3 Graphs That Explain Why 20 MPH Should Be the Limit on City Streets

Graph: ProPublica
A still from ProPublica‘s interactive graph.

Speed kills, especially on city streets teeming with pedestrians and cyclists.

The investigative news nonprofit ProPublica has produced an interactive graph that deftly conveys how just a few miles per hour can spell the difference between life and death when a person is struck by a motorist. ProPublica’s Lena Groeger used data from the AAA Safety Foundation to chart the plummeting likelihood of survival as motorist speed increases.

The average pedestrian struck by a driver traveling at 20 mph has a 93 percent chance of surviving. For a 70-year-old, the chances are somewhat lower but still a robust 87 percent.

As Groeger puts it:

Once cars reach a certain speed (just above 20 mph), they rapidly become more deadly. According to [AAA’s Brian] Tefft’s data, a person is about 70 percent more likely to be killed if they’re struck by a vehicle traveling at 30 mph versus 25 mph.

In collisions at 30 miles per hour, about one in five pedestrians will not survive. For older pedestrians, the odds are significantly worse:

Graph: ProPublica
Graph: ProPublica

The risk of death continues to rise dramatically as speeds exceed 30 mph. At 40 mph, most older pedestrians will not survive:

Graph: ProPublica
Graph: ProPublica

In addition to much higher survival rates in the event of a collision, driving at slower speeds reduces stopping distances, making it easier for drivers to avoid hitting people in the first place.

These graphs are a compelling visualization of the data that undergirds campaigns like “20’s Plenty” in the UK, where residents have demanded 20 mph streets to protect the most vulnerable. As of last fall, more than 14 million people in the UK lived on streets with 20 mph speed limits, according to 20’s Plenty founder Rod King, and the campaign is now focused on extending the 20 mph limit to most streets in the country:

22 thoughts on 3 Graphs That Explain Why 20 MPH Should Be the Limit on City Streets

  1. If twenty is plenty, then let’s push to redesign our streets so that drivers will drive at that speed, without the need for enforcement.

    Lower speed limits do not, in themselves, make anyone safer. We don’t have and can’t afford enough police officers to enforce speed limits on every street where people feel they can drive above 20 mph safely, and automated enforcement is such a battle everywhere it’s tried that it’s simply irresponsible to prioritize lowering the limit before the cultural change that needs to happen to make them feasible. Put another way, by the time we have the political capital and community support we need to put cameras on most streets, in order to enforce a 20 mph limit, we’ll have the political capital and community support we need to redesign our streets to make them safe for all users – except that we won’t, in that case, actually have well-designed streets, with all the side benefits they provide.

    It’s worth noting that the “case” for 20 vs. 25 mph limits seems to be exactly the same as the “case” for 25 vs. 30 mph limits – you’re about 70% more likely to die if you’re hit by a car traveling at 25 mph than you are if it’s going 20 mph. (And 75% more likely when it’s 15 vs. 20; 50% more likely when it’s 10 mph vs. 15.) If this is our sole basis for making the argument, where do we draw the line? The fatality rate only increases by 55% between 30 and 35 mph; 45% between 35 and 40 mph; 33% between 40 and 45 mph. If we’re trying to set the limit before fatality rates significantly increase, we’d have to set the limit unrealistically low; if we’re trying to set the limit where fatality rates even out, we’d have to set the limit absurdly high.

    We should be focusing less attention on how to protect pedestrians and cyclists when they’re hit by drivers – because that inevitably puts us in the position of asking how many people we should tolerate dying in order to allow drivers to continue driving at a reasonable speed – and instead on what we can and should be doing to prevent pedestrians and cyclists from being hit by drivers in the first place. These speed limit fights are a pointless distraction that take us further from our goal of zero-fatality streets. We need to step away from relatively “easy” wins like unenforced speed limit changes, which serve only the interests of politicians and advocacy organizations, and towards the harder work of shifting the cultural thinking on what a street is and whom a street serves, which serves the interests of users.

  2. What we could do is make the streets out of cobblestone, and then drivers will not go much above 20 mph regardless of the posted speed limit. Bike lanes need to be smooth pavement however. Wouldn’t it be a great change if the cars had to ride on bumpy roads while bikes got the better pavement, instead of the way it is now where cars ride on a smooth speedway while bike riders have to suffer with the worst chewed-up parts of the street? Cobblestone streets are charming also, and raise property values.

  3. I disagree, having studied this very issue quite a lot some years back.
    I’m not saying there’s one sequence of changes that every jurisdiction must follow in the same order, nor am I saying that your suggestions for improving safety are no good. But despite all the uphill fights that follow, and despite the hugely deficient enforcement in the interim, I arrived at the conclusion that setting a speed limit before streets are rebuilt to match the new limit is not at all a waste of time, and was indeed in NYC the correct *first step.* I have more to say but I’m sneaking this comment in at my new job and have work to do. One nutshell: virtually never until very recently has any speed limit in the U.S. been set because of the fragility of the human body. That’s always been stupid and wrong. Officially sanctioning speed limits that are dangerous to residents is wrong. “Realistic” or not, it’s wrong, and fixing that is a fine *first step* in making a place safer.

