4 ms·
The 94 million mile mean is for all types of driving, in all types of situations, all types of people, in all types of cars. First and foremost, highway drivin
by rhino369 10y ago
The 94 million mile mean is for all types of driving, in all types of situations, all types of people, in all types of cars.
First and foremost, highway driving is much much much safer than regular driving. There are no intersections (which cause 40% of all accidents), there is high viability, slow turns, no pedestrians, and a super simple traffic pattern. Autopilot is only doing the "easy" driving.
Second, that 94 million miles figure includes motorcycle deaths. Motorcycles have 34X more fatalities per mile driven. That juices the number of fatalities per mile.
Demographics matter too. Telsa owners are disproportionately older than average (not a lot of teens driving Teslas). They are more educated and richer. These people tend to be safer behind the wheel on average.
Telsa's also have a lot of safety features that should reduce fatalities to way below median anyway.
If you want to test if Autopilot is safer you have to test it against a control group that is the same but just didn't use Autopilot.
- gnicholas 10y agoCompletely agree with most of these points, but the demographics one is actually a mixed bag. Fatalities plotted against driver age follow a U-shaped curve [1], with younger and older drivers being the most likely to cause a fatality. The U is higher on the younger driver side of the curve, but this is likely due in part to the fact that teens have less money and therefore drive cheaper/less safe cars. Because Tesla owners probably skew older, we might expect to see slightly more fatalities with Teslas than with other vehicles. The male/female fatality rates [2] are also relevant, and I would guess that Teslas are driven more by men than women. (My estimation of Tesla demographics is based on my commute down Sand Hill road, so not scientific but probably the highest-traffic area for Teslas.) Lastly, I haven't seen data that wealthier people are safer drivers than poorer people. Source? It's believable, but I have looked in the past for data on this and not seen this particular conclusion. 1: https://www.aaafoundation.org/sites/default/files/2012OlderDriverRisk.pdf https://www.aaafoundation.org/sites/default/files/2012OlderD... 2: http://www.cbsnews.com/news/men-vs-women-who-are-safer-drivers/ http://www.cbsnews.com/news/men-vs-women-who-are-safer-drive...
- jsprogrammer 10y agoWe don't know if we have accurate fatality numbers for Teslas though. IIRC, the first fatality was unreported for a couple months.
- CardenB 10y agoThank you! I've been saying this since the accident happened and I'm surprised I never hear this discussion