Well, .00000003% or so of the US population gets eaten by sharks every year, so there must be some circumstances under which it becomes a significant risk. The

Well, .00000003% or so of the US population gets eaten by sharks every year, so there must be some circumstances under which it becomes a significant risk. The question is whether those circumstances are identifiable enough to be worth the cognitive load of avoiding: If 90% of shark attacks happened to people swimming at night in Florida, and only .00000006% of the US population actually swims at night in Florida, it might be an activity worth avoiding.