Why Helmets Aren’t the Answer to Bike Safety — In One Chart


Better street design and getting more people on bikes — not blind faith in helmets — are the keys to making cycling safer, recent research has shown.

Want a good visual to get the point across? The Toole Design Group made this for you.

Of these countries, the U.S. has the highest rate of helmet usage among cyclists — around 55 percent — but also the highest cyclist fatality rate per distance traveled. Meanwhile, in the Netherlands, where helmet use is practically nil, cycling is much, much safer.

While this is just eight data points, higher helmet use seems to be associated with higher fatality rates. Intuitively, that makes some sense. The more dangerous an activity, the more people feel inclined to take steps to protect themselves.

Despite the high rate of helmet use in the U.S., helmet campaigns have clearly failed to make cycling as safe as it should be. If anything, they’ve distracted from the much more important work of designing safer streets and reducing motor vehicle speeds in cities.

Updated at 4:37 p.m to replace Toole’s line graph with Toole’s bar chart, based on the same data.

95 thoughts on Why Helmets Aren’t the Answer to Bike Safety — In One Chart

  1. I recently bicycle toured in the the Netherlands and Germany. The cyclists there who don’t use helmets cycle at 15 mph or less, cyclists on fast road bikes wear helmets. The streets have lower speed limits, 30 kph, where traffic is faster bicyclists use a protected bikeway. I agree helmets have little to do with fatality rate, but street and road design does.

  2. I’m pretty sure the plot would look the same if bicycle *injuries* instead of fatalities were plotted.

  3. You made the classic mistake of only looking at the costs without also looking at the benefits, so your conclusion is false. Per the study you linked to, injury rates from collisions are worse on bicycling. But the benefits from getting exercise through bicycling outweigh the costs making bicycling a net health benefit:


    And let’s not even getting into the moral issues of choosing a mode of transit (driving) so cavalierly without considering the costs. The classic mistake most motorists make is only looking at the benefits of driving (convenience, comfort, ease, etc) without considering the costs (pollution, sprawl, contribution to the obesity epidemic and related heart diseases and diabetes, etc) while amazingly doing the exact opposite for bicycling. This is exactly the kind of societal bias which the helmet debate perpetuates while distracting from this greater issue.

  4. I have an anecdote which illustrates exactly what you’re describing. My late father drove everywhere other than taking the subway to work. He even drove to the local grocery store three blocks away. He had his first heart attack at age 54, which is one year older than I am now. He died from the second heart attack of age 71½. His eating habits were poor also, but it was really the lack of physical activity, and the resulting obesity, which did him in. I can’t help but think if he biked or walked for most of his errands, he might still be alive today.

    In the US there are lots of people in similar condition. Far too many people in this country have heart attacks in their 40s or 50s solely due to the lack of physical activity which driving enables. When we talk of cyclists dying because they’re not wearing helmets, which in itself is a dubious proposition, we’re ignoring the fact most people who ride regularly will have years to decades more quality time than those who don’t. The fact medical science might keep a sedentary person alive until they’re 80 doesn’t mean those were good years. A lot of these people are already virtual invalids when they’re my age. People who exercise regularly often still enjoy life well into their 80s or 90s.

  5. Helmet campaigns are a little like putting that St. Christopher medal in your wallet and expecting it alone to provide safety. The medal does little on its own unless it is a recognition on the owner’s part that safety is a core value.

    Helmet drives put the onus of safety on the end user rather than on the policy wonks and engineers who plan and design our transportation systems. You cannot design an unsafe system and cure it with a helmet.

    Now, which part of my wallet has my St. Christopher medal in it? My mom gave that to me over 40 years ago when I bought my first motorcycle….

  6. Helmet drives usually lack a context, and are cheap ways of distracting from a safety culture that drives the design of our infrastructure. Sure, I wear a helmet for a specific reason–I have a finite chance of falling down and going boom; it is a personal protective device. I don’t wear it to keep me out of a crash.

