Correct prediction where a person predicted not to be violent does not act violently.
The psychological bias of believing two unrelated events are correlated, such as an intuitive feeling and actual dangerousness.
Correct prediction where a person predicted to be violent engages in violence.
The statistical difficulty in predicting rare events, inevitably leading to a high number of false positive errors.
Incorrect prediction where a person predicted not to be violent acts violently.
Incorrect prediction where a person predicted to be violent is not.