  4. My personal take on this is traffic volume is a much bigger driver of fatalities than driving speed. More vehicles equals a greater chance of being hit by one. It also results in road rage, which in turn causes much of the careless driving resulting in fatalities. For a whole host of reasons besides safety, radically reducing traffic volumes in NYC to maybe 10% or 20% of what they are at present will move us closer to Vision Zero than a reduction in speed to 20 mph.

    I also think while the major goal as you say should be preventing pedestrians or cyclists from being hit by drivers in the first place, we should focus on making vehicles more forgiving to vulnerable users in collisions. There’s no reason we couldn’t get the fatality rate at 40 mph to equal the present rate at 20 mph if we focused as much on protecting people outside the vehicle as we do protecting those inside it. Some vehicle designs, particularly SUVs, are just borderline sociopathic. They’re not only dangerous to vulnerable users, but they’re dangerous to anyone driving something more sensible. We shouldn’t have an arms race on our streets.

  5. The question is whether setting the limits lower actually makes anyone safer. Nothing in your comment addresses this core point.

  6. A few points;

    Cobblestones are expensive, and bad for maintenance. Cars going over them are also a lot louder, so it sucks for the residents.

    As for bike lanes being bumpy vs car roads, that’s not really true. Thoug there are some notoriusly bad lanes, have you ridden in the 5 boro bike tour, where you can actually cycle on NYC highways? The road quality is terrible! I was actually surprised how bad it was, given that I drive those roads all the time, and I thought they were okay. I saw plenty of cyclists blow out a tire on the potholes and cracks. Cars just happen to have pretty good suspensions and wide tires, so you don’t feel it as much when you drive.

    As for cobblestones increasing property values… eh, you know that’s not how that works, right?

  7. Okay I’ve edited the comment because in its originally hastily written form it was imprecise in that it made it sound like I disagree with everything you’ve said. Importantly, for instance, I strongly agree with the conclusion of your first sentence: “let’s push to redesign our streets so that drivers will drive at that speed, without the need for enforcement.” Actually I say that a lot and advocates I’ve known for a long time disagree with me because of my attitude that “enforcement” is nothing more than a stopgap and even now is not a very useful focus for safe/livable streets advocates.

    The part I disagree with is your assertion that “speed limit fights are a pointless distraction that take us further from our goal of zero-fatality streets.” I strongly disagree with that. These fights are a public discourse that has educated a large segment of the public–including, importantly, government officials (including elected representatives) who previously gave the subject little or no thought. The discourse has motivated opponents too, but net the “yes do reduce the limits” side has been winning in the U.S.

    Back to your other assertion, that the speed limits themselves do nothing. Purely for purposes of this debate I’ll question whether that really is, as you say in your last comment, “the question” posed by your first comment. I don’t know; that first comment was substantial and made a few points, I don’t know if I agree that that was really its primary point… But, again, my first response was hastily written.

    So I don’t disagree with it, in that it’s purely theoretical, and as a piece of pure theory, it’s not wrong. But I don’t think it’s a helpful thing to spread around if you’re interested in improving safety for everyone, because as a center of a push for safer streets, it provides resistance to that push. That’s because pubic policy vis a vis redesigning streets to make drivers stick to 20mph doesn’t happen in a theory vacuum, and speed limits are a bit of language that the public understands, and that public, and that bit of language, are going to be part of any real changes. So again, it’s not wrong, and it wasn’t the part I was disagreeing with in the first place but please don’t let it be the heart of the case you present to the world for improving safety.

    To repeat my main point, my main disagreement with you is your assertion that public debates over speed limits take us further away from our goal.

  8. Worse is how people will tell you someone’s weight in stone, which is 14 pounds. That’s annoying to calculate in your head. Like “he’s X stone!”. umm, is that super skinny or super fat.

  9. That’s not ironic at all. They’re an AUTOMOBILE association. Why would they try to limit its use?

  10. No informational value in this comment of mine since CtotheC is correct, but maybe some decorum value in that I’ll demonstrate how one can correct someone without being mean:
    The UK does use metric for most things, but road speeds and vehicle speedometers use mph.

  11. I appreciate that pounding on the speed limit “educates” the public, including public officials, but I’m skeptical that it teaches them the right things.

