NTSB: Speed Kills, and We’re Not Doing Enough to Stop It

The federal safety watchdog says speeding is a big blindspot in America's road safety efforts. To change that, it's issuing a new set of recommendations for federal and state officials.

More than 112,500 people were killed in speeding-related crashes from 2005 to 2014 . Image: NTSB
More than 112,500 people were killed in speeding-related crashes from 2005 to 2014 . Image: NTSB

More than 112,500 people lost their lives in speed-related crashes from 2005 to 2014, accounting for 31 percent of all traffic deaths in America over that period. In a draft report released earlier this week, the National Transportation Safety Board says excessive speed is a deadly problem in our nation’s transportation system — one that federal and state officials aren’t doing enough to address.

The recommendations presented at an NTSB board meeting on Tuesday represent a breakthrough for the agency, which is known for investigating plane and railroad disasters. Traffic crashes are less spectacular but cumulatively much more dangerous, claiming tens of thousands of lives each year. When the agency does turn to road safety, it has tended to focus on impaired driving and seat belt use. A thorough look at the dangers of speeding and how to prevent it is a welcome departure.

“The simple truth is that speeding makes a crash more likely. In a crash that’s speeding related, you’re more likely to be injured, your injuries are more likely to be severe, and you’re more likely to die,” NTSB Acting Chairman Robert L. Sumwalt said in his opening statement at the board meeting. “And that’s true whether you’re the speeding driver, another driver, a passenger, a bicyclist, or a pedestrian.”

Unlike driving drunk or not wearing a seat belt, which have been the subject of sustained national attention for decades, there’s little stigma against driving too fast. “Speeding has few negative social consequences associated with it, and does not have a leader campaigning to increase public awareness about the issue at the national level,” Sumwalt said. “We hope that this study helps to bring the issue of speeding back into the spotlight.”

The NTSB’s 19 recommendations should be a wake-up call, especially to the Federal Highway Administration (FHWA), state legislatures, and transportation and police departments across the country.

First off, the NTSB says, the way transportation engineers around the country have been setting speed limits for decades is causing more problems than it solves. Typically, engineers measure how fast drivers are traveling on a road, then set a speed limit so that 85 percent of drivers fall below it, and 15 percent of drivers are above. It’s called the 85th percentile rule. The NTSB points out that it’s based on research from rural roads in the 1950s and aims to accommodate speeding drivers, rather than reduce crashes.

In recent years, states have put the pedal to the metal. Image via NTSB
In recent years, states have put the pedal to the metal. Image via NTSB

The board told the FHWA it should replace the 85th percentile rule with guidelines proven to improve safety. “In general, there is not strong evidence that the 85th percentile speed within a given traffic flow equates to the speed with the lowest crash involvement rate,” the NTSB says. “Alternative approaches and expert systems for setting speed limits are available, which incorporate factors such as crash history and the presence of vulnerable road users such as pedestrians.”

The way motorists think about speed also needs to change. Using GPS or sign-reading sensors, cars can now alert drivers when they’re speeding, and even prevent motorists from exceeding the limit. The NTSB said the availability of these features should be included in car safety ratings, but didn’t endorse them as mandatory equipment.

The board did, however, have strong recommendations backing more speeding enforcement — and camera enforcement in particular.

Studies consistently show speed cameras to be effective at reducing fatalities and injuries, but they’re allowed in only 15 states and the District of Columbia. A total of 28 states have no laws enabling speed cams, and another seven prohibit them. The NTSB urged states to enable automated enforcement on their roads, and said states that already have speed cams should “remove operational and location restrictions” that prohibit their use. New York state, for example, limits speed cameras to a select number of school zones and allows them only to be operated during specified hours — a restriction that achieves no safety objective.

There’s also a need for better speed camera technology, the NTSB says. The United States relies on fixed site cameras, but “point-to-point” enforcement, which tracks vehicle speed over a greater distance, has proven effective in Europe, Australia, and New Zealand and should be used and evaluated here.

