Comment by BugsJustFindMe
14 hours ago
> "The majority of drivers are good, but some drivers are really bad."
I take issue with this. The majority of drivers are "mostly fine as long as circumstances are favorable", not "good". People for the most part do not even know about the concept of defensive driving let alone practice it. Even the best people are fundamentally complacent in basically everything they do. They zone out sometimes, they get too close, they don't always look where they're going. They don't get into more collisions because the world doesn't throw more obstacles at them, not because they're good at driving.
I disagreed with you and went looking in the article for evidence (yeah bad, epistemologically). I actually found evidence for your point. The largest cause of fatal crashes is Generally distracted or "lost in thought" (daydreaming) at 64%. That happens to most people,.
How many of those 64% had their accident because of phones/drugs/recklessness and claimed to be distracted instead? With all the liability and shame involved in fatal crashes there's absolutely no way people are going to honestly self report illegal behavior.
Phones definitely fall in that category. I’m guessing they are the majority, but I don’t know how to substantiate that.
Hopefully police are figuring out that one of the parties in the accident is high while they do the interview.
The human brain was not designed for a task which requires hours of siting down doing mostly nothing while being hyperfocused the entire time ready to take over in an instant on the rare occasion something happens. While running or cycling your brain is actively engaged keeping yourself upright and avoiding obstacles which keeps you alert.
Seems at least somewhat comparable to walking through a forest full of predators, while, say, hunting/gathering.
Evidence that is is compatible with our psyche is the fact that we're at 1.19 fatalities per 100 million miles driven, and ~50% of those being booze and/or speeding.
That's true. Humans are bad at this task. On the railways they have a problem with this too, "Lookout" seems like a ludicrously easy job, but it's basically impossible because you must pay attention to an incredibly boring thing for hours or your friends might die. Getting rid of human lookouts should be a priority but it feels easy and so it's hard to motivate this.
2 replies →
> The human brain was not designed for a task which requires hours of siting down doing mostly nothing while […]
The average one-way commute time in the US is less than half an hour:
* https://www.census.gov/topics/employment/commuting/guidance/...
If you can't pay attention for less than 30 minutes should you really be operating a multi-ton death machine?
2 replies →
You can see it in what people say right after an accident, stuff like "He came out of nowhere".
As a point of evidence in favor of this viewpoint:
Multi-dozen car pileups aren't particularly rare, with the common cause for a lot of these being "one car hits an icy patch, hits another car, car behind them tries but fails to stop in time due to the ice, rinse and repeat a few dozen times". And the root cause of that is every single driver in that collision driving too fast for conditions and unable to make the requisite emergency stop in time.
Good here is in reference to the average crash statistics.
Most drivers are safer than the average, because you can’t go below 0 and some people are really bad.
I know someone who's been at fault in at least 12+ car accidents, and she keeps driving (as someone else put it you almost have to drive in the USA). There are colossally bad drivers out there and we don't yet have a viable system for keeping them off the road.
How is she able to get insurance coverage?
9 replies →
You cannot measure driver goodness (skill and vigilance) by looking at crash statistics, because crashes are definitively circumstantial. Most very clearly bad/dangerous/unskilled/unvigilant operations of automobiles do not lead to crashes.
Population level statistics let you look for statistical aberrations that you can’t detect at the individual level.
Further suppose you define “good driving” as a set of behaviors but failed to include narcolepsy or similar medical conditions in your definition, then some of your “good” drivers would represent a significant risk to the general public. Thus the actual criteria the defines good driver behavior is accidents.
It looks like the median/mode is 1-2 car accidents (50% of drivers), and the average is 3-4. 25% of drivers have no accidents (as far as insurance companies are concerned, ie, no claims or police reports).
I'm curious how this looks with various ways of normalizing for amount driven; is that no-accident crowd the quartile of drivers who don't drive very much? The biggest factor in being in the opposite group (the 10% of drivers with 50% of accidents) is "drive all the time, commercially or otherwise".
