Video-recorded eyewitness identification procedures have heretofore unexamined potential for assisting evaluators in assessing eyewitness accuracy. In Study 1 (N = 240), evaluators, on average, successfully differentiated accurate from inaccurate witnesses based on videos of identification procedures alone, but not when extraneous incriminating evidence was also provided. Study 2 (N = 433) replicated this effect using different stimulus videos and operationalizations of evidence. Study 3 (N = 957) revealed that instructions highlighting the limitations of forensic evidence did not preserve evaluators’ ability to discern accuracy when extraneous incriminating case evidence was provided. Moreover, case information affected other judgments (e.g., perceptions of the witness’s view). Overall, results indicate that evaluators can differentiate between accurate and inaccurate witnesses, simply based on a video record of the identification procedure, bolstering recommendations to record identification procedures. However, these evaluations lose their utility when evaluators know about other case evidence.