    A lot of the resistance we’re facing right now, in NYC, when it comes to making our streets safer and more livable, is based in the impression that our safety-improving proposals are coming at the expense of drivers. We want to limit driving speed to make streets safer; we want to ticket drivers to protect children; we want to remove parking spots to provide protected bike lanes; we want to remove general traffic lanes to provide designated bus lanes; and so on. Every improvement calls for a concession, and every concession invites a fight. As such, we have limited political capital and community tolerance for these changes, and we often end up having to ask for less – e.g., speed cameras that trigger only when you speed more than 10 miles above the posted limit, Class 2 bike lanes, etc. – in order to deflect community objections. And even that usually isn’t enough to avoid the fight.

    Fights over the speed limit feed and perpetuate that way of thinking, and politicians correspondingly conduct their political calculus within that context. So when BdB came out in favor of the 25 mph limit in NYC, and Central Park cut speeds to 20, these were viewed as reasonably “moderate” ways to play to the majority of New Yorkers who don’t drive or who live near Central Park, respectively, while only mildly irritating the driving class (since the lower limits wouldn’t be seriously enforced for most drivers, and the limits in CP are primarily more about cyclist harassment). Meanwhile, BdB’s DOT’s designs have gotten worse, in terms of road safety (again, as part of a bow to the driving class).

    We need to break this dynamic. We need to reframe the fight for road safety and livable streets as something that serves everyone, not just scofflaw cyclists and phone-absorbed pedestrians. We do this, I believe, by strongly making the case that free parking exacerbates parking shortages and the lack of congestion pricing makes driving in the CBD and parts of the outer boroughs worse; by showing how dynamic, livable streets sustain small businesses and improve neighborhood safety; by illustrating how reasonable bike and bus infrastructure actually is, both for traffic flow and improving mobility generally; and so on. We should not be building bike lanes because we want people to be driving less, as BdB recently put it. We should be building bike lanes because traffic flows more safely and efficiently when cyclists have a designated place to be, and especially in the transit-underserved areas of the outer boroughs, they can get lots of cars off the street being used for 1-3 mile trips.

    Unfortunately, the reason our advocates are failing to educate politicians in this way is that it’s not conducive to their organizations’ long-term health. It’s becoming increasingly clear that organizations like Transportation Alternatives feel they need concrete “wins” and clear-cut “campaigns” with black-and-white moral contours in order to demonstrate their efficacy and attract political and monetary support. So instead of breaking the poisonous dynamic I’ve described, they pursue narrow, incremental goals that perpetuate that dynamic, while making the argument (to their supporters) that this incremental approach is the best way to shift the culture. But I believe this is a mistake; the shift of culture they’re working towards is not one based on a consensus of both the driving and non-driving classes as to what is best. They are working towards a political consensus that highlights drivers’ interests as distinctly salient and explicitly subordinates them to other non-driver interests.

    But the only way this can come about is if they successfully dislodge drivers’ interests from the central position they currently have in the attention of the city’s political class, which is not something I foresee ever happening, based on who tends to run for and win public office. BdB is himself a perfect example of someone who still strongly identifies as a driver and with drivers; we are fortunate he has not harmed active transportation and livable streets more than he has. Unfortunately, most of his likely challengers in the upcoming election seem committed to rolling these movements back even further – I can’t think of a single likely contender who wouldn’t be a big disaster for our efforts on this front.

  12. Since when has the point of driving a car been to not kill people? I thought the purpose of a car was to convey you from one point to another.

    If the primary goal is to lower the total number of deaths by car, then let’s insist that people not drive. Period.

    If that strikes you as being stupid, then I submit to you that you don’t know what you’re talking about when you try to use this data as a reason to implement 20mph speed limits.

  13. It was even worse when they still had pounds= 16 shillings, then there were bobs, whatever. I would just hold out my change and let them count.

  14. I feel like instead of there being more of a chance a 70 year old would die from being ran over, it should be more likely a 70 year old would run someone over, when I was pedestrian mainly (a whole decade basically) I was more likely to be ran over by an elderly person than any other demographic. They just did not care pedestrians existed. Now that I’m a driver, I see more near collisions caused by elderly than any other demographic. Of course its the area I live in, the burbs, but imo testing the elderly and revoking licenses would probably go a lot farther in reducing roadway collision than mph on regular streets. The elderly never do the speed limit anyway, always 10 below, and not going with the flow of traffic causes accidents more than just about anything else. More important than speed is probably an in depth study into how age affects driving. Where I live its almost an assurity that anyone over 65 drives like complete ass.I don’t believe we really retest these people with any scrutiny.

Leave a Reply

Your email address will not be published. Required fields are marked *


The Problem With Speed Cameras That Don’t Catch Most Speeders

Residents of urban neighborhoods across the country are increasingly advocating for lower speed limits and automated traffic enforcement. As the above graphic illustrates, the stakes are high for pedestrians. But in some jurisdictions where speed cameras are in use, motorists can drive much faster than the speed limit without penalty. In New York, for instance, […]