Hacker News Reader

Comment by jjeaff | Hacker News Reader

My guess is that self driving cars are already far, far better than human drivers at not hitting pedestrians.

My guess is that it depends a lot on the driver. There are a lot of really reckless drivers, or really bad drivers out there on the roads right now. I know people who've been in several (relatively minor) car crashes, driving the same-ish sort of route to/from college and work that I was during the same period. I think they're just a bad driver. I feel like they're driving recklessly when I'm a passenger in a car they're driving. So a self-driving car might be a better driver than them, while also being worse than the median driver, and much much worse than like the upper portion of the human driver spectrum, like the 1st quintile of the 20% of human drivers who are pretty good.

Given how reckless some drivers are, I'd imagine the distribution of crashes among drivers is very skewed. I don't know numbers, but I'd expect to see something like 80% of car crashes involving (or objectively caused by) 20% of the drivers, with most car crashes involving people who have been involved in numerous crashes, and a large minority of drivers having been involved in zero car crashes.

> My guess is that it depends a lot on the driver.

There is a lot of variation in drivers. Young ones with good reflexes and senses but bad on experience, old ones with worse reflexes and senses but better on experience. Ones that stop at stop signs, ones that blow through them if they think they can get away with it. Cities like New Orleans where drivers are much more aggressive because the cops don't care, etc...

I think the important point is that self driving cars are likely better than the averaged real world population of existing human drivers at avoiding pedestrians.

The average is not the only stat though, and doesn't tell everything. Even if it were true that FSD was safer than the average of human drivers, replacing some manually-steered cars with FSD cars could result in more pedestrians getting hit, depending on the breakdown within the two groups. The overall average isn't enough information and obscures the internal distribution of skill/safety within the overall population of drivers.

It matters who is driving what cars. If you're taking safe drivers and having them drive FSD cars, and comparing the average of that against the average of all non-FSD driven cars (which includes a lot of really shitty drivers in the average), then the net safety impact of this could be nothing. Introducing FSD cars would have no impact to overall car safety, and might even have a negative impact, if the FSD software is worse than the good drivers who are now not driving manually-steered cars but are driving FSD cars instead.

You might think that the average is not too terrible, but wow, there are some people at that one pointy end of the bell curve that make you shake your head.

I present, "Canada's Worst Driver," a series that I watched quite a bit when I had a cable subscription. Besides being hilarious, it should make you think a time or three before crossing the street. https://www.youtube.com/playlist?list=PLHB4WuM7Y_eP-PFv5F6To...

The issue i speculate is the tradeoffs are more imposing than with a human. Sure i accept that the car can drive more safely than a human (all it has to do is never exceed 1mph) but as safety increases i speculate it involves accepting tradeoffs like cars getting stuck, driving slowly, blocking traffic, hard braking when not necessary, or even restricting the operating domain.

Reply on news.ycombinator.com