Bloomberg: The Rules That Made U.S. Roads So Deadly (during the pandemic)

[jeffandnicole]

When there's a ped/bike accident reported in the news, I generally look at this:  If there's a followup story, the motorist was at fault, and the media will do whatever it can to pick on the motorist.  If there's a story that says "It's being investigated/stop back for more updates" and there aren't followup stories or updates, the ped/bicyclist was at fault, and the newspaper decided to drop the story.

Not always.  There was a case where someone was killed jaywalking near Cohoes and not only did the story get continued coverage, it was used as an example of dangerous road design and even became the centerpiece of the political push for rebuilding the road narrower with a 10 mph lower speed limit.

Basically, a road that felt slow at anything below 55 but which was signed for 45/40 was redesigned to be comfortable at 40 but signed at 35/30.  The engineering improvements are good, though with the speed limit drop the road still feels a bit on the slow side, and speed didn't kill the person whose death motivated the change - jaywalking did - but it seems like the answer to every issue on the roads these days, if you ask the advocates at least, is to lower speeds.

(personal opinion)

In cases like this, and I've seen other examples too, instead of blaming the pedestrian for jaywalking, they STILL blame others.

In a case we've talked about on these boards in the past, in Burlington City, NJ, the state actually created a law to make a specific area of US 130 25 mph at all times near a school because a student was killed, and speeding fines were tripled. https://goo.gl/maps/GwHTEY5bH9RCP2Tg7 . It was usually skimmed over that he was killed by a drunk driver after midnight walking on a sidewalk.  This article, written after the speed limit signs were posted, indicates that it's caused more issues on other streets: https://www.burlingtoncountytimes.com/news/20191117/road-of-good-intentions .  Based on what I read (the article was written in 2019), there's been way more tickets written on side roads for people trying to get around congestion than written on 130 itself, and, while not said, few if any tickets may have even been written for speeding on 130.

What I would love to see is a followup article...of the students that lead these rallies to have the speed limit reduced.  What speeds do they do thru this same area.  Have they ever received a speeding ticket anywhere?  Have they ever drank, then gotten behind the wheel of a car?  Did they care much about this issue after they graduated from school?

[kalvado]

That would require states to admit that speed limits are set too low... and if you're going to do that, why not raise the limit?

Because then people will just go even faster than the already high speed they're currently going. And since so many states are reluctant to admit speed limits are low while also refusing/being unable to post higher ones, this would provide a way to bypass that.

The idea that everyone speeds for the sake of speeding has been debunked.  People drive the speed they are comfortable at.  When that speed is well above the speed limit, the speed limit tends to be too low, and that tends to be true often enough that people may ignore a speed limit that's properly set for non-obvious reasons.  There have been studies done where states have raised their speed limits but average speeds changed by less than the increase, and 85th percentile speeds changed by very little if not gone DOWN.

Honestly, the fact that people refuse to accept that this has been debunked and continue to believe that everyone will just go faster if the speed limit is raised are part of the problem.

Whats your source for all this?

Because this is what I found from 2019

"Rising speed limits over the past 25 years have cost nearly 37,000 lives, including more than 1,900 in 2017 alone, a new study by the Insurance Institute for Highway Safety shows.

Maximum speed limits are set by the states, and they have been rising since the mid-1990s. Proponents of raising the speed limit often argue that such increases simply bring the law in line with reality, since most drivers exceed the limit. Once the limit is raised, however, drivers go even faster.

Today, 41 states have maximum speed limits of 70 mph or higher. Six states have 80 mph limits, and drivers in Texas can legally drive 85 mph on some roads.

For the new study, Charles Farmer, IIHS vice president for research and statistical services, analyzed the effect of changes in the maximum posted speed limit in every state from 1993 to 2017. Looking at annual traffic fatalities per mile traveled for each state and taking into account other factors that affect fatality rates – including changes in unemployment, the number of potential young drivers (ages 16-24) and the seat belt use rate – he calculated the effect of speed limit increases.

Farmer found that a 5 mph increase in the maximum speed limit was associated with an 8 percent increase in the fatality rate on interstates and freeways – the roads most directly affected by changes to the maximum speed limit – and a 3 percent increase on other roads. In total, over the 25-year study period, there were 36,760 more deaths – 13,638 on interstates and freeways – and 23,122 on other roads – than would have been expected if maximum speed limits hadn't changed over that time.

Of the 37,133 people who died on U.S. roads in 2017, Farmer estimates that 1,934, or 5 percent, would still be alive if speed limits hadn't changed since 1993."

https://www.iihs.org/news/detail/speed-limit-increases-are-tied-to-37-000-deaths-over-25-years


And also, you say:
"People drive the speed they are comfortable at.  "

That very well may be true. But we're talking about safety, not comfort. 

