3 ms·

> A car accident can kill you at 35 mph but they usually didn't (and usually don't). Cars are safer now and I appreciate that, but they weren't _that_ unsafe: p

> A car accident can kill you at 35 mph but they usually didn't (and usually don't). Cars are safer now and I appreciate that, but they weren't _that_ unsafe: people would not have driven them otherwise. Cars were a lot less safe in 1971, that is a fact. No air bags, no anti-lock brakes, no traction control. Older cars didn't have seat belts, shoulder belts were rare, cars weren't built to crumple, steering columns would impale you, etc. Fact, not anecdote: In 1971 there were 4.46 fatalities per million vehicle miles traveled (MVMT). In 2015 it was 1.18 fatalities per MVMT. So yes, in fact cars were three times as dangerous as today. https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in_U.S._by_year https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in...

giardini 8y ago
1) What I stated was not anecdote: a car accident can kill you at 35 mph but they usually didn't (and don't). 2) I simply don't agree with your use of relative statistics, which reeks of alarmism. My doctors use such arguments in attempts to convince me to take statin drugs, which would at best extend my life for only a few days if I took them for 30 years. He always tells me that I'm "X times as likely to die from a heart attack" unless I take statins. I always remind him of the nature of his statistics, the side-effects of statins, and the non-present longevity payoff of taking them. Once the number of fatalities per MVMT is below a certain value it ceases to be likely in the overall scheme of things and becomes unimportant. Relative statistics ignore the prior probabilities - the fact that I'm unlikely to die driving in the first place. Yes, 4.46 deaths per MVMT is about 3 X 1.18, but both are low deaths per MVMT. I may die from falling off a ladder or get eaten by a shark but neither is likely, nor am I now nor was I likely in the 70s to die in a car accident. Some other damn thing will most likely get me (that isn't anecdotal either).

giardini 8y ago

1) What I stated was not anecdote: a car accident can kill you at 35 mph but they usually didn't (and don't). 2) I simply don't agree with your use of relative statistics, which reeks of alarmism. My doctors use such arguments in attempts to convince me to take statin drugs, which would at best extend my life for only a few days if I took them for 30 years. He always tells me that I'm "X times as likely to die from a heart attack" unless I take statins. I always remind him of the nature of his statistics, the side-effects of statins, and the non-present longevity payoff of taking them. Once the number of fatalities per MVMT is below a certain value it ceases to be likely in the overall scheme of things and becomes unimportant. Relative statistics ignore the prior probabilities - the fact that I'm unlikely to die driving in the first place. Yes, 4.46 deaths per MVMT is about 3 X 1.18, but both are low deaths per MVMT. I may die from falling off a ladder or get eaten by a shark but neither is likely, nor am I now nor was I likely in the 70s to die in a car accident. Some other damn thing will most likely get me (that isn't anecdotal either).