Comment by sivex | Hacker News Reader

Comment by sivex | Hacker News Reader

Parent Full thread

Roadway deaths fall 8%, number of miles driven falls 18%, fatalities per mile increases 14%... What is noteworthy about this? Do most accidents occur near intersections or in urban areas rather than on highways? Where is the data on where these accidents happen? There are a lot less people commuting far distances due to the increase in remote work - and now these statistics are based on fewer people taking lower mileage trips. To go to the grocery store, I need to drive 3 miles and I go through 18 intersections. To go to work I'm driving 50 miles and going through 12 intersections. Intersections create scenarios where some vehicles are going 50mph faster than those stopped or turning. With some location data added, would this show that fatalities are higher in areas where there are larger concentrations of vehicles going drastically different speeds? Fender benders on the highway aren't going to cause deaths. Being T-Boned or being hit by someone running a red light....

I think you're right that it's the intersections/interactions. I commute ~50mi day across varied roads. The intersections on the "not a totally limited access highway but pretty close" seem to be where most of the stupid stuff happens. It's not any different than the normal stuff but now the speed differentials are higher so stupid has the potential to hurt more.

Most people drive around what's reasonable for the road. Some people drive insanely fast and some people drive whatever the speed limit is, reasonableness be damned. With normal traffic the latter group combined with the sheer volume of cars limit the ability of the two former groups to have large speed differentials with any of the other two groups.

Now with 'rona traffic volumes you've got the people that want to drive 100, weaving around the people who want to do 70-85 while the idiots who think it's acceptable to merge at 30mph less than traffic speed do just that. The net situation is just one unsignaled lane change away from a crash.

That said, I've seen a lot less aggressive passing because the roads are so empty that you get lots of chance to pass people so you don't need to take a crappy one.

There's also the fact that any change from the norm is going to increase fatalities. Everyone's driving skills, instincts, muscle memory are tuned to the driving conditions they expect on a particular road at a particular time. Change anything and they make more mistakes than usual.

Exactly. And TFA jumps to conclusions about speeding and reckless driving with only some quoted speculation. FWIW, your speculations make more sense to me.

Your speculation about "fatalities [being] higher in areas where there are larger concentrations of vehicles going drastically different speeds" is consistent with what data I've seen. And it's not just about intersections. In some states, ~official policy is that drivers ought to move with traffic, even if traffic is moving faster than the speed limit.

Edit: carelessness

It would be interesting to know if the number of collisions per mile has increased, or if the lethality of collisions has increased, or both.

Reply on news.ycombinator.com