Safety In Numbers

In his landmark 2003 paper, Safety in numbers: more walkers and bicyclists, safer walking and bicycling, Peter Jacobsen found that accident rates involving motor vehicles and bicyclists/pedestrians decrease as the number of bicyclists and pedestrians on the road increase. From the Abstract:

Results: The likelihood that a given person walking or bicycling will be struck by a motorist varies inversely with the amount of walking or bicycling. This pattern is consistent across communities of varying size, from specific intersections to cities and countries, and across time periods.

Discussion: This result is unexpected. Since it is unlikely that the people walking and bicycling become more cautious if their numbers are larger, it indicates that the behavior of motorists controls the likelihood of collisions with people walking and bicycling. It appears that motorists adjust their behavior in the presence of people walking and bicycling. There is an urgent need for further exploration of the human factors controlling motorist behavior in the presence of people walking and bicycling.

Conclusion: A motorist is less likely to collide with a person walking and bicycling if more people walk or bicycle. Policies that increase the numbers of people walking and bicycling appear to be an effective route to improving the safety of people walking and bicycling.

It’s a common misconception that increased bicycling and walking will lead to higher injury and fatality rates. The Jacobsen study successfully debunks this myth while providing a potent tool for advocates in their efforts to improve conditions for bicyclists and pedestrians.

Safety In Numbers

36 Responses to “Safety In Numbers”

  • Jay says:

    I’m surprised that this is news to anyone. The safest places to ride are generally the places where more people ride. If people expect to see bikes everywhere, then they’re more prepared.

    That’s why I think bike sharing programs like Velib are useful. In my mind it’s the absolute fastest way to get as many bikers on the road as possible, thereby making it safer for everyone.

    I live in a dense neighborhood by a T stop in the Boston area, and there are pedestrians and bikers everywhere. More of people walking and biking doesn’t mean that more of us die, it means people in cars pay more attention to you because they know they have to be vigilant.

  • Christina says:

    It’s cool that someone did a study on this because it’s nice to have facts to back up what is common sense- people (motorists, cyclists, and pedestrians) pay more attention to stuff when there’s more of it. It would be great if we paid attention all the time, but barring that, having more cyclists and pedestrians to pay attention to is what’s going to work.
    It might help with road rage too- if motorists see bicycles as common, one of the many things that commonly hinder their forward progress, instead of a special extra thing that’s slowing them down.

  • Logan says:

    Leaders in Black Hawk Colorado and St. Charles County, Missouri that are banning cycling “due to safety concerns” should read this research. The oxymoron belief that cars are so much safer than bikes when car accidents are the leading cause of death in people under the age of 35 is outrageous, especially when this flawed logic is the foundation of public policy. Jonathan Maus at bikeportland recently wrote a great article reflecting on this problem entitled “Missouri county considers biking ban (and why you should care)”.

  • Nate Briggs says:

    This is the true meaning of Critical Mass, as far as I’m concerned.

    Just getting more riders out there.

    Having been a Lone Rider for many years, I appreciate the changes that are occurring as the number of transportation-oriented riders increases (the Pretend Racers aren’t really relevant in this context – they’re out there, but their relationship with other vehicles is, by definition, adversarial).

    This study also directly contradicts the perspective that motorists will just “do whatever they want” regardless of traffic situations. Motorists DO NOT want to hit people. The vast majority of motorists will be more careful in situations that demand more care.

    – Nate (SLC)

  • Erik Sandblom says:

    Safety in Numbers has been examined in the UK by the CTC, the British cycling organisation. They found that towns with a lower rate of cycling deaths also had more cycling.
    CTC: Safety in Numbers

    The Streetswiki also has more on Safety in Numbers.

  • peteathome says:

    While this may be true in general, the Copenhagen study showed that adding a bicycle facility to a road increased both the number of users and the car/bike collision rate, with the collisions increasing more than the increase in users. So the rate of collisions went up.

    This could be because Copenhagen has already reached a “critical mass” of safety in numbers, if such a thing is a true causality and not a false correlation. Maybe the effect saturates at some point and the increased dangers of the facilities exceed the increased safety of more bicycle users.