Hypothetically, a 17mph speed limit would be really uncomfortable to drive at. But you know what, I have a feeling it would be pretty safe.

A red flag: this is presented as a scientific result, but there is no link to a peer-review paper. I bet they don't expect results to pass scrutiny.

I went through some of IIHS work a while ago - I don't believe their work has actual scientific value. At least this is not the papers I looked at.

[jamess]

If bicyclists and pedestrians are getting hit where they're ultimately at fault, that's not the roadways being unsafe.

Engineering 101.

If many of your customers are using your product incorrectly, you don't have a customer problem, you have a design problem.

That's bicyclists and pedestrians failing to follow the basic signs and signals, many of which were installed for their very safety.

Perhaps if pedestrians and bicyclists were included in the design process - one of the points of the original article - they would explain why those signals and signs are not being used as expected.

In reality, if a slow walker takes 20 minutes to walk 1 mile, it would take them about 2 minutes to walk 500 feet.  No doubt they've walked longer than that to get to the point where they decided to fatally cross the street. 

Surely youre familiar with level of service (LOS).

A delay of 2 minutes is simply unacceptable.

Here is the location of a recent pedestrian fatality.
https://goo.gl/maps/dRFmjQXVMcmi6XzdA

Article:
https://www.app.com/story/news/local/emergencies/2021/03/21/pedestrian-killed-after-trying-cross-hazlet-highway/4792278001/

That is clearly a design problem. According to google, its an 11 minute, 0.5 mile walk to cross a 120 feet roadway. None of the folks who designed the road obviously traveled the area on foot toe experience their designs.

[hotdogPi]

Here is the location of a recent pedestrian fatality.
https://goo.gl/maps/dRFmjQXVMcmi6XzdA

I'm not sure if this would comply with guidelines, but if pedestrians commonly cross here (meaning it isn't a one-off incident), there should be a 4-foot gap between every third barrier to allow for pedestrians to wait in the middle. The gaps would still be raised an inch or two, though; we don't want cars that barely stray into the center by accident to hit the end of a Jersey barrier.

I'm not sure how to handle wheelchairs without making the cars vulnerable to hitting the end or putting crosswalks absolutely everywhere.

[jamess]

Here is the location of a recent pedestrian fatality.
https://goo.gl/maps/dRFmjQXVMcmi6XzdA

I'm not sure if this would comply with guidelines, but if pedestrians commonly cross here (meaning it isn't a one-off incident), there should be a 4-foot gap between every third barrier to allow for pedestrians to wait in the middle. The gaps would still be raised an inch or two, though; we don't want cars that barely stray into the center by accident to hit the end of a Jersey barrier.

I'm not sure how to handle wheelchairs without making the cars vulnerable to hitting the end or putting crosswalks absolutely everywhere.

One of the rules that make US roads so deadly (again pointing to the original article) is that the MUTCD has rules that make it very difficult to add pedestrian infrastructure due to the arbitrary warrants.

https://mutcd.fhwa.dot.gov/htm/2009/part4/fig4c_05_longdesc.htm

So unless you get 107 peds an hour crossing here (such a random number) youre not getting a signal.

An unsiganlized crosswalk wont pass muster here either, using the same MUTCD rules and common sense.

The end result is a rule book that basically says "pedestrian deaths are ok as long as they dont happen too often" which is certainly not the standard being used for motor vehicle safety interventions.

[hotdogPi]

You don't need a pedestrian warrant to install a simple median. (My proposal above assumes that the Jersey barriers are currently serving a useful purpose; if they aren't, you can just replace the barriers entirely with a median.)

It's much easier to cross two one-way lanes twice than four lanes with two in each direction once.

[kalvado]

Here is the location of a recent pedestrian fatality.
https://goo.gl/maps/dRFmjQXVMcmi6XzdA

I'm not sure if this would comply with guidelines, but if pedestrians commonly cross here (meaning it isn't a one-off incident), there should be a 4-foot gap between every third barrier to allow for pedestrians to wait in the middle. The gaps would still be raised an inch or two, though; we don't want cars that barely stray into the center by accident to hit the end of a Jersey barrier.

I'm not sure how to handle wheelchairs without making the cars vulnerable to hitting the end or putting crosswalks absolutely everywhere.

One of the rules that make US roads so deadly (again pointing to the original article) is that the MUTCD has rules that make it very difficult to add pedestrian infrastructure due to the arbitrary warrants.

https://mutcd.fhwa.dot.gov/htm/2009/part4/fig4c_05_longdesc.htm

So unless you get 107 peds an hour crossing here (such a random number) youre not getting a signal.

