In response to the asymmetric advantage that attackers enjoy over defenders in cyber systems, the cyber community has generated a steady stream of cybersecurity-related frameworks, methodologies, analytics, and "best practices" lists. However, these artifacts almost never undergo rigorous validation of their efficacy but instead tend to be accepted on faith, to, we suggest, our collective detriment based on evidence of continued attacker success. But what would rigorous validation look like, and can we afford it? This paper describes the design and estimates the cost of a controlled experiment whose goal is to determine the effectiveness of an exemplar systems security analytic. Given the significant footprint that humans play in cyber systems (e.g., their design, use, attack, and defense), any such experiment must necessarily take into account and control for variable human behavior. Thus, the paper reinforces the argument that cybersecurity can be understood as a hybrid discipline with strong technical and human dimensions.