Hacker News Reader

Comment by ameister14 | Hacker News Reader

In the US in 2013, there were 236 pedestrian deaths of children younger than 14. Thats under 5% of pedestrian fatalities, and the vast majority of pedestrian fatalities are in the dark, with half involving alcohol.

There's about a 17% risk of death if you're hit by a car going 30 mph, vs 12% at 25 mph. This doesn't seem like as big an issue as you're making it out to be.

Your slippery slope argument doesn't hold up - people don't go 45 in a 25 on a regular basis. Almost every driver goes 5-10 miles above the speed limit on the highway and 0-5 miles above everywhere else. The limits are designed with this in mind.

These things can get really personal, but I don't believe the law should.

Nobody thinks it's cool to run through a red light. People go through yellows because that is legal to do.

View on HN

This is a reasonable reply and I appreciate the very relevant data points.

I think you're correct that speed limits are designed knowing that most people will exceed them slightly. I'm curious if there are any studies that have addressed this? Eg. traffic surveys where they either raised or lowered a posted speed limit -- how much impact does that have on behavior?

Sure. https://www.ibiblio.org/rdu/sl-irrel.html

Key points:

Lowering speed limits by 5, 10, 15, or 20 mi/h (8, 16, 24, or 26 km/h) at the study sites had a minor effect on vehicle speeds. Posting lower speed limits does not decrease motorist's speeds.

Raising speed limits by 5, 10, or 15 mi/h (8, 16, or 25 km/h) at the rural and urban sites had a minor effect on vehicle speeds. In other words, an increase in the posted speed limit did not create a corresponding increase in vehicle speeds.

Or another way to state this is that people generally drive at a speed they think is reasonable given the road conditions and mostly ignore posted speed limits altogether.

No surprise to me, many states didn't have speed limits outside of town at all in the US until the federal government withheld highway funding until they posted speed limits. That's the reason the 55mph highway speed limit was originally established. It's also the same method by which we forced the drinking age from 18 to 21.

Given a population of people, some will be reasonable in their behavior, some will not be. Setting limits on that behavior is more of a signal (or suggestion) than anything that really changes how people behave in the large. To consider otherwise is to be naive of the reality.

Fun story, data shows that raising highway speed limits results in less accidents, but the accidents that do occur result in more fatalities.

There are some good studies that basically point to lane width , and road condition being determining factors on speed. E.g. narrower lanes, slower traffic.

The general term for integrating these mitigating factors into roadway design is "traffic calming".

I read their methodology[1] to determine when they measured the speeds, but it wasn't very clear when they did. However, I think a reasonable interpretation of their wording is that they measured just before and just after the change.

In my opinion there is a problem with this approach, in that many motorists will likely pay little attention to signs when going their usual routes, as they already know them all. If a speed limit is changed it may therefore take some time for people to notice that that has happened. To measure the full impact of the change, it seems the measurement would have to be taken a few months after the change.

[1] https://www.ibiblio.org/rdu/sl-irrel/method.html

http://safety.fhwa.dot.gov/speedmgt/ref_mats/fhwasa1304/Reso...

Sorry about that - this the full study, not just the report.

Iirc, its related to the design of the road.

Think of the speed limit as a fine tuning lever and the road itself as the control. If you slap a 30 mph limit on one of those 6-lane boulevards that you see a lot in Florida, there's little change. Likewise, the 70-75 limits on I95 in South Carolina don't really have a material impact on the traffic flow.

Road engineers can slow a road down with all sorts of techniques, and they absolutely do studies and instrument the road to tweak the design.

> This is a reasonable reply and I appreciate the very relevant data points.

Faith in humanity restored. Thanks, this line made my day. Hope I can have the to use that same spirit today.

> People go through yellows because that is legal to do.

IIRC (and this may vary state to state), it actually isn't legal to go thru a yellow. From what I recall (and I'll admit I may be wrong), the actual law is that if you are before the "stop line" or intersection, and the light turns yellow, you are supposed to come to a complete stop. If, however, you are already in the intersection (your front bumper is beyond that "line"), you can proceed with caution, provided you clear the intersection before the light turns red.

Again - I may be wrong about this, and concede that it could vary across jurisdictions.

