Despite high rates of comorbid hazardous alcohol use and posttraumatic stress disorder (PTSD), the nature of the functional relationship between these problems is not fully understood. Insufficient evidence exists to fully support models commonly used to explain the relationship between hazardous alcohol use and PTSD including the self-medication hypothesis and the mutual maintenance model. Ecological momentary assessment (EMA) can monitor within-day fluctuations of symptoms and drinking to provide novel information regarding potential functional relationships and symptom interactions. This study aimed to model the daily course of alcohol use and PTSD symptoms and to test theory-based moderators, including avoidance coping and self-efficacy to resist drinking. A total of 143 recent combat veterans with PTSD symptoms and hazardous drinking completed brief assessments of alcohol use, PTSD symptoms, mood, coping, and self-efficacy 4 times daily for 28 days. Our results support the finding that increases in PTSD are associated with more drinking within the same 3-hr time block, but not more drinking within the following time block. Support for moderators was found: Avoidance coping strengthened the relationship between PTSD and later drinking, while self-efficacy to resist drinking weakened the relationship between PTSD and later drinking. An exploratory analysis revealed support for self-medication occurring in certain times of the day: Increased PTSD severity in the evening predicted more drinking overnight. Overall, our results provide mixed support for the self-medication hypothesis. Also, interventions that seek to reduce avoidance coping and increase patient self-efficacy may help veterans with PTSD decrease drinking.