The cortisol awakening response (CAR) is increasingly recognized as a potential biological marker of psychological and physical health status. Yet, the CAR literature is replete with contradictory results: both supposedly protective and vulnerability psychosocial factors have been associated with both increased and decreased CAR. In this study, we tested the hypothesis that the CAR flexibility would be a better indicator of psychological status than CAR magnitude. Forty-two men measures of happiness, perceived stress and neuroticism, and took saliva samples immediately on awakening, then at 15, 30, 45 and 60. min post-awakening on three study days (i.e., Sunday, Monday and Tuesday). When considering the CAR magnitude, our effects perfectly reflect the inconsistencies previously observed in the literature (i.e., the main effects of the psychological predictors are not consistent with each other, and the effect of one predictor on a given day contradicts the effect of the same predictor on another day). However, considering the CAR flexibility leads to a fully consistent pattern: protective factors (i.e., high happiness, low stress, low neurotiscim) are associated with a flexible CAR (i.e., lower CAR during weekends compared to workdays) whereas vulnerability factors (i.e., low happiness, high stress, high neurotiscim) are associated with a stiff CAR (i.e., same magnitude during weekends and workdays). We conclude that considering the CAR flexibility (e.g., between weekends and workdays) rather than the traditional CAR magnitude might be a way to understand the apparent conflicts in the CAR literature.