Science Friday - The Lack Of Science In Road Design Is Deadly

Episode Date: April 21, 2025

Are traffic engineering decisions based on evidence-based research? Not as much as you might think.If you’ve seen a car crash on the side of the road, you might look at it and think that the person at fault is the driver. But how much blame should be shared by the people who designed those roads in the first place?Well, some traffic engineers are calling for the field to accept more blame for the crashes and the tens of thousands of annual fatalities that happen on our roads, including Dr. Wes Marshall, a professor of civil engineering at the University of Colorado Denver. In his book Killed By A Traffic Engineer: Shattering the Delusion that Science Underlies our Transportation System, he digs into the standards that have dictated traffic design for decades to find out exactly how much science they’re based on. Spoiler alert: It’s a lot less than you’d think.Host Flora Lichtman sits down with Dr. Marshall to talk about how we got to this point and what a safer version of our streets could look like.Transcript for this segment will be available after the show airs on sciencefriday.com. Subscribe to this podcast. Plus, to stay updated on all things science, sign up for Science Friday's newsletters.

This is Science Friday. I'm Flora Lickman. Today in the podcast, are we in the dark ages of traffic engineering? Where people are actually dying or having severe injuries, those tend to be the roads that have most of our engineering science put into them. When you see a car crash on the road, you might wonder which driver was at fault. But what if that's not the right way to think about it? What if that crash could have been avoided entirely if the roads were designed better, designed with more updated science and data in mind? That's the thesis of my next guest, Dr. West Marshall, a professor of civil engineering at the University of Colorado, Denver. He's a traffic engineer who wrote a book called Killed by a Traffic Engineer, which digs into the standards that have dictated traffic design for decades to find out exactly how much science they're based on. Spoiler alert, it's a lot less than you'd think or probably want. Dr. Marshall, welcome to Science Friday.

Hi, Flora. Thanks for having me. Okay, you go really hard in this book. Like, you're quite critical of your own field. The driving gloves come off. Why did you feel compelled to write it? Anger, maybe a little bit. No, I mean, really, I've been doing research, I mean, for decades. And I feel like a lot of that stuff is good, but it's sort of chipping away at the tip of the iceberg. So what I was learning, I mean, the deep right dug is that foundation was really where the problems were. So it needed to be something. bigger. And I'm also not really one to pull punches on this kind of stuff. So, I mean, maybe that's a little bit of my personality, but that's where I ended up going. Okay. So, I mean, you're a researcher, but you've also worked as a transportation engineer. Is there a major misconception that you're hoping to clear up for people about road design? Well, I mean, it was a misconception, like for me, too. As a young engineer, I mean, you take a couple of transportation engineering classes, you get into the workforce,

and they drop, you know, a half dozen of these thousand-page manuals on your desk. And you assume that whoever wrote these, like, knows more than you did. You assume that they did their homework. You assume that there's 100 years of science. And more than that, safety science steeped into these documents. And, you know, what I'm trying to show is it's not quite so. So I'm going down these different rabbit holes to try to find the origin story. I'm not just pointing out what we do was wrong and where, but why we do what we do.

And oftentimes it wasn't nearly. what I was led to believe. It wasn't nearly as scientific as I was taught. So, you know, for a lot of people in the general public, it's great information and understanding like how these things come together. But even for my profession, I was helping us, you know, a little bit of self-reflection and understand a little bit more about why we're doing what we're doing. Well, give me an example of that. Is there an accepted standard and road design that isn't based in science or that's actually harmful? Oh, there's so many. I mean, That's a fun question. Down a lot of these rabbit holes were just really interesting stories and stories that were so different.

I mean, some of the basics are like, we think that we were, I mean, I was taught that wider roads are safer. And if you look at the original studies on that, that's exactly what they found. There's a study from mid-1930s, 1940s, and for decades, we cited those studies. And then we stopped citing them and just the idea that wider roads are safer perpetuated through time. But if you go back to those original studies, like the widest road they looked at in the first study was only 22 feet wide curve to curve, which is smaller than any road we would put out there today. So yes, a 22-foot road was safer than a 20 or 18-foot road. But it doesn't extrapolate to the kind of things we're building now that are 60, 80, 100 plus feet wide. Well, also, I'm like Googling right now, what did cars look like in the 1930s?