  • Erik Sandblom says:

    peteathome, which study are you referring to? The one below states that cycle tracks increase bicycle traffic by 20%, while injuries increase 10%. See abstract. Injuries at intersections increased 18%, see table 3.

    Bicycle Tracks and Lanes: a Before-After Study (PDF) Trafitec 2008.

  • peteathome says:

    Yes, this is the study. It id not show a safety in number effect. it showed that diverting car traffic off a street made that street safer for bicyclists. But the increase in bicycling numbers on a given street did not increase safety. And both the cycle track and bike lane designs decreased safety. Please read the entire study. They DO say that in spite of the increase in bicycle injuries they felt the designs were OK because bicyclists felt more comfortable on these facilities and that was a positive thing in itself, good for health and the environment because it might encourage more people to bicycle in the long run ( or slow down any decrease in bicycling).

    Table 3 is showing the expected number of accidents and injuries based on a regression between total number of automobiles and bicycles before and after the cycle tracks went in. They lump bicycles with mopeds together, but elsewhere say that bicycles made up 95% of that traffic category so the numbers can essentially be considered bicycles.

    A positive number means more injuries than expected, a negative number less.

    In table 3 it states that the actual number of total injuries for bicyclists were 10% higher after the cycle tracks went in than would have been expected for the total number of bikes and cars observed.

    Interestingly, the total number of pedestrian injuries were 19% higher, and automobile drivers 4% higher ( probably not statistically meaningful), so that ALL road users had higher injury rates than expected after the cycle track went in.

    These numbers aren’t terrible, just surprising to the authors who thought the tracks would improve safety, not decrease it slightly.

    Bike lanes were, surprisingly, much worst, with the bicycle injuries 49% higher than expected. Pedestrians had their injuries go down 17% and drivers go up 12%.

    I think the pedestrian problem with the cycle tracks, based on my reading of the more detailed Danish language study, was that people getting off buses were getting hit by bicycles as they tried to cross the cycle track to get to the sidewalk. I wonder if there is an easy solution for that.


    In the introduction of this paper it did note that ridership went up on the cycle paths more than the injuries did – in the full paper you see that is because they got rid of street parking so many few cars used the street. But the number of accidents was much higher than they should have been with the large reduction of motor traffic, especially if there would have been a safety in numbers effect. That’s what the regression analysis is trying to account for: what is the INTRINSIC safety of the design.

    You can say – well, getting rid parking made it OK in terms of safety. But Copenhagen does not feel that is a sustainable approach to traffic design. They figure everyone has to go somewhere and they can’t simply get rid of the cars on cycle track streets – sooner or later that approach would restrict the streets bikes could use too much and make bicycle transportation unfeasible, so it’s not a good solution.

    Also – the thrust of this study is what are the best facility designs to minimize accidents. It’s trying to estimate the intrinsic safety of the design – which would include any safety in numbers effect from increased bicycling. The study showed getting rid of parking made the street safer for bicyclists and putting in the tracks made it less so. So the best practice design, in terms of safety, would be to get rid of parking and do nothing else. Don’t put in either tracks or lanes as they make safety worst.

    This is one of the best and most complete studies I’ve seen, with extensive bicycle counts and injury measurements before the street was modified and then performed again after the modification. I don’t think anyone else has done an extensive study like this. Most facility studies just compare the rates of bike injuries on streets with lanes, say, to those without, without taking into consideration that the streets with lanes might have already been safer for bicyclists without the lanes.

    This study was performed by a “bike friendly” agency, in a “bike friendly” city with high ridership which really expected to see a safety improvement but didn’t.

  • John Schubert says:

    Hi all,

    The Jacobsen study is, unfortunately for its fans, mathematical garbage. It doesn’t prove anything.

    I know how enticing it sounds. Add more cyclists, and they become safer. Just like that.

    When you peel apart the causes of crashes, this seems questionable. And when you look at how Jacobsen presented his data, the kindest thing you can say is, “Jacobsen did not prove his hypothesis. He proved a tautology.”