An unsiganlized crosswalk wont pass muster here either, using the same MUTCD rules and common sense.

The end result is a rule book that basically says "pedestrian deaths are ok as long as they dont happen too often" which is certainly not the standard being used for motor vehicle safety interventions.

ANYTHING in safety is " deaths are OK as long as they dont happen too often". It is also "deaths are OK as long as it is too expensive to avoid them"

Regarding 107 number - you realize that it is about situations when traffic doesn't warrant the signal otherwise?  And would it make you feel better if it was 100 pedestrians per hour?

[jamess]

You don't need a pedestrian warrant to install a simple median. (My proposal above assumes that the Jersey barriers are currently serving a useful purpose; if they aren't, you can just replace the barriers entirely with a median.)

It's much easier to cross two one-way lanes twice than four lanes with two in each direction once.

I agree with you, but I see that proposal being stuck in circular logic.

-Adding a gap will encourage pedestrians to cross
-It is dangerous for pedestrians to cross due to the high speed limit and traffic volume
-Ergo, we cannot add a gap for pedestrians.

ANYTHING in safety is " deaths are OK as long as they dont happen too often". It is also "deaths are OK as long as it is too expensive to avoid them"

Actually, our peers in Scandinavia which developed Vison Zero have show that this is in fact not true. Deaths are preventable. The fact that our guidance is not designed around that, and instead prioritizes traffic movement, is the problem the article is talking about.

Regarding 107 number - you realize that it is about situations when traffic doesn't warrant the signal otherwise?  And would it make you feel better if it was 100 pedestrians per hour?

Nope. If there is demand for 99 pedestrians every hour to cross here, they dont deserve a safe crossing because of arbitrary reasons?

[kalvado]

ANYTHING in safety is " deaths are OK as long as they dont happen too often". It is also "deaths are OK as long as it is too expensive to avoid them"

Actually, our peers in Scandinavia which developed Vison Zero have show that this is in fact not true. Deaths are preventable. The fact that our guidance is not designed around that, and instead prioritizes traffic movement, is the problem the article is talking about.

Yes, unfortunately education problems and prevalence of agenda over facts is not a US-only problem 

Regarding 107 number - you realize that it is about situations when traffic doesn't warrant the signal otherwise?  And would it make you feel better if it was 100 pedestrians per hour?

Nope. If there is demand for 99 pedestrians every hour to cross here, they dont deserve a safe crossing because of arbitrary reasons?

More like they don't create enough problems in a low-traffic area to warrant a light. It's an < OR > list of conditions, and summary is that if there are no pedestrians and no cars - there no need for traffic signal.

[SEWIGuy]

You don't need a pedestrian warrant to install a simple median. (My proposal above assumes that the Jersey barriers are currently serving a useful purpose; if they aren't, you can just replace the barriers entirely with a median.)

It's much easier to cross two one-way lanes twice than four lanes with two in each direction once.

I agree with you, but I see that proposal being stuck in circular logic.

-Adding a gap will encourage pedestrians to cross
-It is dangerous for pedestrians to cross due to the high speed limit and traffic volume
-Ergo, we cannot add a gap for pedestrians.

ANYTHING in safety is " deaths are OK as long as they dont happen too often". It is also "deaths are OK as long as it is too expensive to avoid them"

Actually, our peers in Scandinavia which developed Vison Zero have show that this is in fact not true. Deaths are preventable. The fact that our guidance is not designed around that, and instead prioritizes traffic movement, is the problem the article is talking about.

I believe that Vision Zero doesn't claim as a goal to prevent ALL traffic related deaths.  That's not possible in any meaningful way.

[jeffandnicole]

If bicyclists and pedestrians are getting hit where they're ultimately at fault, that's not the roadways being unsafe.

Engineering 101.

If many of your customers are using your product incorrectly, you don't have a customer problem, you have a design problem.

That's bicyclists and pedestrians failing to follow the basic signs and signals, many of which were installed for their very safety.

Perhaps if pedestrians and bicyclists were included in the design process - one of the points of the original article - they would explain why those signals and signs are not being used as expected.

In reality, if a slow walker takes 20 minutes to walk 1 mile, it would take them about 2 minutes to walk 500 feet.  No doubt they've walked longer than that to get to the point where they decided to fatally cross the street. 

Surely youre familiar with level of service (LOS).

A delay of 2 minutes is simply unacceptable.

