I'm not quite implying that events happen with dice. More that if we knew the true frequency, we would gauge there to be far less risk (because there is actuall

I'm not quite implying that events happen with dice. More that if we knew the true frequency, we would gauge there to be far less risk (because there is actually far less risk). Meanwhile, you don't fully control if you slip in a bathtub, a person ascribes much less risk to the bathtub than they should (due to lack of confirmation bias and lack of knowledge of the risk stats, bathtubs are surprisingly riskier than you think; but if you put it in the context of what is the chance of a fall, you would start to think the odds are not that bad - and any serious fall with that many hard surfaces around can easily have a very bad outcome. So it's more consider the risk of any fall, now compare that to the statistics of 'stranger danger'. The point is that everyday common events, your car, your bathtub, your daily commute, can pose magnitudes more risk than even the shadiest of homeless person).