Controversy remains as to which obesity measures better predict type 2 diabetes (T2D) risk in overweight or obese individuals. The objective of this study is to determine which commonly used obesity measures better reflect beta cell function and predict T2D risk in participants with overweight or obesity and to validate the findings using prospective cohort data. Cross-sectional data from the Obesity Clinic of the Xiangya Hospital of the Central South University and prospective cohort from UK Biobank. BMI, waist circumference (WC), waist-to-hip ratio (WHpR), and waist-to-height ratio (WHtR) were measured. Primary outcomes included beta cell function indices in the cross-sectional study and the occurrence of diabetes obtained from UK Biobank data. One thousand four hundred and ninety-seven participants with overweight or obesity (median age 29 years, 41% males) and 322,023 UK Biobank participants without diabetes at baseline (mean age 56.83 years, 50.4% males) were studied. WHpR had a stronger association with beta cell function and central body fat distribution than the other three obesity measures irrespective of glucometabolic states. WHpR associated positively with diabetes risk in participants using the hazard ratio scale (HR per SD increase of WHpR, 2.311, 95% CI 2.250-2.374). WHpR is a superior index in reflecting central body obesity, estimating beta cell function, and predicting T2D risk in people with overweight or obesity compared to BMI, WC, and WHtR.