Police use of excessive force is under constant scrutiny. Recent high-profile killings of unarmed Black individuals have bolstered calls for reform regarding police use of force. De-escalation has been recommended as a solution to help police reduce their use of force and help repair community relations. However, there is much that we still do not know about de-escalation: What is it exactly? What are de-escalation tactics? Which are considered the most effective? By examining a group of peer-nominated top de-escalators in the Tempe (AZ) Police Department, this dissertation serves to uncover the black box of de-escalation. This in-depth case study of top de-escalators is presented in three parts. Part 1 utilized data from an officer perception survey (N=101). The perceptions of the officers were categorized into three groups: 1) the importance of de-escalation tactics, 2) the frequency of these de-escalation tactics, and 3) the perceptions of de-escalation training. The results in Part 1 highlight which tactics top de-escalators view as most important and which tactics they report using more frequently. Additionally, differences between the top de-escalators and their peers were examined. In Part 2, the author utilized data via systematic social observation of body-worn camera footage (N=228 coded interactions). The behaviors of the officers were categorized into three groups: 1) objectively observable de-escalation tactics, 2) “to-do’s” tactics, and 3) “do not’s” tactics. This phase of the study examined what de-escalation tactics top de-escalators were observed using most frequently (or least) and how they differed from their peers. In Part 3, the author utilized data from a social network question that was administered alongside the officer perception survey (N=64). This phase of the research examined if top de-escalators were perceived as more influential than their peers regarding de-escalation. The empirical, methodological, theoretical, and practical implications of these findings are also discussed.