Introduction: Ballistocardiography (BCG) captures cardiac- and blood flow-induced body vibrations. Aortic valve stenosis (AS) leads to prolonged left ventricular ejection time (LVET) for a similar stroke volume (SV), potentially altering BCG signals. However, the physiological mechanisms underlying disease-related BCG alterations remain insufficiently understood. This study aimed to characterize linear and rotational BCG (BCGlin and BCGrot) systolic metrics in AS and to identify echocardiographic descriptors linked to BCG signals. Methods: Fifty subjects were prospectively included: 30 patients with AS and 20 controls with normal valves. Sixteen AS patients underwent transcatheter aortic valve implantation (TAVI) with repeated post-procedural measurements. Echocardiographic measures, including SV and LVET, were followed by a five-minute supine recording of BCGlin and BCGrot, synchronously acquired with ECG for cardiac timing. Mean transvalvular aortic flow rate was estimated as SV / LVET. Mean BCG systolic kinetic energy (KQT) was computed from signal amplitude across the three orthogonal BCG components as the integral of kinetic energy over the QT interval, accounting for the inertial parameters of each participant and normalized by QT. Group comparisons used independent and paired statistical tests, and correlations were assessed using cross-sectional and repeated-measures analyses. Results: LVET was significantly longer in AS versus controls (p=0.013) and shortened after TAVI (p<0.001), while SV did not differ between groups or pre/post TAVI. BCGrot KQT was significantly reduced in AS versus controls (p=0.014), whereas BCGlin KQT did not differ between groups. After TAVI, both rotational and linear BCG KQT increased (p=0.004 and p=0.043, respectively). In TAVI patients, intra-individual changes in flow rate correlated with changes in BCGlin KQT (rm=0.63, p<0.01), with a stronger association observed for the cranio-caudal component (r=0.72, p<0.001), but not with BCGrot KQT. Conclusion: Reduced BCG KQT characterize AS, with linear BCG components reflecting reduced transvalvular flow rate, while rotational components capture additional disease related mechanical alterations.