    I work in what some might consider a high hazard occupation, but am actually very safe at work. Its cycling to work that worries me. Personal protective devices are the very last layer of safety–for when all of our carefully considered primary layers of safety fail. I teach layers of safety in my LCI classes, and stress that the helmet is the innermost ring of safety and one never wants to get there….

  7. Yes better street design and getting more people on bikes make cycling safer, but also a tighter rein on speeders would result in remarkable achievements. Nowadays if speeders are going 10mph over the posted speed, the likelihood of being pulled over is pretty slim. What helps these speeders is other people are going over the posted speed too. All this ends up doing is making the roads unsafe for all!

    Start declaring that speeders will start getting cited for going 5mph or more over the posted speed and watch how streets all the sudden become safer. That’s without spending a billion or two “updating” the infrastructure or convincing everyone that that’s the way to go.

    Everyone understands they were speeding when they were cited. They might be a bit PO’d getting cited for going ~8mph over the posted speed, but they were speeding.

    And stop giving speeders a free pass for going 10mph or more over the posted speed!

  8. Regarding rings of safety and such, one should also consider the likelihood of reaching the innermost ring of safety and the downsides, if any, of providing protection against it. From my own experience and research, head injuries are exceedingly unlikely if you’re a competent cyclist, probably to the point you’re more likely to get struck by lightning. And yet we wouldn’t advocate for lightning-proof suits for cyclists precisely because the low probability of such an event would make suffering the downsides of such a suit almost pointless. Helmets have enough downsides to similarly tilt the equation in favor of not wearing them at all for a large fraction of the cycling population. The few times I tried wearing helmets lent to me by friends it just sapped all the joy out of riding, to the point I just would give it up if I had to wear a helmet. For starters I was sweating profusely when it was 40°F. The sensations and sounds of the wind going past were gone. It was like watching a cycling video instead of actually participating in cycling.

  9. Helmets were never meant to be “the answer to bike safety” the same as bullet proof vests were never meant to be the answer to world peace, but if you are going to a war zone you just might want to have one on you…

    Helmets are meant to protect you from a head injury IN THE CASE of an accident, they are not meant to prevent the accident from happening.

    Head injuries can lead to death or worse, and the laws of physics apply the same in the Netherlands as they do in the US.

    A helmet won’t stop a driver from spilling oil on the street or from texting while driving. But when you slip on this oil or this driver hits you, the helmet might determine if you end up with a few broken bones and bruises or weather you are going to loose the ability to read silly graphs on Facebook due to death or worse a debilitating head injury

  10. Another possible interpretation is an idea I’ve seen previously—drivers are more careful around cyclists not wearing helmets.

  11. ” if you are going to a war zone you just might want to have one on you…”

    Making an analogy of “war zone” with cycling is bad for cycling. It’s over the top.

    It’s not a war zone out there. It’s not as safe as it should be, but the lack of deaths in bike share shows it’s not THAT dangerous. And as we’ve seen on this blog, pedestrians die due to bad drivers too, and we don’t push helmets for them.

    Or do you? Do you call for pedestrians hit by cars to wear helmets?

  12. “Any effort spent promoting helmets by bicycle advocacy groups means less effort spent teaching proper riding techniques.”


    1. Don’t crash.

    2. Advocate for safer streets.

    Those are the top priorities.

  13. I lived in Amsterdam for three months as a junior in college in the late 60s. Bicycling was normal back THEN and everyone did it. It was part of the culture and everyone understood and accepted it. I think that long period of integration of bicycles in public awareness and culture may have something to do with the low fatality rate. You would never see a driver yelling at a bicyclist for getting in his way!

  14. Due to the prevalence of such bikeways in The Netherlands and Denmark, the figures for those two countries are a decent proxy for what you want to know.

  15. Yeah, but almost nobody in the Netherlands and Denmark wears a helmet, so it isn’t really what he wants…

  16. Yep. Did a search for “cyclist killed in Chicago,” and here’s the first mainstream coverage of a fatal crash I found (Woodstock is a Chicago suburb):

    “WOODSTOCK, Ill. (WLS) — Police are investigating the death of Christopher Michols, 18, who was struck and killed by a van while bicycling in Woodstock, Ill. Police say the accident may have been Michols’ fault.