Traffic safety advocates hailed the NTSB’s report.

“We hope this report will continue to bring attention to the important yet often overlooked role of speeding in traffic injuries and fatalities for everyone who uses our streets,” said the National Complete Streets Coalition.

“For too long, proven policies to reduce speeding in our communities have been held hostage by outdated standards, costing more than 10,000 lives lost each year,” said Leah Shahum, director of the Vision Zero Network. “We urge every state and community to adopt NTSB’s recommendations to stem the tide of preventable suffering on our roadways.”

“Speed is just as significant a factor in traffic deaths as alcohol,” said Jake Nelson, director of traffic safety advocacy and research for the American Automobile Association. “AAA hopes the NTSB report and recommendations spur a constructive and long overdue discussion about how to curb speeding and improve safety on our nation’s roadways.”

  • Urbanely

    Does anyone have any information on the role of vehicle weight in these crashes? SUVs and trucks are very popular in the U.S., and they tend to be heavier vehicles. Many of the reports of fatal crashes that I read in NYC involve these larger vehicles as well.

  • Larry Littlefield

    This is the one thing we’ll get out of AVs. There should be a mandated 5 mph speed limit when turning a corner at an intersection.

  • Vooch

    Actual crash statistics suggest SUVs are some of the deadliest vehicles around for their occupants.

  • Vooch

    5 mph ?

    that’s too fast – more like 2-3 mph max

  • lol

    A full stop and 15-second wait. Then 2mph until around the corner.

  • 1ifbyrain2ifbytrain

    Just as NYC has no right on red NYC should have no turns without a full stop.

  • Wells

    Autonomous vehicle speed limits (a driver may go slower but not exceed) and emergency braking are the main safety benefits the technology offers. Eliminating the steering wheel, accelerator and brake pedals eliminates the safety devise of driver assessment of road conditions and ability to take precautionary and evasive actions to avoid accidents. Self-driving technology alone therefore cannot be considered as safe as possible, nevermind the get rich quick schemes devised and promoted by Uber and Lyft road predators.

  • Ludicrous. 15 extra seconds at every street corner will just run up freight costs. Where there are a lot of peds and bicyclists that ignore common traffic and crosswalk laws maybe there needs to be some targeted enforcement of that problem too. Most suburban intersections are designed for much-higher speed turns especially here in Colorado where we have continuous right turn and acceleration laws.

    Not every city is like NYC, Boston, Philadelphia, or Washington, DC with their narrow roads and an immense amount of pedestrian traffic.

  • Loaded 18-wheel trucks take an average of 3 times the distance to stop as does the average family sedan at the same initial speed. SUVs and pickup trucks can take 50% longer to stop than some small sports cars with 4-wheel vented disc brakes. If you want to increase safety perhaps all cars should have relatively similar braking distance, as lots of 1970s full size cars with drum brakes take twice as long to stop from the same speed as new smaller sports cars these days.

  • Vooch

    agreed; in the boonies – people’s lives are worth all that much, let the traffic carnage continue.

    especially if it’s to save 10 seconds on a delivery route

  • Vooch

    great idea – reduce stopping distances by reducing speed limits on city streets.

    30 MPH on collectors
    20 MPH on small streets

  • ZeroVisionPhila

    Trying to SELL SPEED! Because it’s the Has the highest profitability. Of anything else you can use Citation for Taxation$ I never seen anything on Liquid Courage and Chemical Powers and distraction. Because the only profits to be made are in ones the funeral business. Not in the photo enforcement industries as well as all of government. Plus it fur fills all the agendas.

  • Joe R.