None of these arguments make any sense.  There's no "delay" of 2 minutes.  That's just 2 minutes walking on a sidewalk.  Peds and bicyclists are included in design processes...which is how ped signals are now timed, and how bicyclists are given the very specific lanes they requested (bicyclists themselves can't agree on what design they want though).   And the road design is fine in most cases.  You say how there was a single pedestrian death...yet what about all the other pedestrians that managed to get from one side of the highway to the other?  The design clearly works...it's someone who decides they want to shortcut the system that often causes their own death.

[webny99]

I typically don't wait for the walk signal to cross; I just cross when it's clear, taking advantage of medians when possible. (It helps that the Northeast has few undivided multilane roads.) I make sure that if the cars don't see me at all, I'll still make it across. I also cross behind cars quite often so that I'm not blocking their way, even at intersections.

If I absolutely need to cross at a specific spot, and there's no traffic signal nearby, I'll step into the shoulder if it's wide enough and wait for the cars to stop (where I typically did this before I moved into an apartment, it would be almost guaranteed within the first 6 cars). I would never do this on a road that's more than one lane in each direction, though.

Is this a problem?

The first part is certainly not a problem. I often do the same.

The second part depends on the context. If it's an urban or heavily developed area, it's probably fine, but there's no expectation for the drivers to stop unless there's a crosswalk, so I wouldn't do it on a rural road or any road with a speed limit higher than 40.

[kalvado]

If bicyclists and pedestrians are getting hit where they're ultimately at fault, that's not the roadways being unsafe.

Engineering 101.

If many of your customers are using your product incorrectly, you don't have a customer problem, you have a design problem.

That's bicyclists and pedestrians failing to follow the basic signs and signals, many of which were installed for their very safety.

Perhaps if pedestrians and bicyclists were included in the design process - one of the points of the original article - they would explain why those signals and signs are not being used as expected.

In reality, if a slow walker takes 20 minutes to walk 1 mile, it would take them about 2 minutes to walk 500 feet.  No doubt they've walked longer than that to get to the point where they decided to fatally cross the street. 

Surely youre familiar with level of service (LOS).

A delay of 2 minutes is simply unacceptable.

None of these arguments make any sense.  There's no "delay" of 2 minutes.  That's just 2 minutes walking on a sidewalk.  Peds and bicyclists are included in design processes...which is how ped signals are now timed, and how bicyclists are given the very specific lanes they requested (bicyclists themselves can't agree on what design they want though).   And the road design is fine in most cases.  You say how there was a single pedestrian death...yet what about all the other pedestrians that managed to get from one side of the highway to the other?  The design clearly works...it's someone who decides they want to shortcut the system that often causes their own death.

There were two fatal accidents within a month with people trying to cross same road in the same general area.
There is a clear issue of pedestrian infrastructure along the road, if you move around in google maps. But, as far as I can tell, this is one of those roads which collect traffic at a highway and operate close to the limit - if not beyond the limit. 
Unfortunately, the way things are set up - for  example, general American public being unable to use underground cossings due to crime issues, and ADA effectively banning such crossings - there are no good ways to resolve such issues.

[kphoger]

So unless you get 107 peds an hour crossing here (such a random number) youre not getting a signal.

107 strikes me as the opposite of random.  100 would strike me as random, but 107 strikes me as a number that was reached by doing actual research and calculation.

The end result is a rule book that basically says "pedestrian deaths are ok as long as they dont happen too often" which is certainly not the standard being used for motor vehicle safety interventions.

Yes, pedestrian deaths are OK as long as they don't happen too often.  This is true.  In order to ensure roads have zero deaths, the only approach that would actually work is to physically prevent anybody from using those roads.  Humans by nature make mistakes and poor choices, and some percentage of those mistakes and decisions will result in injury or death.  Somewhere between zero and 'too many', a line needs to be drawn.  You may disagree about where the line is drawn, but the line still needs to be drawn.

[jamess]

So unless you get 107 peds an hour crossing here (such a random number) youre not getting a signal.

107 strikes me as the opposite of random.  100 would strike me as random, but 107 strikes me as a number that was reached by doing actual research and calculation.

The end result is a rule book that basically says "pedestrian deaths are ok as long as they dont happen too often" which is certainly not the standard being used for motor vehicle safety interventions.

Yes, pedestrian deaths are OK

Odd, I dont remember us covering that topic in ethics class, nor in my annual ethics recertification

[Scott5114]

You must have had a shitty ethics professor, then. They covered it in mine back in 2007.

[hotdogPi]

So unless you get 107 peds an hour crossing here (such a random number) youre not getting a signal.

107 strikes me as the opposite of random.  100 would strike me as random, but 107 strikes me as a number that was reached by doing actual research and calculation.

