Fifty‐four moderately trained men were evaluated following resistance training to determine the contribution of structural dimensions to maximal bench press lifting capacity. Muscle circumferences were taken around the flexed upper arm and the relaxed chest. Skeletal dimensions included the length of the upper and lower arm, the drop distance of the bar during the bench press lift, and the width between the hands on the lifting bar. Skinfolds were used to estimate body composition. Multiple regression analysis selected upper arm muscle plus bone cross‐sectional area, age, upper arm length, and lean body mass as the best predictors of bench press performance (r= 0.79, standard error of estimate = 13.8 kg). Prediction of Schwartz rating bench press (r = 0.59) and bench press/ kg (r= 0.56) from anthropometric dimensions were less successful but used similar variables. Body mass, either in the form of total body mass or lean body mass, appeared to be the major factor influencing bench press performance. In addition, individuals with shorter, more muscular arms at a given body mass performed better in the bench press lift.