Situational theories of crime assert that the situations that people participate in contain the proximal causes of crime. Prior research has not tested situational hypotheses rigorously, either for lack of detailed situational data or for lack of analytical rigor. The present research combines detailed situational data with analytical methods that eliminate all stable between‐individual factors as potential confounds. We test seven potential situational causes: 1) presence of peers, 2) absence of adult handlers, 3) public space, 4) unstructured activities, 5) use of alcohol, 6) use of cannabis, and 7) carrying weapons. In a two‐wave panel study, a general sample of adolescents completed a space–time budget interview that recorded, hour by hour over the course of 4 complete days, the activities and whereabouts of the subjects, including any self‐reported offenses. In total, 76 individuals reported having committed 104 offenses during the 4 days covered in the space–time budget interview. Using data on the 4,949 hours that these 76 offenders spent awake during these 4 days, within‐individual, fixed‐effects multivariate logit analyses were used to establish situational causes of offending. The findings demonstrate that offending is strongly and positively related to all hypothesized situational causes except using cannabis and carrying weapons.