This paper describes a detailed empirical comparison of a 3-D Static Strength Prediction Model with a set of strength performance data. The data were acquired from 29 male volunteers who performed four different static exertions (e.g., lift, push-down, pull-in, and push-out), with one hand and both hands using a force cell located at six different positions around the body. It is shown that the biomechanical static strength prediction logic is extremely sensitive to errors in postures and somewhat sensitive to errors in anthropometry used as inputs to the model. It does appear that with good postural data the model is capable of predicting a population's mean static strengths reasonably well.