2 ms·

Statistics often miss a huge amount of threatening behaviour, and in many countries you just outright cannot trust the stats at all. Plus, people will take more

Statistics often miss a huge amount of threatening behaviour, and in many countries you just outright cannot trust the stats at all. Plus, people will take more precautions in areas that are genuinely more dangerous - if they didn't, the statistics would look much worse. In general I think people are actually quite good at perceiving the risk present to them. That doesn't mean 'always correct' but it does mean 'has a good idea of which places are at much greater risk of turning dangerous at 2am'. Finally, don't forget survivor bias: people who were killed or whatever in a random dangerous area simply cannot go on to report how unsafe an area or activity is. Overall your chances of a bad outcome might still be low in terms of absolute risk but being in the wrong area could easily be hundreds of times more dangerous than anywhere else you've been.

clairity 4y ago
no, people are mostly terrible at perceiving actual risk vs. the imagined, because evolutionarily, it was much less harmful to get it wrong one way than the other. we humans no longer live in a world of such unpredictable mortality. and even if imprecise, statistics provide relevant information, especially on relative and general magnitude. moreover, serious crimes are the ones most likely to be reported, and therefore having more accurate stats, so survivorship bias isn't a significant factor. humans consistently overestimate the dangers of the unknown and consistently underestimate the dangers of the known, often by orders of magnitude. we're also wired to be highly overconfident in our estimations of just about anything. your conjectures include prime examples of such mis-estimation, e.g., "much greater risk" and "hundreds of times more dangerous", implying all that avoidance behavior is justified. but death by firearm, for instance, is caused 90+% of the time by the self or familiars, yet we spend all of our time and resources worrying about strangers and extremely rare scenarios (armed robbery, home invasion, mass shootings, terrorism, etc.) rather than the common (suicide, accidental shooting, crimes of passion, etc.). mostly, we worry about the wrong things most of the time.

clairity 4y ago

no, people are mostly terrible at perceiving actual risk vs. the imagined, because evolutionarily, it was much less harmful to get it wrong one way than the other. we humans no longer live in a world of such unpredictable mortality. and even if imprecise, statistics provide relevant information, especially on relative and general magnitude. moreover, serious crimes are the ones most likely to be reported, and therefore having more accurate stats, so survivorship bias isn't a significant factor. humans consistently overestimate the dangers of the unknown and consistently underestimate the dangers of the known, often by orders of magnitude. we're also wired to be highly overconfident in our estimations of just about anything. your conjectures include prime examples of such mis-estimation, e.g., "much greater risk" and "hundreds of times more dangerous", implying all that avoidance behavior is justified. but death by firearm, for instance, is caused 90+% of the time by the self or familiars, yet we spend all of our time and resources worrying about strangers and extremely rare scenarios (armed robbery, home invasion, mass shootings, terrorism, etc.) rather than the common (suicide, accidental shooting, crimes of passion, etc.). mostly, we worry about the wrong things most of the time.