BACKGROUND : Overuse accounts for 82% of injuries in military personnel, and these occur predominantly in the spine and lower limbs. While non-linear analyses have shown changes in overall stability of the movement during load carriage, individual joint contributions have not been studied. The concept of entropy compensation between task, organism and environmental constraints is studied at a joint level. OBJECTIVE : The aim of this study was to investigate whether using different methods of loading by military personnel would have an effect on the sample entropy of the joint ranges of motion. METHODS : Eleven male reserve infantry army soldiers (age: 22 ± 2 years; height: 1.80 ± 0.06 m; mass: 89.3 ± 14.4 kg) walked an outdoor, 800 m course under 5 load conditions: unloaded, 15 kg backpack, 25 kg backpack, 15 kg webbing and backpack and 25 kg webbing and backpack. Kinematic data was recorded at 240 Hz using the Xsens motion capture system. The ranges of motion (ROM) of the spine, hips and knee were calculated for each gait cycle. Mean ROM, coefficient of variation (CV) of the ROM and the sample entropy of the ROM were compared between conditions. RESULTS : Spine side flexion ROM decreased significantly from the control condition in all loaded conditions, while sample entropy of the spine side flexion ROM increased in some conditions with no significant change in CV. Conversely, the hip flexion ROM increased significantly from the control, while sample entropy of the hip flexion ROM decreased. CONCLUSIONS : These results suggest that entropy compensation may propagate at a joint level. Understanding that a decrease in certainty with which a joint angle is selected, may be accompanied by an increase at a neighbouring joint. This could be significant in monitoring injuries as a result of environmental or task constraints.