The present study attempted to find alternative methods to measure training stress when cortisol blood analysis is not available. The relationships during exercise between cortisol and six non-invasive physiological measures (heart rate, lactate, lactate:RPE ratios, oxygen uptake, oxygen pulse, and ratings of perceived exertion) were assessed. Subjects (n = 18) participated in four experimental sessions (40%, 60%, and 80% VO2max, and control). Blood samples were taken pre- and post-exercise along with non-invasive measures being recorded throughout the exercise session. All correlations between cortisol and the non-invasive measures were significant. However, the small variance accounted for between each non-invasive measure and cortisol lead to low predictability. Individually, non-invasive measures are not viable indicators of training stress. However, exploratory step-wise multiple regression analysis revealed that when combined, blood lactate and lactate:RPE ratio are highly predictive of cortisol, suggesting that in combination they may be a plausible alternative to assess training stress.