    Michols was struck around 7:40 p.m. while crossing Rte. 14 at Kishwaukee Valley Road. Witnesses say he tried to cross the intersection against the red light. Police say he was wearing dark clothing on an unlit bike, and was not wearing a helmet.

    The driver of the van was not charged.”

  17. Thanks for boiling that down to “one fatality per 13.8 million miles.”

    What is the fatality rate for other modes — driving, walking?

  18. It’s a bit over one fatality per 100 million miles for auto travel and about one per 70 million miles walking ( http://www-nrd.nhtsa.dot.gov/Pubs/810968.pdf ).

    I much prefer fatalities per hour of exposure ( http://www.mrmoneymustache.com/2013/06/13/bicycling-the-safest-form-of-transportation/ ). That more or less equalizes the rates between driving and cycling in the US. In countries with good cycling infrastructure the hourly risk for cycling is far lower than driving. If you do a net analysis of the health benefits of cycling, you find cycling extends your life on average, while driving shortens it.

  19. I thought that was a little over the top myself but it still doesn’t change his basic message by much.

  20. Not the only thing a bit over the top. Per passenger mile cars cost ~26 cents, ~35 cents per vehicle mile.


    Not the 50 cents he gives. He tries to explain away a lot of the cyclist fatalities:

    Remember the US cycling fatality ratio of 6.9 per 100 million miles? That’s with our current group of cyclists: a disproportionate number of children under 14 with no driver training, homeless people, DUI-convicts who have lost their license, competitive road racers and downhill mountain bikers, and the less than 1% of adults who actually ride bikes to work like they should be doing. When you and I ride our bikes, we stop at the red lights and stop signs, obey the lane markings and use arm signals, use bright lights and reflective clothing at night. We plan our routes to pick the safest roads and paths. By following these steps, our own crash rate can be much lower than the national average. Probably even safer than the average for cars.

    A lot of people die in single vehicle crashes or from driving drunk or being in a car with a drunk driver. If you live in a compact enough urban area to bike, the motor vehicle fatality rate is much lower than in rural areas etc…If he waves his hand to explain away a lot of the risk from cycling because it’s avoidable/not applicable to his readership, why no comment that you can say the same about a decent chunk of driving risk?

    or every hour of exercise you do, you extend your lifespan somewhere between 3 and 9 hours.

    He acknowledges this at the bottom, but that doesn’t extend indefinitely of course. The more you exercise, the less you gain from the next hour of exercise.


    It tapers off a lot beyond 10 MET-hours/week.


    That’s only ~150 minutes of leisurely cycling/week.

    Studies show that even mild exercise like riding 2 miles a day also saves you from missing about two sick days of work per year. Assuming your days are worth about $300, you spent 60 hours riding to earn $600. An additional $10 per hour. And how do we account for those extra 2.5 hours of life you gained? Since one of my rules is that your spare time is worth more than $25 per hour, you get another $62.50 in pay for each hour you ride your bike.

    $300 a day, 250 days working a year works out to $75k. Is that typical for the ” smarter-than-average people stuck in average situations” he says he writes the blog for? Cycling gets credited with monetary value of spare time from life gained from exercise, but not penalized for the reduced spare time from being slower than driving?

    First of all, in the entire United States (Population about 310 million), there were only 623 cyclist deaths in the year 2010. For perspective, there were about 26,000 deaths due to each of “falls” and “alcohol”, and 35,000 caused by car crashes. So for every cyclist who dies on a bike, 56 die in cars. Out of the MMM readership alone (roughly 0.1% of the US population), 3 people die in car accidents every month.

    Should really throw in a bit about per hour exposure if he’s talking about per hour risk/gain from an activity in the article.

    All that said, if people could get in the habit of replacing even just a few short car trips with bicycles per week, even if it’s only 3-9 months out of the year when the weather is nice, it would be really great.

