Hacker News Reader

Comment by averageweather | Hacker News Reader

I understand this statistic and get why it's often referenced, but how about this question instead? "Per accident, which has higher likelihood of death?"

I feel that's what worries people. Ya, per mile traveled its less likely, but per accident I imagine is a much different story.

View on HN

"Probability of death conditional on an accident" sounds like a statistic designed to fool people who really just care about the probability of death, preying on a cognitive error.

But that statistic is pointless because you're so much more unlikely to get into an accident. It's irrelevant that you're more likely to die in a plane crash than a car crash. By that logic, you should never drive more than 5 miles an hour, because with each mile an hour you increase your speed, you're more likely to be in a fatal accident.

That makes it worse actually - not only are you more likely to die in a car per mile, but you are many more times likely than THAT higher number to get injured.

It may be what worries people, but that's only because people are irrational. That number is utterly meaningless. If you use that number to decide safety, then stubbing your toe before you go BASE jumping makes it safer!

you you're saying that people prefer to be maimed more and so would choose driving over flying because there's less chance of them being maimed when they fly?

honestly that makes no sense

But that's an inherently wrong way of looking at it. If you're traveling somewhere in a plane vs on the road, you're way more likely to die. Even if you factor own your own human error (but not others').

Reply on news.ycombinator.com