Hacker News Reader

Comment by AStellersSeaCow | Hacker News Reader

FTA: "Fortunately, American roadways are safer than ever, with highway fatalities at historic lows. Roads can be dangerous, but the perception of roads getting increasingly dangerous is a false one." ... "Published Jul 23, 2014"

And then: https://www.nhtsa.gov/press-releases/traffic-fatalities-shar...

View on HN

> Almost half of passenger vehicle occupants killed were not wearing seat belts. Research shows almost one in three fatalities involved drunk drivers or speeding. One in 10 fatalities involved distraction.

Someone has a pretty big axe to grind when they have to lump speeders in with drunk drivers in order to get the statistics they need... I suspect that the number is almost the same even if you remove the non-drunk speeders from it. (I'm being a little snotty and disingenuous--I suspect that the issue is that the number of non-drunk, speeding fatalities is almost nil and that drunk and fatality almost always implies speeding as well).

And I still can't believe how many people die because they won't put on a bloody seat belt. Sheesh.

So, what we know is that using electronic devices in cars is more dangerous than speeding. So, we clearly need to remove the cellular towers near highways, right?

Indeed. Looking at the local accident statistics and investigation reports where I live (Finland), it is easy to see that small breach of speed limit is fairly negligible as a cause for accidents. In some part of accidents it is one contributing factor, but there are quite large groups where speed limits are not relevant at all:

- suicides (somewhere between 10 % and 20 %) - spontaneous deaths (around 10 % of traffic deaths, driver dies of old age or similar; they are included in statistics even though the accident was not actually cause of death) - gross negligence (at least 30 %: the driver is severely drunk, does not have a license, the vehicle has not passed inspection and the driver is trying to escape the police at 160 km/h on a 50 km/h zone; sometimes you have all these at the same time!) - freak observation errors (at a rail crossing, just drive under the train)

> American roadways are safer than ever

Well, even then, the US still has somewhere around 3-4 times more deaths on the roads per capita than other western countries.

Per capita, or per miles driven? My limited experience is that people in the U.S. drive a lot more in general, so per mile driven would be a more meaningful measure of road safety.

IIRC it's still like 30-40 % higher when considering distance.

Some of this may be due to the skillset to obtain a license in most states in the US is a quick written test and a 15 minute driving test in a parking lot, whereas other countries require actual studies, hand-on road instruction, and about 30x more cost ($32 for Georgia vs ~$1000 for many European countries).

To be fair, distracted driving is a larger problem for fatalities than speeding. Also, I think we probably have far more than 3-4 times the people who drive and the corresponding roadways than most other western countries.

It would be interesting to know what the ratio of deaths per driving regularly capita is for all western countries.

To be fair, that has more to do with everyone now having mini super-computers in their possession distracting them than with speed limits.

And even at the historic lows, numerous studies suggest fatality rates would be even lower if we'd stuck to the national 55/65 speed limits.

55? That's ridiculous! Sure, it'll save a few lives, but millions will be late!

You jest, but given that the average lifespan is about 622k hours, I think you can have a serious moral argument about whether it's better to kill one person or make a million people late by an hour.

Ooo, this will be a good one for the ethics panel. How many of those million people could have saved, or ended a life in that time? Its a juicy one.

I think the argument rests on the fact that on net, people generally do positive things with their time.

If the break-even point isn't 1 million people, where is it? 622k? 100 billion? 3^^^3?

This is Eliezer Yudkowski's dust speck argument all over again. http://lesswrong.com/lw/kn/torture_vs_dust_specks/

In my experience, all this does is create a dangerous speed differential between the people who go 80 regardless of the posted speed limits, and the people who follow the speed limits with room to spare.

And it's the large differentials in speed which create the most dangerous conditions and accidents.

Even rigorous enforcement can't catch everyone, and the high speeds quickly become the norm; with people going 65 the dangerous exceptions.

Your experience is directly contradicted by what people studying this observed as the national limit was relaxed and then removed completely.

Rates would be lower if we all drove at the same speed, no matter what that speed is. Setting the speed limit at 55/65 doesn't make everyone drive at 55/65.

That may be true, but doesn't change the fact that we've done this experiment nationally, and adjusted for other factors that are known to affect speed, there were fewer fatalities when the limit was lower.

Setting the limit at 80 doesn't make everyone drive 80, either.

That's why you set it to the 85th percentile speed, as the article explains, rather than picking some arbitrary speed and expecting everyone to go at that speed.

> And even at the historic lows, numerous studies suggest fatality rates would be even lower if we'd stuck to the national 55/65 speed limits.

Evidence, please.

We did this experiment and the evidence is against you.

Fatalities did not significantly increase for any of the states that increased their limits after the 55mph thing ended and there was some weak evidence that they went down.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2724439/ https://www.theguardian.com/news/datablog/2013/may/13/speed-... http://www.livescience.com/33202-higher-speed-limits-cause-m... http://www.jsonline.com/story/news/2016/09/23/crashes-spike-... http://www.iihs.org/iihs/news/desktopnews/speed-limit-increa...

Did you even read the report, or did you simply pull a summary? The numbers are all over the map and the P values look hacked to generate a paper.

In addition, if you look at the fatality and injury numbers for urban interstate they went down for 65MPH and up for 55mph and 75mph. That makes no sense.

Basically, the statistics are sufficiently close to a wash that the confounding factors (like texting or rerouting due to more convenience) matter more than the speed limit.

Not only that, but there is a study [1] that shows that traffic speeds do not change significantly even if the speed limit dies have a significant change.

[1] http://www.ibiblio.org/rdu/sl-irrel/

Reply on news.ycombinator.com