    If there are state maximum speed limits then that means on some roads the limit isn’t set at the 85th percentile. A good example I’m familiar with in the NYS Thruway. The 85th percentile speed was easily 80 mph two decades ago, which is the last time I rode on it as a passenger. Generally due to improvements in automotive technology highway speeds increase roughly 5 mph per decade. That would put the 85th percentile at maybe 90 mph now. NYS has a 65 mph maximum state speed limit. The NTSB should do some fact checking before making recommendations. The fact is unless we’re talking local streets or two-lane country highways, the speed limit is most likely legislated because the 85th percentile is well above the state maximum speed limit. Why should states be allowed to have maximum speed limits anyway?

    Numerous studies have pointed to the introduction of the national 55 mph speed limit as the start of drivers ignoring first speed limits, and later all traffic laws. While I certainly support the idea of designing local streets which run through cities for lower speeds, this report is off-base if it’s recommending we reduce highway speed limits even further from their already too low legislated speed limits. Most carnage happens on either two-lane country highways or urban streets. That’s where the focus should be. Maybe if we set speed limits properly on highways drivers would eventually start respecting speed limits everywhere. Or at least it seems to work in Europe. Normal highway speed limits there are typically 120 kph to 160 kph. Compare this to the US. If speed really killed then Europe would have a higher death rate than we do. It doesn’t. Perhaps they’re doing something right. The entire “speed kills” thing is something mostly backed by police officer unions (it creates jobs for cops) and cash-starved municipalities (when 90% of the drivers are exceeding the speed limit the highway is a virtual ATM).

  • Johnny

    Well said. Giving someone a wide open road constrained by a speed limit is a bit like handing them Pandora’s box. Of course they’re going to open it. In urban areas, the solution lies in taking away the wide open road. Use roundabouts, on-street parking, and narrow street designs for the residential streets.

    In some suburban developments, I kid you not, I’ve seen street widths of 50 ft. With a 25 mph speed limit. Madness!

  • Corvus Corax

    Most carnage happens on either two-lane country highways or urban streets.

    . Really? Do you have a citation for this sweeping claim?

  • Corvus Corax

    What does the width of a road have to do with it? And what city does not have a lot of pedestrian traffic?

  • Joe R.

    See table 2 on page 102 here: http://faculty.wcas.northwestern.edu/~ipsavage/104-09.pdf

    Interstates have a much lower fatality rate than other types of roads. Rural roads of all types have the highest fatality rates by far.

  • John Lindstrom

    Trump voters apparently are more apt to approve, or need to speed. A BLUE state, Oregon, has a 65mph limit on I-5 in the Willamette Valley corridor from Eugene to Portland. New England (BLUE) also has lower speed limits. Is there a political correlation?

  • LinuxGuy

    All people who like speed cameras should read this:

    https://www.motorists.org/blog/ntsb-wants-turn-highways-profit-police-state/

  • Old Town Outsider

    Motorists.org and the National Motorists Association are pro-car and pro-speed. Of course they’re going to write that article. That doesn’t make it true. Most of us prefer to be pro-safety. People’s lives matter. Speed kills. Drive slower so less people die.

  • LinuxGuy

    The NMA is pro-everyone, and has a Streets That Work program geared towards cities. Did you pull it up? Nobody is pro-speed, but setting realistic 85th percentile speed limits increases safety. Why would you oppose that? The blog entry has a slew of facts, so are they all wrong too? The NMA must be super powerful then to somehow coerce lots of independent sources to put out bogus data, right? How could that be done? You obviously missed the decades of research that shows trying to force people to drive slowly increases the crash rates. Please also keep in mind that motorized vehicles bring everything you buy to the stores, even bicycles. All you need to do is look at what happened when states were allowed to set their own speed limits. All the insurance companies were whining, along with interest groups, but they were all wrong, the roads have never been safer.

    Along with setting proper speed limits, making yellow lights longer, and using stop signs only where needed would help.

    How many crashes may be caused by non best-practice engineering, ticket cameras, and overzealous manned enforcement? Cases of this have been well-documented.

    If anti-car people were pro-safety, then why do they support policies which increase crash rates, cite safe drivers, cite the wrong drivers, and take away constitutional rights?

