AIM: To investigate the influence of postoperative intraocular lens (IOL) positions on the accuracy of cataract surgery and examine the predictive factors of postoperative biometry prediction errors using the Barrett Universal II (BUII) IOL formula for calculation. METHODS: The prospective study included patients who had undergone cataract surgery performed by a single surgeon from June 2020 to April 2022. The collected data included the best-corrected visual acuity (BCVA), corneal curvature, preoperative and postoperative central anterior chamber depths (ACD), axial length (AXL), IOL power, and refractive error. BUII formula was used to calculate the IOL power. The mean absolute error (MAE) was calculated, and all the participants were divided into two groups accordingly. Independent t-tests were applied to compare the variables between groups. Logistic regression analysis was used to analyze the influence of age, AXL, corneal curvature, and preoperative and postoperative ACD on MAE. RESULTS: A total of 261 patients were enrolled. The 243 (93.1%) and 18 (6.9%) had postoperative MAE<1 and >1 D, respectively. The number of females was higher in patients with MAE>1 D (χ2 = 3.833, P=0.039). The postoperative BCVA (logMAR) of patients with MAE >1 D was significantly worse (t=-2.448; P=0.025). After adjusting for gender in the logistic model, the risk of postoperative refractive errors was higher in patients with a shallow postoperative anterior chamber [odds ratio=0.346; 95% confidence interval (CI): 0.164, 0.730, P=0.005]. CONCLUSION: Risk factors for biometry prediction error after cataract surgery include the patient’s sex and postoperative ACD. Patients with a shallow postoperative anterior chamber are prone to have refractive errors.