Residual transprosthetic gradient (TG) after transcatheter aortic valve replacement (TAVR) with balloon-expandable valves (BEV) may be due to suboptimal valve expansion. To compare hemodynamics after TAVR with small BEV according to postdilation strategy. This observational, retrospective cohort study included 184 consecutive patients from a single center treated with 23 mm Sapien 3 Ultra (Edwards Lifesciences) BEV implantation in the aortic position and enrolled between January 2020 and April 2023. Patients treated with routine postdilation (RP, n = 73) were compared to patients treated according to local standard practice (SP, n = 111). Primary endpoint was 30-day mean TG. Secondary endpoints were incidence of 30-day prosthesis-patient mismatch (PPM), technical success and device success. Thirty-day mean TG was lower in RP versus SP (12.3 ± 4.6 mmHg vs. 14.1 ± 5.7 mmHg, p = 0.031), and incidence of PPM was less common with RP versus SP (47.3% vs. 71.0%, p = 0.006). Technical success (98.6% vs. 99.1%, p = 0.637) and device success (93.1% vs. 90.1%, p = 0.330) did not differ between groups. Differences in 30-day mean TG were driven by patients at normal flow (12.1 ± 4.0 mmHg vs. 15.0 ± 5.5 mmHg, p = 0.014), while no differences were evident among patients at low flow (12.5 ± 5.5 mmHg vs. 11.7 ± 5.5 mmHg, p = 0.644). RP decreased height and increased width of BEV, and a linear regression established that final BEV width could predict 30-day mean TG (r = -0.6654, p < 0.0001). RP after TAVR with small BEV was associated with more favorable forward-flow hemodynamics than SP.