Objectives Military personnel are required to train and operate in challenging multi-stressor environments, which can affect hormonal levels, and subsequently compromise performance and recovery. The aims of this project were to 1) assess the impact of an eight-day military training exercise on salivary cortisol and testosterone, 2) track the recovery of these hormones during a period of reduced training. Methods This was a prospective study whereby 30 soldiers (n = 27 men, n = 3 women) undergoing the Australian Army combat engineer ‘Initial Employment Training’ course were recruited and tracked over a 16-day study period which included an eight-day military training exercise. Non-stimulated saliva samples were collected at waking, 30 min post waking, and bedtime on days 1, 5, 9, 13, 15; measures of subjective load were collected on the same days. Sleep was measured continuously via actigraphy, across four sequential study periods; 1) baseline (PRE: days 1–4), 2) field training with total sleep deprivation (EX-FIELD: days 5–8), 3) training at simulated base camp with sleep restriction (EX-BASE: days 9–12), and 4) a three-day recovery period (REC: days 13–15). Results Morning cortisol concentrations were lower following EX-FIELD (p<0.05) compared to the end of REC. Training in the field diminished testosterone concentrations (p<0.05), but levels recovered within four days. Bedtime testosterone/cortisol ratios decreased following EX-FIELD and did not return to pre-training levels. Conclusions The sensitivity of testosterone levels and the testosterone/cortisol ratio to the period of field training suggests they may be useful indicators of a soldier’s state of physiological strain, or capacity, however inter-individual differences in response to a multi-stressor environment need to be considered.