    First, the causes of accidents: Roughly two thirds of all reportable bike accidents in the US do not involve a collision with a motor vehicle. These accidents — “reportable” because they generate a hospital visit — involve a collision with a fixed object, a collision with a pedestrian or other cyclist, or loss of control not involving any collision.

    These accidents do not go away because of numbers alone. I don’t think anyone reading this is less likely to blunder into a phone pole because there are other riders on the road.

    Now… of that remaining third … somewhere around half are clearly the fault of the cyclist: blowing red lights, riding at night without lights, wrong-way riding so they sneak into intersections from unexpected directions, that sort of thing.

    That leaves one sixth of all bike accidents that are collisions with motorists that are the motorist’s fault. If awareness could wipe out all of these (a rather optimistic “if”), that would generate a one-sixth (17 percent) drop in accidents.

    But the benefit Jacobsen claims is far greater.


    Jacobsen got data from several different places, graphed it, and the resulting graphs look dramatic. The X axis is the amount of activity; the Y axis is the accident rate, and when X is near zero, Y is quite high. The curve drops steeply as X increases, then the slope decreases as the accident rate approaches zero. It looks terrific. More cyclists, more pedestrians, more safety. What’s not to like?

    Here’s what:

    These graphs are a mathematical tautology, and at least two researchers have proven so.

    The first of these researchers is cycling author John Forester, who did us the service of explaining the math succinctly:

    “[Jacobsen’s] plots from many sources of data all show what I would call a quasi-hyperbolic curve or a quasi-decreasing-power curve,” Forester e-mailed me.

    “If N is the number of accidents to cyclists, C is the number of cyclists, and P is the size of the population, Jacobsen is plotting N/C on the Y axis and C/P on the X axis. Note that C appears as the numerator of one fraction and the denominator of the other fraction.”

    But it turns out that you don’t need to count all the cyclists in Copenhagen to get that nice looking curve. Any old numbers will do.

    “I produced 300 random 4-digit numbers by running down the columns of a standard telephone book,” Forester continued. He plugged these random numbers into Jacobsen’s equations. . . and got essentially the same steep-slope graph.

    Got it? That’s no conclusion Jacobsen reached. It’s a tautology. You plug any numbers into Jacobsen’s equation, and that’s the shape of the curve you get.

    And Forester is not alone in experimentally verifying this tautology. So did Ryan Conrad, a Portland, Oregon graduate student in traffic engineering. Conrad used a random number generation feature in an Excel spreadsheet to get his random numbers. Same result.

    Forester explained why it works this way:

    “It’s easy to see what happens. Each of the three numbers is a random selection from 10,000. Now consider the effect of the random number labeled C as number of cyclists. When C is small within its range, then, on average, N/C will tend to be large, and C/P will tend to be small, and when C is large within its range then N/C will tend to be small and C/P will tend to be large.”

    I’m sure there ARE safety factors that correlate with numbers. The average rider will have more years of acquiring knowledge and experience, greater average age, better access to education, and so on. But these other factors are the causes. Numbers aren’t the cause. Let’s not assign “safety in numbers” magical powers that it doesn’t really have.

    John Schubert

  • Erik Sandblom says:

    @peteathome, thanks for your reply. Now I know what study you are talking about :)

    As you know, it’s known that bicycle paths do not increase safety a intersections. However, this might be changing with advanced stop lines and pre-green for cyclists at traffic lights. Did the studied bikeways have these? It seems the study covered a long time period, when these were uncommon.

    @John Schubert, 90% of cyclist deaths in the USA involve a motor vehicle. Regarding the math, I need to think about it but Forester’s claims are discussed here:

    Ask the Experts: Paul L. Jacobsen, and Dr. Lon D. Roberts, PhD.

  • peteathome says:

    Advanced stop lines are more useful for bicycles making left turns. Since Danish law doesn’t normally allow bikes to make in-street left turns, that’s not very useful.

    Advanced green lights are problematic. They only work if the bicyclist is already at the intersection waiting for a green light. If the bicyclist arrives after the advanced green is over it does him no good.

