Members of the legal system (e.g., experts, jurors, investigators) are often skeptical of the information provided by intoxicated witnesses given the negative stigma surrounding alcohol and memory. However, studies examining the relationship between alcohol and witness memory often find that alcohol has no effect on peoples’ recall or their ability to identify a previously seen face. While insightful, the validity of these findings has been questioned given the low-moderate blood alcohol concentration (BAC) levels tested in these predominantly laboratory-based studies, which may not be high enough to consistently expose the cognitive deficits alcohol is expected to create. The present study examined how elevated BAC levels affect witnesses’ recall. In addition, it examined how identification format (i.e., showup versus lineup) impacts witnesses’ identification decisions at elevated BAC levels. Bar patrons (N = 132) were asked to participate in a study examining the effects of alcohol on cognitive and motor functioning. Consenting patrons’ BAC levels were recorded and they were given instructions for their first motor task. Midway through this task a confederate intruder entered the room and caused a disturbance. Participants were then asked to recall the intrusion via a mock interview and identify the intruder from a lineup or showup in which she was pictured (target-present) or was not pictured (target-absent). This procedure yielded participants with BAC levels as high as .24%. Linear regressions showed that elevated BAC levels reduced both the quantity and quality of information provided by participants. Logistic regressions showed that alcohol had no effect on identification decisions, regardless of identification format or target presence. These data highlight the importance of testing witnesses’ memory across a broad BAC spectrum and suggest that the legal system may benefit from expert information on alcohol’s lack of effect on memory for faces, despite what jurors may believe.