Exercise addiction is characterized by excessive exercise behavior with potential negative consequences. The symptoms consist of salience, mood modification, tolerance, withdrawal symptoms, conflicts, and relapse. The purpose of this study was to test the psychometric properties of the exercise addiction inventory (EAI) and to estimate the prevalence of exercise addiction in a Danish normal weight sport population. A sample of 780 habitual fitness and football exercisers were contacted and 590 completed the EAI and an in-house questionnaire containing questions about variables related to exercise addiction: (a) exercise frequency; (b) continuance despite injuries; and (c) personal perception of addiction. The results demonstrated an overall prevalence of exercise addiction of 5.8%. There was no significant difference between fitness and football prevalences. The internal reliability of EAI was acceptable with a Cronbach's α of 0.66. The criterion validity was tested toward the three variables related to exercise addiction. The dependent group had significantly higher scores on the three variables than the non-dependent group. Exercise addiction seems to exist in both fitness and football. The EAI is a useful screening tool and might be applicable in future screening and prevention of exercise addiction. However, further investigation about the population is needed to understand the phenomenon and to identify the risk group.