Falls are a great concern for poststroke patients. Various interventions have been developed over the past few decades to prevent falls. However, the effectiveness of these interventions remains to be investigated. These authors aimed to evaluate the effects of exercise interventions on the prevention of poststroke falls. CNKI, Wan Fang, VIP, SinoMed, PubMed, Embase, Cochrane Library, and CINAHL were searched for randomized controlled trials (RCTs) on the prevention of falls after stroke from inception to September 2021. The primary result was the number of falls. Two reviewers independently screened and extracted data and assessed the risk of bias for all studies. In Stata 15.1, the effects of multiple interventions were compared using Bayesian networks. A total of 15 RCTs with 8 kinds of exercise interventions were included. Balance training (BT) was the most effective way to prevent falls (odds ratio [OR] = 0.24, 95% confidence interval [CI] = 0.13-0.46, p < 0.05). Moreover, cognition and movement multitask training (CMM) (OR = 0.30, 95% CI = 0.09-0.96, p < 0.05); Multimodal Exercise (OR = 0.31, 95% CI = 0.11-0.84, p < 0.05) and Resistance Exercise (OR = 0.35, 95% CI = 0.15-0.84, p < 0.05) were ranked as second, third and fourth most effective, respectively. The effect of Walking-based Intervention was the worst (OR = 1.63, 95% CI = 0.57-4.67, p > 0.05). BT and CMM are the preferred exercise interventions for the prevention of poststroke falls. A further investigation is needed to compare the effectiveness between BT and CMM for populations at high risk of falling after stroke.