  21. Thanks, really interesting info. It looks like I’m already at the point where the benefit of exercise is starting to taper off just walking a few hours per week at ~4 to 4.5 mph for errands. Add in the cycling, which sadly has averaged only 1000 miles per year, and I’m at maybe 25 or 30 MET-hours per week. Guess there might not be much life extension benefit once I get back to my more usual 3000 miles per year but I enjoy it.

    All that said, if people could get in the habit of replacing even just a few short car trips with bicycles per week, even if it’s only 3-9 months out of the year when the weather is nice, it would be really great.

    I think a larger benefit here is years of quality life gained versus just years of life gained. I’ve known lots of out-of-shape people who made in into their 70s or 80s, but frankly many of them were virtual invalids by their mid 50s. They could hardly walk from their car to a store. They had to talk a bunch of pills which had their own side effects just to stay alive. My own feeling is keeping active probably gives you 2 or 3 more decades of quality life even if it has a relatively small effect on your life span. Another thing worth noting is if not for the state of medical science, the gains in life from exercising might be a lot greater. Those out of shape people would probably die by their early or mid 60s without modern medicine.

  22. Before all you Americans hang your head at the misplaced attention paid to helmet use and its negligible impact on actual safety for cyclists, please spare a thought for those of us in Germany who are subject to a concerted effort from the Transportation ministry and pushed by the auto lobby (VW, Bosch, Daimler, BMW) to lobby for ‘Helmflicht’ and bombard cyclists with helmet Propaganda.


    Brought to you by the auto lobby:


  23. There are more head injuries in car accidents than on a bike. Everybody riding a car should wear a helmet?

  24. If there are that many more cyclists on the roads, the drivers will eventually be kinder as a result (because many of them would be cyclists themselves), so that the fatality rate would not go up.

    Not all cycling deaths happen because of collision with automobiles. Fatality can occur with cyclists hitting other objects or each other.

  25. After estimated 300000-400000km on pedals in my life all across the World I must insist that “…Better street design and getting more people on bikes — not blind faith in helmets — are the keys to making cycling safer…” is just partially true, part “not blind faith in helmets”.
    It is NOT about more people on bikes or “safer” street design (latter may actually be more dangerous!-I’ll explain). It is in need to install basic understanding of traffic, traffic rules and behaviors from early on without ANY excuses for bikes. This includes police actually stopping and punishing cyclist for idiotic violations.
    Also, I can trivially explain USA “perplexing” numbers by what I have seen by my very eyes, perfect example: Long Island NY. Intersection of small two-way street with 8-10 lanes superhighway Sunrise Hwy near the place where it starts having street-function. Traffic light red for the small street. Speed limit on Sunrise Hwy is 55mph but people regularly drive 75-80mph. Well dressed middle aged man in all bicycling regalia AND HELMET on excellent new bike approaches intersection on the small two way road. Driving on the LEFT side. He does not even slow down on the red light. He pedals through. By sheer luck no reckless drivers in immediate vicinity, some nearby cars evade and horn, he could care less. Because he has all-powerful helmet I guess and might of rights above all others being “socially responsible” by driving the bike.
    What in part creates this “superman” in the USA?- Focus on only two issues: 1) Helmets 2) “Safer” street design and bike lanes. Helmet gives false feeling of invincibility.
    “Safer” street design and bike lanes alienate car driving public and bike riding public in both directions. Neither is exposed to real normal street interaction. Most of driving/riding happens in apartheid. One that can’t be recreated through the whole country or even a city. Not knowing how to drive along, once they encounter each other on a normal street-both sides do stupid things, things helmets can’t make safe.
    Solution?-Educate everyone (in my country of origin it happened in 1st grade) on driving bicycle on real wold streets both in theory and practice (yes, every class in 1st grade drove real bicycles through real busy streets as a part of education). Place higher demand on car driving education (tests I needed to pass in NY and NJ to get divers license are on par with tests in my country of origin to allow one to start driving w instructor on streets,in order to learn how to drive, for several months…). Finally, stop using traffic design and regulation for social engineering.

