Exercise intervention (EI) is a promising and economical way for elderly patients with hip fracture, but it is unclear which type of exercise is optimal. To assess the effectiveness of different EIs and identify the optimal EI of improving the prognosis for elderly patients with hip fracture. We searched Medline (via PubMed), Web of Science, Embase, Cochrane Central Register of Controlled Trials, CINAHL, CNKI, Wan Fang, VIP and CBM from their earliest records to June 2022. The included randomized controlled trials (RCT) included at least one type of exercise for patients with hip fracture. The methodological quality of these trials was assessed using the Cochrane Collaboration Risk of Bias Tool, Stata 14.0 software was used for NMA. The primary outcome was hip function, and the secondary outcomes were activity of daily living (ADL), walking capacity and balance ability of the patients. This study included 46 RCTs involving 3,286 patients. Based on the recovery of patients' hip function, resistance exercise (RE) ([MD]: -11.07, [Crl]: -15.07 to -7.08) showed significant benefits over the control group (CG). RE ([MD]: -8.86, [Crl]: -15.01 to -2.71) ([MD]: -9.09, [Crl]: -14.20 to -3.98) showed significant benefits over the weight-bearing exercise and aerobic exercise. In addition, RE (surface under cumulative ranking curve values [SUCRA]: 94.8%) and balance exercise (BE) ([SUCRA]: 81.1%) were most likely to be the optimal exercise methods for hip function recovery. BE may be the most effective exercise in improving ADL([SUCRA]: 98.4%) and balance ability([SUCRA]: 88.4%). These results suggest that RE and BE might be the best approach to improve prognosis for elderly patients with hip fracture. More well-designed and high-quality RCTs are needed to verify the conclusions drawn from this study.