    The Netherlands normally use bicycle-only phases for their lights. This would reduce most of the intersection problems. However, most intersections are not controlled and that’s where a bike gets into trouble. You can’t put lights at all intersections, it would slow everybody to a halt. The Dutch approach slows bikes down quite a bit as it is ( bikes can only go when they have a bicycle green, so they are sitting at lights a lot more, as are cars when they have red and bikes green) and I’m told there is some resentment against them.

  • Rick says:

    @John Schubert:

    John, one of the basic tenants of finding truth in the modern age is to “follow the money”; when I’m trying to decide if someone is an either an impassioned believer or a corporate shill usually comes down to this little test: does someone get paid to say what they just said?

    Where does one get their funding? If one expresses a view deriding something new or different, the first thing I ask myself is, “I wonder who writes the checks to those guys?” If one has corporate masters, and those corporate masters make money off the status quo, then I usually disregard anything that person says, because it just comes down to their being a corporate flack in another uniform. Just watch any political talk show, and you’ll see what I mean.

    So, John, where does your company get its money from?

  • rdhd says:

    Before Rick accuses me of taking money from some anti-bike people: I get no money from anyone but I do have a PhD in economics with plenty of training in econometrics. I’ve read Jacobsen’s study. It shows a nice correlation and nothing else. In particular he does not account for differences in physical facilities, educational campaigns, or other things that can affect bicycling safety. At best, his time-series data could be used as a proxy for driver education, but even then he does not account for it in his functional form. The author simply provides no legitimate model that shows what things can affect safety and then statistically estimates the magnitude of those affects.

    Sure, my gut reaction is that there is safety in numbers; but this study simply does not show it.

  • Erik Sandblom says:

    rdhd, we know that correlation is not the same thing as causation. You don’t need a PhD in econometrics to understand that.

    But it’s a stretch to say that Jacobsens study does not show that there is safety in numbers. With the same logic, seeing the moon does not prove it exists, and the only acceptable proof is to go there and walk on it.

    To the best of our knowledge, there is safety in numbers.

  • Rick says:


    But that’s my point: when I wrote, “when I’m trying to decide if someone is an either an impassioned believer or a corporate shill usually comes down to this little test: does someone get paid to say what they just said?”, I meant exactly that. If you believe the study has little validity, can give a reasoned explanation as to why you feel that way (and you certainly did), and aren’t getting money from folks who would profit from the information in the study being squashed, then we’re having a real debate. Which I love.

    I’m sure you can appreciate that I’m just trying to bring to light the problem that happens when those folks who should recuse themselves from that debate don’t.

    Thanks for your thoughts–I’ll try to come up with a reasonable reply after the first cup of coffee sinks in… :-))

  • Rick says:

    I’m beginning to get my feet wet in the world of transportation advocacy, and one of the things I’m asked—on a regular basis—is “why aren’t there more cyclists?” Actually, I thought there were , but then again I have very little to go on: my world exists primarily within a 15 mile circle of the California capitol building, and there always appears to be swarms of people on their bikes. However, the very few times I do get to explore the Sacramento suburbs, I find very few people cycling. So, yes: why aren’t there more cyclists?

    I’m hardly an expert in the field—I’m a critical care nurse by training—but because of the work I do, I suppose makes me at least an expert in annectodal bike safety: having to work on someone in the ER because of a bike/auto accident isn’t a lot of fun, and afterwards you’re always wondering if the accident could have been prevented somehow—which got me here, I suppose.

    The biggest problem I see in the Vehicular Cycling approach is that it comes down to trust: In short, while adhering to the V.C. tenets, I assume the driver will know what to do when I ride correctly. And frankly, in my experieince, they don’t. My biggest fear while riding is that I’ll be puttering along at 12 mph on a lovely day, riding safely on my side of the road in a bike lane, and some idiot, going a few miles over the speed limit because he’s late for something, will be fussing with (a) his kids in the back seat, (b) another car, (c) his cell phone, (d) his radio, or (e) all the above, slip over into the bike lane and kill me. It’s for no reason at all, and of no fault of my own, but I’ll be just as dead as if there was. I don’t need numbers to prove I’m not safe with a car going 40 mph within two feet of my elbow; I don’t feel safe, period. And neither does anyone else.

