A machine belt squat is a piece of equipment designed to allow the performance of squats while loading weight on the lifter’s hips using a belt. The purpose of this investigation was to determine if belt squats differ from back squats in activation of the primary movers, and to determine the predictive capabilities of back squat load, training status, and anthropometric data on belt squat load. Thirty-one participants (16 males and 15 females) completed anthropometric measurements, a demographic questionnaire, a familiarization visit, and two testing visits, completing a 5 repetition maximum test for back squat and belt squat. Surface electromyography was used to measure muscle activation for the left and right vastus medialis (VMO), vastus lateralis (VLO), rectus femoris (RF), and gluteus maximus (GM). Comparison of muscle activation between the two exercises showed significant differences in the left GM (back squat: 0.84 ± 0.45, belt squat: 0.69 ± 0.22, p=0.015) and right GM (back squat: 0.86 ± 0.45, belt squat: 0.71 ± 0.29, p=0.004). Regression analysis computed significant prediction equations for belt squat load for general population, males, females, and advanced lifters. Overall, results indicate that belt squats may significantly differ in GM activation from back squats. Back squat load, as well as other variables, may be effective in accurately estimating appropriate belt squat load. These findings may help to more appropriately program for training with machine belt squats as a back squat alternative.