BARD1 has been shown to play tumor suppressive roles in human cancer. We performed this meta-analysis and firstly evaluated the association between three common BARD1 polymorphisms (Arg378Ser, Val507Met and Pro24Ser) and cancer susceptibility. We performed this meta-analysis following PRISMA guidelines. A comprehensive search of PubMed, EMBASE, Cochrane Library, OVID and Web of Science databases was done from database inception to August 2014. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were combined to measure the association between BARD1 polymorphisms and cancer risk. On the basis of 10 studies about BARD1 polymorphisms and cancer, we found that BARD1 Val507Met (G/A) polymorphism was associated with decreased cancer susceptibility (allelic model: OR = 0.76, 95% CI: 0.66-0.87, P < 0.00001; dominant model: OR = 0.77, 95% CI: 0.65-0.91, P < 0.00001; recessive model: OR = 0.64, 95% CI: 0.55-0.74, P < 0.00001; homozygote comparison: OR = 0.58, 95% CI: 0.49-0.70, P < 0.00001; heterozygote comparison: OR = 0.85, 95% CI: 0.74-0.99, P = 0.0008). BARD1 Pro24Ser (C/T) polymorphism was also associated decreased cancer risk in allelic model (OR = 0.72, 95% CI: 0.60-0.88, P = 0.0009), dominant model (OR = 0.70, 95% CI: 0.56-0.87, P = 0.004), recessive model (OR = 0.70, 95% CI: 0.56-0.87, P = 0.004), homozygote comparison (OR = 0.55, 95% CI: 0.39-0.78, P = 0.0007) and heterozygote comparison (OR = 0.75, 95% CI: 0.62-0.91, P = 0.004). And in our sensitivity analysis, when deleting the study performed by Capasso in 2009, we found that BARD1 Arg378Ser polymorphism was associated with decreased cancer risk in allelic model (OR = 0.81, 95% CI: 0.67-0.97, P = 0.02), dominant model (OR = 0.72, 95% CI: 0.56-0.91, P = 0.007) and heterozygote comparison (OR = 0.72, 95% CI: 0.57-0.91, 0 = 0.006). In conclusion, BARD1 Arg378Ser, Val507Met and Pro24Ser may be associated with decreased cancer risk. More studies with larger samples and gene-environment interactions are needed to confirm our findings.