Abstract Background Along with increasing research on acupuncture for chronic pain, the validity of sham acupuncture (SA) has also been argued. Methods Nine databases were searched for randomized controlled trials (RCTs) from the inception dates of the databases to July 5, 2022. With Markov Chain Monte Carlo methods, a Bayesian multiple-treatment network meta-analysis (NMA) with random-effects model was conducted. Results A total of 62 RCTs with 6,806 patients and four kinds of treatments (real acupuncture RA, non-acupuncture NA, penetrative SA PSA, and non-penetrative SA NPSA) were included. The results indicated that both NPSA and PSA were not superior to NA in improving chronic pain (NPSA: mean difference MD= –4.77, 95% confidence interval CI –11.09 to 1.52; PSA: MD= –4.96, 95% CI –10.38 to 0.48). After NPSA and PSA were combined into the SA group, the weak trend of pain relief from SA was still not statistically significant (MD= –4.91, 95% CI –9.93 to 0.05). NPSA and PSA had similar effects (MD= 0.18, 95% CI –5.45 to 5.81). RA was significantly associated with pain relief, compared with NPSA and PSA (NPSA: MD= –12.03, 95% CI –16.62 to –7.41; PSA: MD= –11.85, 95% CI –15.48 to –8.23). The results were generally consistent regardless of pain phenotype, frequency, duration, acupuncture methods, analgesic intake, or detection bias. Conclusion These results suggested that acupuncture was significantly associated with reduced chronic pain. The two kinds of placebo acupuncture, NPSA and PSA, have similar effects. Both NPSA and PSA, with a weak but not significant effect, are appropriate to be inert placebo controls in RCTs for chronic pain.