Skip to main content
ABC News
Why The Rules Of The Road Aren’t Enough To Prevent People From Dying

On a foggy morning last February, Jelani Irving rode his bike in the darkness just before sunrise. He was returning to his home in Brooklyn’s Crown Heights neighborhood after finishing a shift driving a taxi, a job he’d started two weeks earlier.

According to police and family accounts, Irving crossed Eastern Parkway, a busy thoroughfare, and rode down Washington Avenue, along the eastern side of the Brooklyn Museum. A car driving under the 30 mph speed limit approached from the other direction as Irving veered left onto a cross street. But the car was turning right onto that street. Irving was in its path. The two roads meet at an obtuse angle, which allows motorists to turn without slowing down; the car and bike collided. An ambulance whisked Irving, 22, to Kings County Hospital Center, but he died four days later.

The death was ruled an accident, and there’s no evidence anybody broke the law. But does that mean we should view this tragedy as unavoidable?

How speed limits are set

In 2013, 32,719 people died in motor vehicle crashes in the United States, and 2.3 million were injured, according to the National Highway Traffic Safety Administration (NHTSA). Those numbers were down from the previous year, but motor vehicle crashes remain a leading cause of death, and speed is a leading cause of accidents. The NHTSA estimates a $277 billion annual price tag1 for those accidents, with an additional $594 billion for “harm from the loss of life and the pain and decreased quality of life due to injuries.”

Given the social and economic toll of speeding, one might assume that we set speed limits with careful calculations aimed at maximizing safety. But that’s not exactly how it works, and a history of questionable applications of data is partly to blame.

Here’s how speed limits are established in most states, according to Federal Highway Administration research: Traffic engineers conduct a study to measure the average speed motor vehicles move along a road. The speed limit is then set at the 85th percentile. From then on, 85 percent of drivers would be traveling under the speed limit and 15 percent would be breaking the law. Sometimes other factors2 are taken into consideration, but in most places, speed limits are largely determined by the speed most people feel safe traveling. Some states, including Louisiana and Michigan, go so far as to call limits determined by this method “rational speed limits,” stating that achieving compliance is possible only if the speed limits are reasonable.

Drivers travel at the speed a road allows, and speed limits are set accordingly. So, what determines how fast people can drive and still feel safe?

Roads are planned according to a concept known as design speed, basically the speed vehicles are expected to travel.3 Engineers often apply the 85th percentile rule to a similar road to arrive at the design speed for the proposed road. It might make sense, then, that the design speed would become the speed limit. However, in practice, the design speed is often used to determine the minimum speed of safe travel on a road.

Confused? So was I. Norman Garrick, a professor of engineering at the University of Connecticut, explained how this works using the example of a commercial office building.

“The engineer will usually calculate the load a beam must bear and then design it to hold some percentage of higher load, for safety. When building roads, the 85th percentile calculates the speed the engineers hope or intend people will travel, but then it’s used to design a road to meet that speed at a minimum, with a factor of safety allowing for faster travel,” he told me.

In other words, by adding additional “safety” to the road, it is designed to make people comfortable going faster than the engineers’ intended speed. This is known as the interpreted design speed (the speed people actually feel safe traveling), which is often significantly higher than the intended design speed. Think of a subdivision with wide, flat roads. The speed limit may be 25 mph, but you feel utterly comfortable doing 40.

The 85th percentile idea grew out of research that is now a half-century old. Several people I spoke with traced it to a 1964 study prepared for the U.S. Department of Commerce.4 The main finding of the study was that when a car travels at a vastly different speed from the average speed of all traffic (whether faster or slower), there is an increased risk that the driver will be involved in an accident.

There are lots of problems with the study and its applications. It was conducted 50 years ago, when the size and distribution of cars and trucks on the road were very different from today.5 The study included crashes only on two- and four-lane rural highways, where the risk of head-on collisions is much higher than it is on divided interstates. Many of the “speed variations” were related to slowing down in congestion or at intersections, situations that are unavoidable. But perhaps most notably, the study did not include anything related to pedestrians or cyclists, who share the road space with cars in most urban and suburban environments. Without a 2,000-pound vehicle for protection, they are also more likely to experience serious injury, even at slow speeds.

Death on the highways

In a study of how impact speed affects a pedestrian’s risk of severe injury or death, Brian C. Tefft of the AAA Foundation for Traffic Safety found that above 15 mph, relatively small changes in vehicle speed lead to large increases in risk of death. Tefft calculated that the risk of death for a pedestrian is 10 percent with an impact speed of 23 mph but rises to 25 percent at 32 mph. The limit is 30 or 35 mph in many cities, so 32 is not an uncommon speed.

But the majority of motor-vehicle-related deaths involve vehicle passengers. There’s evidence that speed limits have an effect on that number as well. From 1974 to 1995, a federal law called the National Maximum Speed Law restricted speeds to 55 mph on all interstate highways (though an act in 1987 raised the limit to 65), with the intention of reducing fuel consumption after a 1973 oil embargo. When that law was removed in 1995, every state raised its speed limits. A study published in 2009 in the American Journal of Public Health researched the effects of the changes and found a 3.2 percent increase in road fatalities attributable to the raised speed limits. The researchers calculated that the speed increases led to 12,545 deaths over the 10-year period they studied.