    So what seems to me to be common sense—since we can’t trust drivers, and we want to bicycle more and use our cars less—that we need to go ahead and plan for more segregated pathways. I believe that we’ve reached a saturation point, where the cyclists we see on the roads now are the ones most easily able to comes to terms with the inherent risk of interacting with cars, and until we’re able to build more infrastructure just for bicyclists, we’re going to be in a no-growth phase, and no matter what the price of gas. Until then, those more adverse to risk aren’t going to get on a bike no matter what, and until we can build that infrastructure to where we see more women and their kids on bikes for everyday errands (let’s call it the “Mom Test”), we’re just going to be stuck where we are.

    I’m sure that the advocates for Vehicular Cycling are well-intentioned—although I would also caution that it also seems to me to be the perfect answer to those developers and cities wanting to do things on the cheap as well, and as such should have their motives challenged whenever possible—but I believe the time has come for us to create a new paradigm when considering Active Transport issues in this century.

  • Alan says:


    I agree with you on just about every point other than the part about being in a no-growth phase regardless of gas prices. We saw a significant jump in bike ridership and sales when gas approached $5 per gallon a couple of years ago, which leads me to believe another spike in gas prices has a high probability of doing the same again. From my view, it looks like a bit of a chicken/egg scenario: we need more and better infrastructure to attract new riders, but we need the political support of large numbers of new riders to obtain the funding for the type of infrastructure you’re envisioning. Perhaps a spike in gas prices is just the spark we need to break out of these doldrums.


  • peteathome says:

    it’s not a matter of VC-types being well intentioned or not, it’s about facts. VC is about something else all together – how to ride safer on the streets, not whether streets are safer than paths.

    The facts show that the separated pathways you feel more comfortable on are actually more dangerous than riding in the street. In spite of all the driver misbehaviors you describe.

    OTOH – you feel safer the more segregation you achieve from automobiles. While you might be less safe, the feeling of safety attacks more bicyclists.

    Many facility people say that the feeling of safety is more important than actual safety because, by attracting more bicyclists, the “safety in numbers effect” will eventually make them actually safer.

    So it’s important to understand if this effect is real or just a false correlation resulting from other factors. For instance, countries that build a lot of facilities may have slower roads, fewer cars, and that is why they are safer for bicyclists. If so, it’s not the facilities you want, but slower speeds and fewer cars and one should look for a more direct way of achieving that goal instead of potentially dangerous facilities.

  • rdhd says:

    @Erik Sandblom But my whole point is that Jacobsen’s methodology does NOTHING to assign causal effect. It is only a correlation he is showing. List all of the things that you think make cycling safer. Were any of those things included in his analysis? Only one: number of cyclists.

    Our best knowledge may be that there is safety in numbers, but this study does not show this.

  • Sharper says:

    I like the way you think! For my part, I’ve been working on my own formulation of the chicken/egg problem, looking at it from the transportation bicycle advocate’s standpoint.

    At least here in Sacramento (and midtown specifically), I’m seeing a shift from the more traditional advocacy and its focus on more infrastructure and political pressure towards “social” advocacy, where the social act of riding draws people together and celebrates and extends bicycle culture. You see this in newly minted bicycle-friendly businesses, overwhelming local support and media coverage for the Bicycle Kitchen, fashion shows, the Bicycle Film Festival, pub rides, and myriad other bicycle-as-social-tool events, small and large.

    I suppose the difference is loosely analogous to the difference between “supply-side” economics and demand-driven markets. The former states that sufficient infrastructure will create riders, the latter that sufficient riders will prompt infrastructure improvements (no apologies from myself to the businesses whose doors I occasionally lock my bike to, however; get a proper rack).

    Sacramento seems to be at that tipping point where traditional advocacy might need to give way to social advocacy, and I wish I had the time and resources it would take to figure out some better guidelines.

  • Rick says:

    Peteathome, when you write, “it’s not a matter of VC-types being well intentioned or not, it’s about facts”, please note it’s a fact that for many people, riding in the streets is terrifying, and as long as they’re terrified, they won’t travel by bike. Period.