A huge percentage of the cars I see on the highway are tailgating. People are awful drivers.
The fatality rate is 1.19 per 100 million vehicle miles [1], ~83 million miles per fatality. The mean driver averages ~15,000 miles per year. So, the mean driver encounters a fatal crash after ~5,500 years of driving. If you started driving when the Great Pyramid was built, you would still have another 1,000 years before you would expect to die in a fatal crash.
The mean driver in the context of the USA driving environment is shockingly safe. Being safer than even the mean human driver in the USA is hard and takes a shocking amount of data to verify. Anybody downplaying it as "lol humans are so unsafe" to argue the problem feels easy to solve or can be solved by the end of the year with a little bit of elbow grease is utterly clueless.
[1] https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/...
https://en.wikipedia.org/wiki/List_of_countries_by_traffic-r... has the US at 7th highest out of 22 countries (for which this data is available) for the most accidents per km driven.
So it's clearly not that hard to be safer than the mean human driver in the USA. 3 of the 22 countries manage to have less than half the US rate.
Sir this is a Wendy's, I mean a discussion about autonomous vehicles. The benchmark for the autonomous vehicles in the USA is in comparison to humans in the USA. That means autonomous vehicles need to go ~5,500 person-years worth of driving between crashes to equal the mean driver. The mean US driver is a very high bar to reach for technology, that is why Waymo has just maybe barely reached that bar after nearly two decades and billions of dollars.
3 replies →
80% of the data (for fatalities per km) in that table is missing
Remember, I'm responding specifically to how good people are at driving, not how likely they are to die while driving. How likely a person is to die is mediated by all kinds of things that have nothing to do with how good they are as a driver, like airbags and seatbelts and crumple zones.
In thread context, you're conflating a bunch of things, and it screws up your position as each additional conflation gets you further and further from your headline statistic into "??? -> Profit" underpants gnomes territory.
1) Fatalities vs non-fatal injuries. Not the same thing.
2) Crashes that caused injury vs crashes that did not cause injury. Not the same thing.
3) Crashes that did not cause injury vs crashes that could not have caused injury. Not the same thing.
4) Crashes vs unsafe behaviors that could cause a crash. Not the same thing.
5) Being "safe" in the sense of "enjoying fortunate circumstance" vs being "safe" in the sense of "actively contributing to safety". Not the same thing.
Here is a thought experiment for you to consider:
If you run a red light and the cross traffic brakes and swerves around you, so no collision occurs, were you driving safely? No, right? Then why would you think that fatalities are the true measure of driving goodness?
The people OP is talking about, the 0.05% absolute worst drovers, are not merely complacent, they're doing things like getting hammered and going street racing (shamefully I can say, I was 19 once...) or speeding away from crimes etc
edit: someone else replied with data contrary to this, maybe they are right but Im leaving this here to represent the counterargument
It's a category error to say that everyone else are good just because they're not the worst you can imagine. It doesn't matter what your worst 0.05% do. It doesn't make most people actually good drivers.
It's more of a category error to imply that robotaxis are 92% safer than human-driven taxis if it turns out that the raw accident rate they're baselining against is disproportionately caused by people who not only aren't full time professional drivers but shouldn't be allowed near a car [in their current state]...
1 reply →
The other reply doesn't contradict what you're saying. Someone who was street racing is going to tell the cops / insurance company that they were lost in thought.
I mean when comparing to machines though the only thing the machines are better at is maybe not getting distracted while humans tend to be better at adapting which is usually important for safety to put it mildly
After all the first Tesla auto drive fatality was literally "that is probably a sign not a trunk"...
This is not supported by crash data. In Australia alcohol is involved in some 30% of fatal crashes. A lot of crashes are caused by a very small number of hoon drivers, distracted drivers etc.
Autonomous cars beating the average that includes the above are still worse than a diligent driver, which anyone can resolve to be.