The end result is a rule book that basically says "pedestrian deaths are ok as long as they dont happen too often" which is certainly not the standard being used for motor vehicle safety interventions.

Yes, pedestrian deaths are OK

Odd, I dont remember us covering that topic in ethics class, nor in my annual ethics recertification

The cost of a human life is estimated at somewhere around $7 million (and non-fatal injuries also have numbers, but obviously much lower). It all just gets factored into calculations.

[Dirt Roads]

<snipped> In order to ensure roads have zero deaths, the only approach that would actually work is to physically prevent anybody from using those roads.  Humans by nature make mistakes and poor choices, and some percentage of those mistakes and decisions will result in injury or death.  Somewhere between zero and 'too many', a line needs to be drawn.  You may disagree about where the line is drawn, but the line still needs to be drawn.

Indeed.  In my world (mainly rubber-tired driverless AGT systems), the only way to achieve a zero-fatality safety certification is to prevent anything from accessing the guideway.  This not only includes passengers and maintenance personnel, but also large animals, trees, rocks, and other debris.  This means that the guideway must be fully enclosed with barrier walls and all access points must be fault alarmed using fail-safe detection.

Even with all of that, the current methodology of system safety (and related calculations) have eroded so far from the standards set in the mid-1970s that the industry now occasionally experiences some accidents that were never envisioned.  Once upon a time, it was required that an AGT contractor would need to mathematically prove that there would be no unacceptable risk accidents per every million hours of operation (for the uninformed, that's more than 100 years of continuous operations).  The rules were even more stringent for failure modes that could occur per train, per vehicle, per axle, per doorset, etc.  That paperwork was mighty expensive, but resulted in several generations of near-perfect operations (even with less stringent safety designs).

[Duke87]

Mixed bag here.

I give the author credit for doing her homework and bringing up the Green Book first, as opposed to simply dumping on the MUTCD as has become trendy. The Green Book is what actually covers geometric design of roads, not the MUTCD (which is, as the name implies, strictly focused on "traffic control devices").

The author is likely also correct that the large pickups and SUVs that have become popular in the US are one contributing factor to the rate of pedestrian fatalities seen here.

On the other hand, the author starts to falter by correctly identifying that adjustments to geometric design of streets are needed to slow traffic, but then also dumping on the 85th percentile rule. While her argument against it sounds reasonable, this is a classic case of "people don't understand how speed limits work". The 85th percentile rule, while recommended by the MUTCD, is rarely actually followed in practice because a speed limit at the 85th percentile speed is frequently in excess of what state law permits to be posted, and even in cases where it isn't speed limits are usually posted below this level due to politics.
There is a problem with the 85th percentile rule, but that problem isn't that the rule is misguided - it's that in practice it's not actually followed. We'd be much better off if it were, since this would permit lower speed limits to be posted where they are actually warranted (due, perhaps, to a road feeling deceptively safe at higher speeds) and people would pay attention. Instead, drivers have been conditioned to speed limits always being underposted and just routinely disregard them. Thus, changing the speed limit on a given road without making any other changes often fails to alter vehicle speeds much (and this is true of raising speed limits too, not just lowering them!). We've destroyed our ability to use speed limits as a safety tool by constantly misusing them to the point of dulling their effect.

And then of course, the "we need a wider group of people involved in making the rules" bit is a load of anti-intellectual garbage. If anything there is already too wide a group involved because politicians keep meddling in it instead of leaving it to the people who are experts in the field.

[kalvado]

And then of course, the "we need a wider group of people involved in making the rules" bit is a load of anti-intellectual garbage. If anything there is already too wide a group involved because politicians keep meddling in it instead of leaving it to the people who are experts in the field.

I doubt that would help. Average engineering qualification takes a nosedive as baby boomers expire...

[Dirt Roads]

And then of course, the "we need a wider group of people involved in making the rules" bit is a load of anti-intellectual garbage. If anything there is already too wide a group involved because politicians keep meddling in it instead of leaving it to the people who are experts in the field.

I've railed on this before, but sometimes local officials with really good intentions cause some of these issues.  Here in Orange County, North Carolina our planning regulations require new subdivisions over a certain size (it was 14 or more lots in my zoning area back 10 years ago) to provide for wide roadways on a 60-foot right of way with no cul-de-sac.  The hope was encourage more development creating a new network of through roads where emergency vehicles and school buses could access from both directions.  For developments that built to these standards, we now have high speed thoroughfares posted at 25 mph and cobbled with speed tables or speed bumps to try to keep traffic "calm".  Fortunately, Orange County residents are for the most part very sensible, so high speed chases through neighborhoods are still very rare.