I mean, they're quite different. Like, I imagine not just the roads are different, but the cars are different. The volume of cars must be different. Like, is that a relevant study to even invoke today? Probably not. And that's the thing. A lot of our stuff is based in theory. And so much has shifted over the years.

Like, we think so much will help safety. Like, I mean, right now, there's a lot of technology being added in the cars. And you think it will sort of improve safety. I mean, the data would make you think that some of these things, technologies would almost eliminate some crashes. But then when you actually see them getting into the world, you know, it eliminates maybe a fraction of the crashes that we think it will. Well, like specifically like the crash avoidance technology, like the beeps when I'm about to hit something? Yeah, I mean, things like that. I mean, the research on some of that stuff

shows that it works well during the day and it works best if like the pedestrian in front of you or something is light skin. It doesn't work as well at night, at dusk, at dawn, or the darker the skin on the pedestrian. So there's a lot of stuff that, yes, it would be safer, especially if we do the exact same thing we always done. So like when you're changing lanes in the highway, you know, before you would use your mirrors and look over your shoulder and use the technology in addition to it, but when you start relying on just the technology, you might not get the benefits you think. What is the ultimate goal for traffic engineers? I mean, is it to eliminate traffic? Is it to be as efficient as possible? Is it safety? I mean, well,

We always say that safety is our first priority, but I mean, honestly, it's not true. Our priorities are usually around reducing congestion or increasing capacity or increasing speed. And if you look at the way our protocols are set up, like that's what they're leading to. It's like things along those lines or even sometimes reducing costs. Like safety at best is third. So that is part of the thing that we don't like to admit, but that's the way the system has been set up. So if you look around most cities, like where people are actually dying or having severe injuries, those tend to be the roads that have most of our engineering science put into them.

Really? Yeah, it's not the rows that were built before traffic engineers existed. Like those ones tend to be safer today than a lot of the newer roads. Why is that? Oh, that's sort of the problem. Like, we think a lot of the stuff we're doing is steeped in the safety science. And it's not. It's really more steeped in trying to increase capacity, not even just capacity.

today, like capacity 20 years into the future at some distant off peak hour. And it's not steeped in trying to help the safety of the people that live and work and play in those areas today. Are we collecting the data we need to actually make road design better? Oh, that's a good question, too. I mean, to some extent, yes, but not nearly enough. I mean, one of the problems I see, and it's sort of fundamental. Like, we all want to have a more data-driven approach to making roads safer. And when any... any city, any researcher like me, when you dig into the data, all the data is telling is that we have this huge human error problem. It will be people speeding and people jaywalking.

And this goes back to your original question too. When you start looking at those actual crashes and asking why, like why were those road users doing what they're doing? Oftentimes you can come to a different conclusion. Like, yes, for the sake of liability and insurance and police, it's useful to sort of blame the bad actors in the system. but for engineers and transportation planners and folks like us, if you start asking why, like, why did this person jaywalk? And you start looking at the situation we put them in and you realize, well, the nearest crosswalk is like a half mile away.

And even if they go to that crosswalk, it's not that safe of a crosswalk. And you look at the built environment between where they are and where their crosswalk is. It might be missing sidewalks or things like that. And you start to think, well, what they did in that situation was for them the most rational decision. Like so from my perspective, it's us who's failed to provide a safe place to cross. And we put them in a situation where jaywalking was their best option. And then we blame them when they do so and they get killed. And it happens for almost any crash you can kind of think of. You can start asking why. And you start getting a different perspective. We start kind of going

down that rabbit hole of the why these people did what they did. I mean, you've called it the dark ages of traffic engineering. Well, I think someday it might be thought of that way. I mean, I think part of the issue, too, is that we as traffic engineers, and I thought the same thing. I thought it was all steeped in all the science. And as you start looking at it, you're like, wait, what? Like, that's why we do this. And almost every time you get to something along those lines that it's like, oh, my gosh, like, that's the only reason. Like, that's the science behind what we're doing.

