Studies of stress and coping effects on college performance have yielded inconsistent results, which might be clarified by stricter adherence to stress theories. Here, university students (n = 1212) were surveyed during their first year, said to be the most stressful. They completed measures of stress overload (the pathogenic state identified by theory) and coping preferences (the strategies most reported in the literature), and granted access to their official grade and enrollment records. Regression analyses showed stress overload to relate to poorer performance, and avoidance coping to relate to greater stress overload. SEM models indicated a theory-consistent causal sequence fit the data better than the reverse. However, after model modifications, only avoidance showed an effect on performance mediated by stress overload, as predicted, while other strategies showed direct effects. Study limitations in regard to causal inferences and generalizability, and larger problems in applying medical stress theories to academia, are discussed.