The birth of jaywalking

It turns out, we weren’t always obsessed with speed in the United States. When cars were first becoming popular at the beginning of the 20th century, terms like “road hog” and “joy rider” were called out at “speeding” vehicles, which in urban areas tended to mean they were going faster than 10 mph. In his book “Fighting Traffic: The Dawn of the Motor Age in the American City,” Peter Norton, a professor of history at the University of Virginia, argues that in the early 20th century, people thought we needed to keep roads safe from cars, not for cars. City speed limits were set below 20 mph until automakers and industry groups realized that such low limits were going to hurt sales. To change public opinion, campaigns were started to criminalize pedestrians in the road (“jaywalking”) and to shift the blame for accidents from cars to “reckless drivers.” By 1930, city speed limits were just about what they are today.

There’s evidence from Europe that high speeds in the United States over the past century have caused many deaths. The Netherlands and Sweden have overhauled the design of their roads and cities, resulting in enormous declines in motor vehicle fatalities. In urban areas, curbs are removed, giving the perception of shared space (though cyclists, pedestrians and cars are still separated), which encourages drivers to slow down. And in most of Europe, driving lanes are much narrower, which also fosters slower travel.6 In Sweden, a program called Vision Zero treats all traffic fatalities as preventable, and this idea has recently crossed the ocean to U.S. cities.

Soon after New York Mayor Bill de Blasio took office in 2014, he announced an initiative to end motor vehicle deaths, using Vision Zero as the guide. The first order of business was to lower the default speed limit.

First, state law had to be changed. Since 1964, the law had required default speed limits to be at least 30 mph throughout the state. By November of last year, that was lowered to 25 mph for New York City. The speed limit reduction was the most public of Vision Zero’s initial changes, but to date, 29 of its 63 initiatives have been implemented. Preliminary data suggests the program may be paying off: 2014 saw the lowest number of pedestrian deaths since the city started keeping records in 1910. New York also had a decline in overall motor vehicle deaths from previous years. But with 134 pedestrian deaths and 250 overall, the city is a long way from zero.

Data released Tuesday by the New York City Department of Transportation also showed that controversial speed cameras near schools may be working to slow traffic at the 19 sites where they have been placed since September7. Overall, there was a decline of 58.7 percent in the number of daily speeders found on these cameras (which ticket only at speeds of 10 mph above the posted speed limit), with individual cameras’ declines ranging from 21 to 75 percent. This suggests the mere presence of the camera can help reduce speeds.

Paul Steely White, executive director of Transportation Alternatives, a group that advocates for pedestrians, cyclists and public transit in New York City, noted a culture of acceptance that some people will die in accidents if we live with cars. But he thinks design, laws and education can change that.

“We can analyze the causal factors and control for them. We can mitigate those factors. We know that human beings are going to make mistakes, but we can design the system so that those mistakes won’t be fatal,” White said.

It’s impossible to know whether a different road design or a lower speed limit could have saved Jelani Irving’s life. But White argues that the goal of zero traffic fatalities is achievable.

“It’s completely unacceptable for someone to die in a plane crash or an elevator,” he said. “We should expect the same of cars.”

The city’s deputy commissioner for transportation planning and management, Ryan Russo, said that in a dense urban environment like New York, improving safety doesn’t necessarily increase travel time from A to B. And while the speed limit change was an important symbolic first step, and a meaningful one for safety, the city is only a year into the program.

“The speed limit change was important functionally for those streets,” Russo told me, “but also for starting conversations, getting media coverage and discussion, and to get Vision Zero off to a strong start.”

CORRECTION (Jan. 15, 11:02 a.m.): A previous version of this article incorrectly stated the number of traffic fatalities in New York City in 2014, and the speed at which drivers in the city receive a ticket from traffic cameras.


  1. The figure represents the direct costs of the fatalities, injuries and damaged vehicles.

  2. These can include proximity to schools, statutory speed limit, daily traffic, specifics of the terrain (mountain roads, for example) and the functional use of the road (whether driveways back out onto it, etc.).

  3. Most U.S. cities have inherited the roads of previous generations; redesigning roads for a new speed often involves adding features, painting lines or narrowing lanes, changes otherwise known as traffic-calming measures. But it’s helpful to understand how our roads were designed historically.

  4. I also traced references on dozens of research papers back to the same source.

  5. The size and weight of a vehicle changes the dynamics of a crash. Although cars weigh less than they did 50 years ago, there are more SUVs on the road.

  6. More information about the Netherlands’ “self-explaining” roads and Sweden’s Vision Zero can be found here and here.

  7. The city has 21 fixed cameras and 28 mobile cameras in areas near schools where crash data and speed data identified hot spots.

Anna Maria Barry-Jester is a senior reporter at Kaiser Health News and California Healthline, and formerly a reporter for FiveThirtyEight.