Like, we need to do better. And, you know, for a lot of this stuff, I'm not saying that wider roads are necessarily safer and safe. I'm just saying the science we're using to tell us they are safer. isn't quite what we think. And we need to have more research and more basically empirical results to really tell us, because it might change in different contexts, too. Like, it's not as simple as we often make it same. We have to take a quick break, but don't go away. More on this when we come back. If we can admit that these things are steeped in the science that we think, that is really the first step to try and move in the right direction. You know, we put out a call for Lister questions. And of course,

we got tons. Here is one that a lot of people were wondering about. Hi, this is Joey. I have a question about traffic calming circles. So my husband's usually very even tempered, but traffic calming circles make him growl. What problem are traffic calming circles trying to solve? And is there good evidence they actually work?

Thanks. That is a great question, Joey. So, I mean, the fundamental thing they're trying to solve is a physics problem. It's speed. And our bodies can only take so much force. So when there is a crash, like the best thing you can do is, like, reduce the speed. Like force equals mass times acceleration and reducing speed and things related to traffic calming.

So not just traffic circles. Are we talking about roundabouts? Sort of. I mean, roundabouts are a little bit of a distinction. These are oftentimes neighborhood traffic circles. It's a little bit of a similar to a roundabout smaller oftentimes. it's more just trying to help drivers decide to go slower through different neighborhoods. It's almost like a self-enforcing type design, which oftentimes these roads, they're over-designed.

They're designed with a design speed or factor safety to sort of entice you to go faster than we want you. And then we put up a speed limit trying to ask you to go slower. And that often is where we get the disconnect and people speeding through these neighborhoods. So things like traffic circles can help. And really the best thing, a lot of those round type intersections help is it reduces like the T-bone type crash, which is one of the most dangerous ones. Like if you have a crash in most roundabouts or traffic circles, it tends to be slower and it tends to be more of a fender bender. And like for me, I'm much more focused on trying to save lives and keep people from being severely injured than trying to reduce the number of fender vendors. So if we have a few more of those, it seems like a worthwhile tradeoff.

And it especially helps in making these streets feel better for people that want to walk or bike. And it still allows people to drive too. So it kind of gives people more options and more freedom to do what they want to do. You know, you're talking about saving lives. How big of a problem is this? I mean, how many people die from road deaths every year? Well, we treat it as the cost of doing business, but it's not. It's huge.

And in this country, United States, we're looking at over 40,000 deaths every year around the world. It's like 1.35 million each year. I did some math recently, and I tried to figure out the total number of deaths we had in the United States, you know, since we started collecting this date in 1890. And around this time last year, we crossed the four million total deaths mark. By the end of this decade, a million of those will be pedestrians and cyclists. And within that, there's hundreds of thousands of kids that have died. So you look on the news every night and you can always see a story of somebody dying, but we don't sort of treat it the same way that we do. like on a plane crash or a train crashes, like those are more like a fire hose that we have to shut off.

These fatalities are treated more like a faucet dripping. And we all sort of just live with it and think it is what it is, but it doesn't have to be. How do you design roads to make them pedestrian-friendly? Are there specific things you can do other than adding crosswalks that make a road good for pedestrians? Well, there's a lot of things you can do. I mean, I will say that where stars is usually sidewalks. I mean, I've been in a lot of cities, like even near where I am, currently in downtown Denver, I do not have to go far to find the sidewalks just will disappear, they're missing. Oftentimes, like, I have pictures of places where we built these amazing ADA curb ramps and there's a bus stop, but there's no sidewalk between them. So when you look at these things, just like, it's funny that we're so focused on technology saving the day when we can't even get the sidewalks right in so many places. But there's a lot of things we can do. It's hardly like

pinpoint one, but, you know, typically it's making your designs prioritize those folks to some extent, because for decades, they've been an afterthought. Here's another question that a lot of people called us about. Commute from Northern Rhode Island into Cambridge, Massachusetts, and I'm always blown away by the amount of traffic I see. What really bothers me is that I just don't understand why the slowdowns happen, and I'm expecting to see some large accident or some major incident way above that's creating all the clogging, but I don't see anything. I don't understand it. I don't get it.