  26. Thank you. You’ve stated very clearly and logically why wearing a helmet when bicycling is a good idea. I particularly liked:

    “Helmets are meant to protect you from a head injury IN THE CASE of an accident, they are not meant to prevent the accident from happening.”

    This one sentence clearly describes the stupidity of the graph in this article.

  27. There are also many more people driving/riding in cars than on bicycles in developed countries. Not a valid comparison.

    You deeply missed the point of Mendi’s post. We all must draw a line (for ourselves) as to when an activity has enough inherent danger that wearing a helmet is warranted to increase our chances of survival from a crash. Helmets don’t stop crashes from happening, make us better bicyclists, make drivers more aware of bicyclists, etc. They only provide an extra level of protection for the bicyclist’s head if it hits something during a crash.

  28. Why, how often and what type of crashes occur is not correlated with wearing a helmet. Put another way…. how US drivers behave is not related to bicyclists wearing or not wearing helmets.

  29. In terms of numbers car occupant head injuries are an order of magnitude higher, with associated higher treatment costs (which seems to be the primary bugbear of most helmet zealots). If it save just one life, after all…?

  30. “real normal street interaction”

    We have that in almost every town and city in the English speaking world. And it invariably leads to death and injury among vulnerable road users. Not very equitable, I think you’ll agree.

    Without dedicated cycling infrastructure, no one habituated to using their car for every journey is going to change modes. The difference in places such as the Netherlands and Copenhagen is that many more people know they can safely change modes and use a bicycle when it better suits the kind of journey they wish to make.

  31. It’s still more or less the same. Especially in Amsterdam, with those narrow streets along the canals and a shortage on parking spots, bicycles are an cheap and practical way of transportation. And in the Netherlands bikes are very,very popular. So much that as good as every driver in a car will also be on a bike sometimes, so everybody knows how to handle the situation.

  32. “There are also many more people driving/riding in cars than on bicycles in developed countries…”
    And that’s not true for the Netherlands.

  33. Not objectively true! If you perceive less danger to a cyclist because they are wearing a helmet you are less likely to accept your own responsibility to share the road in a way that allows the greatest safety for the cyclist exercising her or his legal right to use the road.

  34. I see so many posts about cyclists not obeying traffic laws. The only laws that cyclists should be required to observe are the ones that keep the cyclists themselves and others safe. An example of a traffic regulation that cyclists frequently disobey because it is not usually relevant to them is the rolling stop at a stop sign. The bike approaches that sign at a much slower speed, and, usually, with much clearer sensory input than a person in a car. A cyclist does not need to come to a full and complete foot-down stop to know if it is safe or not to proceed.

  35. In Montreal, in 2018, I’m pretty sure that more pedestrians were killed than cyclists.
    And most were due to head injuries.
    Why don’t they force pedestrians to wear helmets at all times? I’m sre there would be a dramatic decrease pedestrian mortality, particularly among children.
    WELL, ISN’T IT?”®

Leave a Reply

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


More Evidence That Helmet Laws Don’t Work

If you want to increase cycling safety in your city, drop the helmet law and focus on getting more people– particularly women — on bikes, with street designs that offer separation from vehicle traffic. That’s the finding of a new study by researchers at the University of British Columbia [PDF] evaluating safety outcomes for cyclists across Canadian […]

What’s Behind the Rise in Cyclist and Pedestrian Deaths?

The National Highway Traffic Safety Administration is out with traffic fatality data for 2011, and the news is not good for cyclists, pedestrians or, for that matter, anyone who uses U.S. roadways. While motor vehicle deaths declined to a still mind-numbing 32,000, cycling deaths were up 8.7 percent, and 3 percent more pedestrians were killed. […]

Don’t Believe the Headlines: Bike Boom Has Been Fantastic for Bike Safety

The Governors Highway Safety Association released a report Monday that, the organization claimed, showed that the ongoing surge in American biking has increased bike fatalities. Transportation reporters around the country swung into action. “Fatal bicycle crashes on the rise, new study shows,” said the Des Moines Register headline. “Cycling is increasing and that may be […]