Objectives To compare the predictive ability of six anthropometric indices for identification of metabolic syndrome (MetS) and to determine their optimal cut-off points among Chinese adults. Methods A total of 59,029 participants were enrolled. Body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHtR), a body shape index (ABSI), body roundness index (BRI), and conicity index (CI) were measured. Receiver-operating characteristic curves analyses were performed to determine the discriminatory power of these indices for the identification of cardiometabolic risks and diagnosis of MetS. The differences in the area under the curve (AUC) values among the indices were evaluated. The Youden index was used to determine the optimal cut-off points. Results WHtR and BRI exhibited the highest AUC values for identifying MetS and most cardiometabolic risk factors in both sexes, whereas ABSI showed the lowest AUC value. The general optimal cut-off points in women were 23.03 kg/m2 for BMI, 77.25 cm for WC, 0.490 for WHtR, and 3.179 for BRI; those in men were 24.64 kg/m2 for BMI, 87.25 cm for WC, 0.510 for WHtR, and 3.547 for BRI. The AUC values and cut-off points of the indices were also analyzed in each age and BMI category. Conclusions In Chinese adults, WHtR and BRI showed a superior predictive power for MetS in both sexes, which can be used as simple and effective screening tools for cardiometabolic risks and MetS in clinical practice.