I don't like it. Thanks. Any thoughts? Yeah, I mean, that's a good question, Larry. And I will say that usually it's oftentimes on the highway, but it's the pinch points when you get on and off the highway. So the on-off ramps when you get into cities like Cambridge and like that sort of stuff happens. But it's also, you know, I live in. Denver and I grew up in that area in Boston. And, you know, as someone that didn't have to commute from

Rhode Island to Cambridge, you know, I can get around without having to even be in a car. I can bike. I can take the train. But I could drive too. And getting on the highways oftentimes limits your options. So in Denver, when I hit traffic in a car, it's a grid so I can move over a street or two. When they're on the highway, you can't do so. You're sort of stuck, sort of waiting for, in the kind of situation that I was talking about to dissipate. And oftentimes, like, Cambridge isn't this, but when I was commuting out to, like, Chelmsford, Massachusetts, and the highway was fine, but trying to get from the end of the highway to my final destination was so difficult because there was only, like, one option for a road. So redundancy and grids can help so much. And you can

get much more traffic through a grid as opposed to one giant multi-lane road. Hmm. Okay. Are there specific? changes supported by science that should be made? Like, do we know what we need to do or do we need to do more science? Both. I mean, to some extent, we know what we need to do. We know that what we put out there isn't the safest designs we could have and we can do better. At the same time, we need a lot more research to try to give us more examples. And I think the best thing you can do is look for the empirical results for where things are working, like where the streets that people love, the streets that people find charming and lovely and safe and useful or even efficient,

whatever it is, like measure those streets. Let's try to learn from those empirical examples, as opposed to just letting our theories guide the way. So yes, we need more research, but there is enough out there where we need to start embedding that research back into those manuals because that's one of the big problems I found is those manuals weren't steeped in like the best research. They're steeped in our old theories. Do you have examples of great streets? Like roads that are getting it right. You know, it's hard because it's so context sensitive. If I point out a great street in one city, another city is like, oh, we can't possibly do that because we're not so-and-so.

So, you know, I try to, you know, every city and, you know, when I grew up in the Boston area, people could point the streets that they love. And if you look at the way our mandals are set up, it would be impossible to redesign those streets today because they don't fit our typical. criteria or theories about what makes for good, safe designs. And that seems to be the case in most cities you go to. They have these older neighborhoods that the streets are almost the legal to build. And it's often not times due to the mandibles, sometimes due to municipal regulations, both things like level service. But those are the places people love. And we put ourselves in a corner where we can't redesign them. So I would say instead of looking for some pie in the sky place from some mother city in a different context with a different culture, look around your own neighborhood,

city, region for places you love and see if you can replicate those. How do we get out of the dark ages? What's our first step? Admit that we're in it is probably the first step. I mean, I think that's... That's always the first step. Yeah. There's been some denial. I mean, there's been a few engineers that take offense with the book title, but if they actually read it, they'll see I'm not blaming them. I'm like, well, we're just doing what we were taught and what we were taught isn't quite so. So if we can admit that these things aren't steeped in the science that we think, that is really the first step to trying to move in the right direction. Wes, thanks so much for taking the time. Thanks for having me. This was fun.

Dr. West Marshall, professor of civil engineering at the University of Colorado, Denver. And if you want to read an excerpt from killed by a traffic engineer, head to our website at Science Friday.com slash traffic. And that is about all we have time for. Lots of folks helped make the show happen, including... Dee Petersman. Praise of Gucci. Kathleen Davis.

Santiago Flores. I'm Flora Lichtman. Thanks for listening.