The anomaly score is intended to help rank review priority. It answers a workflow question, not a certainty question: which symbols or portfolios look unusual enough to inspect now?
Research workflows break down when every chart feels equally important. An anomaly framework helps narrow attention by highlighting conditions that stand out from the platform's current thresholds and checks. That makes the score useful for triage, even though it is not a direct measure of opportunity or quality.
A high score usually means the current situation is unusual relative to the configured logic. That can happen during upside expansions, downside breakdowns, sharp reversals, or portfolio concentration problems. The score should lead to inspection of the driver details, not a default action.
A single score is easy to misuse. Driver breakdowns provide the "why now" layer behind the number. If one symbol is being flagged because of momentum divergence while another is being flagged because of concentration or event sensitivity, those are not equivalent situations and should not be treated the same way.
The strongest use case is comparison. Users can line up several symbols, inspect the score, then compare the chart, news, and visible drivers. That is a better workflow than using the score in isolation. Scores are heuristic outputs and may change when thresholds, data inputs, or market conditions change.
The most common mistake is to confuse "unusual" with "actionable." Many unusual conditions are worth noting but not worth trading. The score is useful when it starts a better review process, not when it ends one too early.
Related: Methodology | How Bias Read Works | FAQ