Likelihood and Severity: How Behavioral Risk Levels Actually Get Determined

When people picture a threat assessment, they often picture a single verdict: dangerous, or not dangerous. The research behind modern behavioral threat assessment points somewhere else. The U.S. Secret Service's National Threat Assessment Center has studied years of averted and completed attacks and found no single trait, background, or warning sign that reliably marks someone as a future attacker. What shows up instead is a pattern: behavior that builds over time, shaped by a person's situation, relationships, and setting, and assessed as a whole rather than checked off a list.

That finding changes the central question. Instead of asking whether a person is dangerous, it helps to ask two narrower, more answerable questions: how likely is this situation to escalate, and how severe would the outcome be if it did. Those two questions, likelihood and severity, are a practical way to think about the level of risk a behavior carries. Homicide Zero's tools are built to be compatible with the National Threat Assessment Center model, and these two questions are a useful lens on what each tool is looking for.

Two Questions, Not One Verdict

The Homicide Threat Screener (HTS) is a rapid screener built to take 5 to 10 minutes for a non-clinician such as a teacher, manager, or HR staff member who has already noticed something concerning. It does not try to produce a final risk level. It is designed to sort cases quickly into ones that can be monitored and ones that need a closer look from someone trained to go deeper.

That closer look is the Violence Safety and Risk Assessment (VSRA), a 20 to 30 minute structured assessment for trained professionals. The VSRA looks across several domains, including historical, clinical, and contextual factors and the protective factors that offset them, and the professional brings those findings together into a single severity rating that runs from low to imminent. Likelihood and severity are a useful way to read what those domains capture. How specific a plan is, whether there is access to a target, and whether behavior has recently escalated speak to how likely a situation is to progress. Access to the means to cause harm and the scale of harm that could follow speak to how serious it would be. A situation that seems unlikely to escalate but could be very serious if it did, and one that is likely to escalate but less dangerous, both deserve attention, but not the same attention.

A risk level is not a verdict on a person. It is a measurement of how likely a situation is to escalate and how severe the outcome could be, reassessed as the facts change.

Building the Two-Question Habit

The value of separating likelihood from severity reaches past any single tool. It gives frontline staff a shared vocabulary for concern that does not require a clinical license to use. A teacher, supervisor, or security officer does not need to decide whether someone is dangerous. They need to describe what they are seeing, clearly and specifically. The HTS gives that description a structure, and the VSRA gives trained professionals the depth to turn it into a decision with a documented basis.

Organizations that build this two-question habit into policy, rather than leaving the first read of a situation to one person's instinct, get a consistent, documented process that can be revisited as new information comes in. Prevention does not start with certainty. It starts with asking the right two questions about the behavior in front of you, and treating risk as something to measure, not something to guess.