> Almost every driver goes 5-10 miles above the speed limit on the highway

That's really only the case in states that have kept their NMSL (National Maximum Speed Limit) era laws on the books. In states that have speed limits 75 mph or above, you're not going to find traffic flowing at 85 to 90 mph. Conversely, in states where they still have large sections of highways still posted at 55 mph, you will routinely see traffic flowing at 70 to 80 mph.

> and 0-5 miles above everywhere else.

I have seen plenty of instances where people drive at 35 to 40 mph in 25 mph zones. These are roadways that were formerly posted at 35 mph, but the town, for some odd reason decided to reduce the speed limit to 25 mph.

A 40ish% increase in risk cannot simply be handwaved away. That's huge.

I think what he meant is that pedestrian child deaths number (especially in the conditions that the only problem was speeding) is so small in comparison to total traffic child deaths that increasing it by 40% percent is still almost nothing.

So the argument "don't speed no matter what or you might kill a child that would otherwise not die" while technically true is not particularly strong.

I think the existence of speed cameras and red light cameras are because it was the problem that could be automated easily. Automatically detecting a car failing to yield to pedestrians, backing up on freeways, etc, until recently was too difficult of a problem to automate at scale. I think that has probably changed. If someone is breaking one of these things it is a proxy for other dangerous driving behavior.

The flip side to this is instead of being used to give tickets this technology will be used to drive the car. On the other hand, now drivers are being invisibly ticketed in the form of higher insurance rates by their car insurance companies collecting data on their driving behavior, and they don't even know it.

If you run down the street waving a gun at people, there is a high chance the police are just going to shoot you. Yet, if you have a car and pedestrians or bicyclists didn't get hit because they dodged you, it is considered normal behavior. It isn't.

I think this analysis (without any citations) is like big O notation. Sure, 17% is not that large compared to 12%, except there's a hidden constant: the value of life, which makes that 5% quite significant.

If you can show that bringing the speed limit down to 20 from 25 will save lives, I'll get on board.

I don't think it will, though - the data I have read says it will have little to no effect, and actually could have a negative effect. Most residential limits are already abnormally low in terms of safety from a traffic engineering perspective.

http://safety.fhwa.dot.gov/speedmgt/ref_mats/fhwasa1304/Reso...

The chance of dying from being hit by a car at different speeds is not the only factor, when you are going 20, you will have more time to react, the person about to be hit has more time to react and the distance you'll travel while braking is less.

So it would probably save lives, and while 1 or 2% doesn't seem like much, the parents of those childs would beg to differ.

> So it would probably save lives, and while 1 or 2% doesn't seem like much, the parents of those childs would beg to differ.

Not sure if an appeal to emotion is going to work too well on HN. While I certainly sympathize with those parents this is the absolute wrong way to go about implementing policy or law. Most of US law that has eroded our rights over the years has been enacted by similar emotional arguments. I prefer my policymakers to be sober dispassionate people who look at raw data to make as an informed decision as possible.

> Not sure if an appeal to emotion is going to work too well on HN.

Killing a human being is not the same as killing a process. We live in a society because it's safer. We have a constitution, rights, and laws to protect us (at least in theory). The optimal law would save as many lives as possible. I don't see how saving lives makes a poor heuristic for implementing laws.

>The optimal law would save as many lives as possible.

If that were true, cars would be illegal.

We're operating in a framework of likely accidents, benefit to society and overall cost. There is an optimal law that saves as many laws as reasonably possible, and in most cases it has already been implemented and is being enforced. Where it has not or is not, sure.

My younger brother was hit by a car when crossing the street. Although he didn't die, he sure came close and was in a coma for over a week.

The driver was doing 40 in a 30 (mph) and was on his cell phone.

I still don't think every traffic law should be modified to prevent this scenario.

It's silly to emotionally invest in the law. Preventing all traffic deaths is just not realistic currently.

"was on his cell phone" is a much more important detail than the driver's speed.

While you're talking about speed limits, my post was about actual car speeds, and how a difference of 5% (from your data) means a great deal when it refers to your survival chances.

> There's about a 17% risk of death if you're hit by a car going 30 mph, vs 12% at 25 mph.

While I understand that, I wasn't just talking about that statistic independently. I was responding to a person who brought up the 'think of the children' argument and isolating them as a smaller group. The total number of fatalities amongst children we're talking about within that speed is quite small and potentially 0.

That's why I brought up the effectiveness of change in posted speed.

But with so few deaths overall it does become insignificant.

Reply on news.ycombinator.com