3 ms·

> This is consistent with previous reports from Waymo. If one of their vehicles gets in an accident, it's extremely likely that the other (human) driver was at

> This is consistent with previous reports from Waymo. If one of their vehicles gets in an accident, it's extremely likely that the other (human) driver was at fault. It looks like in 2018 Arizona had 1.4 crashes per million miles https://azdot.gov/sites/default/files/news/2018-Crash-Facts.pdf https://azdot.gov/sites/default/files/news/2018-Crash-Facts.... and Waymo released data that they had 2.9 crashes per million miles, over twice as many. 7.7 if they hadn't had human intervention, almost at another order of magnitude than the average driver. That's rough. I'm really not interested if the other driver is almost always at fault, that's not the number to get down. Their mid-term goal seems to be to drive on public roads with mostly human drivers, so they need to get better results. If they're getting bad results over lots of data, it's not just bad luck: the way they're driving is causing more crashes, even if they're not at fault. Additionally, fault classification is worrisome, since there is the potential for bias. Various factors might make this remark wrong - "Crashes per mile" isn't the right metric: injury, fatality, or similar should replace or augment it. - I compared data from Arizona to data from Waymo's zone, which might be more crashy. - Crashes in the doc I linked underestimate total crashes, since many go unreported. - I am not comparing apples-to-apples -- new commercial minivans might be far more crash-prone than the average vehicle.

marwatk 6y ago
This report [1] seems to imply that around 35% of minor crashes aren't reported which is significant, but not enough by itself to move the needle too much. Another factor is highway vs surface miles. Highway miles would seem significantly safer on a per-mile basis (but I can't find any numbers to back that up), and with Waymo excluding highways for now they'd be all surface miles. I can't find data that breaks down crash rate by street type, but I suspect surface traffic has a much higher crash rate than highway miles. [1] https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812183 https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/...
- dmoy 6y ago
  Freeway-style fatality rates tend to be way lower per vehicle mile, yes. Freeway is about 0.3 per 100 million, and the aggregate total is 3x higher than that around 1-1.2 per 100 million.
RivieraKid 6y ago
To put Waymo's 7.7/million accident rate in perspective, it would mean an average driver, who drives 13,500 miles in a year, has one accident in 10 years. In other words, I don't think this is a concern at all. It's similar to humans already and will only get better. I think economics and scaling are the main challenges and unknowns for Waymo. I hope they will soon finish phase 1 (perfect self-driving in a small area) and start pahse 2 (expand as fast as possible).
- caturopath 6y ago
  It sounds like it's not similar to humans in Arizona according to the Arizona DOT, although I find it surprising if the average Arizonan only gets in one accident about every 50 years. I wonder if a huge amount of the driving are things like OTR trucking and have a low accident rate compared to the average civilian commuter.
  - RivieraKid 6y ago
    The 2.6 per million miles is just the lower bound. From Waymo's blogpost: > The majority of incidents in our report fall into this category of typically unreported minor collision or contact.
asdfasgasdgasdg 6y ago
I would argue that the type of crash matters quite a bit too. Low speed delta rear end fender benders are one of the most benign forms of traffic accident.
jniedrauer 6y ago
Have you driven in Arizona? There are thousands of miles of road where you can be almost completely alone. Waymo is not driving on those roads. I think you'd need data on human-caused crashes within the test area to have a reasonable comparison.
- caturopath 6y ago
  I have - I've driven across it on the highways and in the Phoenix metro area. One thing that I noticed was that when I was in Phoenix, there were lots of people...most of the people in Arizona, actually. My hope in matching up those statistics was that they did most of the driving too, so it wasn't too apples-to-oranges, but of course I did bring up this factor that made my comparison suck. I would have loved to find the right statistic to compare to--did you have any luck finding it?

marwatk 6y ago

This report [1] seems to imply that around 35% of minor crashes aren't reported which is significant, but not enough by itself to move the needle too much. Another factor is highway vs surface miles. Highway miles would seem significantly safer on a per-mile basis (but I can't find any numbers to back that up), and with Waymo excluding highways for now they'd be all surface miles. I can't find data that breaks down crash rate by street type, but I suspect surface traffic has a much higher crash rate than highway miles. [1] https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812183 https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/...

dmoy 6y ago
Freeway-style fatality rates tend to be way lower per vehicle mile, yes. Freeway is about 0.3 per 100 million, and the aggregate total is 3x higher than that around 1-1.2 per 100 million.

RivieraKid 6y ago

To put Waymo's 7.7/million accident rate in perspective, it would mean an average driver, who drives 13,500 miles in a year, has one accident in 10 years. In other words, I don't think this is a concern at all. It's similar to humans already and will only get better. I think economics and scaling are the main challenges and unknowns for Waymo. I hope they will soon finish phase 1 (perfect self-driving in a small area) and start pahse 2 (expand as fast as possible).

caturopath 6y ago
It sounds like it's not similar to humans in Arizona according to the Arizona DOT, although I find it surprising if the average Arizonan only gets in one accident about every 50 years. I wonder if a huge amount of the driving are things like OTR trucking and have a low accident rate compared to the average civilian commuter.
- RivieraKid 6y ago
  The 2.6 per million miles is just the lower bound. From Waymo's blogpost: > The majority of incidents in our report fall into this category of typically unreported minor collision or contact.

asdfasgasdgasdg 6y ago

I would argue that the type of crash matters quite a bit too. Low speed delta rear end fender benders are one of the most benign forms of traffic accident.

jniedrauer 6y ago

Have you driven in Arizona? There are thousands of miles of road where you can be almost completely alone. Waymo is not driving on those roads. I think you'd need data on human-caused crashes within the test area to have a reasonable comparison.

caturopath 6y ago
I have - I've driven across it on the highways and in the Phoenix metro area. One thing that I noticed was that when I was in Phoenix, there were lots of people...most of the people in Arizona, actually. My hope in matching up those statistics was that they did most of the driving too, so it wasn't too apples-to-oranges, but of course I did bring up this factor that made my comparison suck. I would have loved to find the right statistic to compare to--did you have any luck finding it?