitudinal peak ԑ was reduced compared with control subjects. In contrast, ԑ_ES and Sslope were lower in both pathologies compared with control subejcts (P < 0.01), but also different between AS and AR (P < 0.05). UNCOUP_ED and UNCOUP_LD were significantly higher in both patient groups compared with control subjects (P < 0.05). Receiver operating characteristic curves revealed that loop characteristics (AUC = 0.99, 1.00 and 1.00; all P < 0.01) were better able then peak ԑ (AUC = 0.75, 0.89 and 0.76; P = 0.06, <0.01 and 0.08, respectively) and LV ejection fraction (AUC = 0.56, 0.69 and 0.69; all P > 0.05) to distinguish AS vs control, AR vs control and AS vs AR groups, respectively. Temporal changes in ԑ–volume characteristics provide novel insight into the haemodynamic cardiac impact of AS and AR. Contrary to traditional measures (i.e. ejection fraction, peak ԑ), these novel measures successfully distinguish between the haemodynamic cardiac impact of AS and AR.