    People in cities have many choices. If they do not want to drive or use bicycles, they can take mass transit, get motorcycles or mopeds, etc.

  • Old Town Outsider

    NMA is pro-car and pro-speed. It is not concerned with safety of pedestrians in any way. Here is their mission statement. https://uploads.disquscdn.com/images/33213f36f9a18fb1d9e993f8291076b999027ceddb3eeeef8d1d9ef2fb7ce44d.png

  • LinuxGuy

    But speed causes somewhere below 5% of crashes, and I have seen it as low as 0.7%, so what is the true goal here?

    Here is a federal study that disproves everything you said. I highly doubt the NMA is powerful enough to scare the people behind this to change their numbers.

    http://www.ibiblio.org/rdu/sl-irrel.html

  • Old Town Outsider

    LOL…

    That study says changing speed limits doesn’t change driving behaviors. It says nothing about safety or pedestrians.

    Quote:
    “The results of the study indicated that lowering posted speed limits by as much as 20 mi/h (32 km/h), or raising speed limits by as much as 15 mi/h (24 km/h) had little effect on motorist’ speed.”

    No mention of safety.

  • LinuxGuy

    You will argue if I say the sky is blue, so believe what you want, it will not make you any more correct. How many crashes involving bicyclists or pedestrians are caused by those people? This has also been well-documented. They think that since they cannot be identified, that they need not obey the rules and also do not pay attention. I am glad to see stings for that unsafe behavior.

  • Old Town Outsider

    Now you’re changing the subject and making false statements. Or are they alternative facts?

    Admit that speed kills and that enforcement of lower speed limits would save lives. Then we can discuss whatever you want. Otherwise we are done. Which will it be LinuxGuy?

  • Why would anyone who isn’t extraordinarily intellectually dishonest recommend a piece that is so blatantly intellectually dishonest?

  • “The NMA is pro-everyone”

    Lie.

    “setting realistic 85th percentile speed limits increases safety”

    Lie.

    Waste o”If anti-car people were pro-safety, then why do they support policies
    which increase crash rates, cite safe drivers, cite the wrong drivers,
    and take away constitutional rights?”

    Lie, lie, lie, lie.

    You’re a bad person.

  • “speed causes somewhere below 5% of crashes, and I have seen it as low as 0.7%”

    Lie. lie.

    “so what is the true goal here?”

    The true goal is for car fundamentalists to obscure the truth.

    “Here is a federal study that disproves everything you said.”

    Lie.

    You are a bad person.

  • “You will argue if I say the sky is blue”

    Lie.

    ” believe what you want, it will not make you any more correct”

    Projection.

  • LinuxGuy

    Since all you can do is insult people, you lost the argument. You cannot debate people with facts, because you have only slanted items to support your beliefs. If you reject numerous unbiased studies, then there is no point in wasting my time further, so after tonight, there will be no further replies.

  • LinuxGuy

    Again more insults, and a rejection of facts.

  • LinuxGuy

    Still more insults, you really make great arguments! Not! A typical liberal, offering only insults.

  • LinuxGuy

    I am stating facts about driving and how you have acted on this topic. The truth is hard for some to believe. I worked at an ENR 500 engineering firm and have done traffic studies, now have you done anything like this? I doubt it. I will not admit the false ideology you hold, because it was disproven many times over. I am glad we are done, because this is a waste of time, if you do not like facts. Speed kills was debunked decades ago, and further debunked when states set their own limits. Just recently, Australia showed that strict speed enforcement causes crashes. Are the Aussies wrong as well?

  • Lie.

  • LinuxGuy has no argument at all. He’s a fundamentally bad person, and whining about the observation doesn’t change that. And he’s too dumb to realize how completely he shows his hand with this hypocritical nonsense about liberals.

  • LinuxGuy

    I thought you were not replying anymore? You had to write back to hurl more insults though, as that is all you have.

  • LinuxGuy

    Still writing back when you said you would not?