    I know I need to do more research about this, but please understand that I’m fairly skeptical of using statistics when public policy is being debated: simply put, everybody lies, and numbers can be (and often are) manipulated to come to a pre-ordained outcome. One of my favorite aphorisms is: “There are three kinds of lies: lies, damned lies, and statistics”, and that pretty much sums up my position on the subject.

    When rdhd said, “Sure, my gut reaction is that there is safety in numbers; but this study simply does not show it.”, all I would ask him (and you) is to let go of your trust in numbers, and just ask yourself this: do you really think your mother would be safer having cars at speed pass within a couple of feet of her while she’s on her bike? I just can’t believe it, really, and like Occam’s Razor, the easiest answer is usually the best: how could you trust your mother’s life to someone who you don’t know, and have no idea what shape they–or their car–is in? The answer, of course, is that you wouldn’t. Why would you if you didn’t have to?

    If you think segregated facilities are more dangerous, why don’t we think about ways to make them safer? Let’s steer the conversation in that direction, because I’d love to hear what you think about it.

  • Erik Sandblom says:

    @rdhd, “But my whole point is that Jacobsen’s methodology does NOTHING to assign causal effect. It is only a correlation he is showing.”

    I already conceded above (July 19th, 2010 at 7:09 am) that correlation and causality are two different things. Are you saying that a correlation is of no interest? Isn’t that like saying “dropping a ball 100 times doesn’t prove that gravity exists”?

    Jacobsen refers to a before-and-after study:

    Automobile speeds were reduced as anticipated. However, it seems as if the positive effect of this was more or less canceled out by increased bicycle speeds. The safety per bicyclist was still improved by approximately 20%. This improvement was primarily caused by an increase in bicycle flow, since the data show that more bicyclists at a given location seem to benefit their safety.

    Of course this might be due to other factors, but I think the evidence is compelling. One of the factors affecting pedestrian and cyclist safety is the safety in numbers effect.

  • peteathome says:

    First – the dangers of sidepaths have been documented MANY times, both by studies in teh US and in Europe. They are considered dangerous enough that they are not a recommended design under the US DOT design guide . I suggest you do some research.

    I already agreed that many bicyclists feel safer on the more dangerous facilities.

    The only way to make sidepaths safer is to eliminate or control the intersections they cross. That’s where the accidents happen, just like on a regular road. But sidepaths make the intersections worst and thus increase the accident rate. Studies going back to the early 70’s show this and even recent studies by cities like Copenhagen, a very bike friendly city that wanted to show an increase in safety, found they were more dangerous. Sometimes reality cannot be overcome by beliefs.

    I’ve already mentioned one way to make them safer: Do not allow minor intersections on a sidepath ( driveways, commercial enterances, etc.) and have a light at most intersections with a bike-only phase. Unfortunately, the first consideration is very restrictive as to where one can put sidepaths and the second is an expensive solution that will also slow down motorists and bicycles.

    Instead of lights, one could have the sidepath cross intersections using overpasses. Expensive and won’t work in urban areas – too much real estate is required for overpasses.

    Or do like the Dutch do – merge the sidepath into the road at intersections and have a bike-only light phase. Slows everybody down – bikes and cars alike – and that improves safety. Expensive as I said, but that’s what it takes.

  • Sharper says:

    Off the top of my head, I’d expect that segregated facilities aren’t inherently more dangerous, but combinations of segregated and mixed facilities are, because they prevent motorists from learning how to interact safely and effectively with bicyclists on the roadways. As your ER experience shows, poorly shared roads are death traps for the two-wheeled.

    But then this feeds back into the same decades-old circular problem bicycle advocates face. Poor sharing makes the roads seem less safe, depressing bicycling rates. With fewer bicyclists around, drivers get even less experience on safe driving around bicyclists. And the feedback loop works wonders: with low bicycle mode share, there’s little demand to expand bicycle infrastructure…

    If we want to make segregated facilities safer, we have to make all roads safer (and heck, substitute “safer” for safer if you prefer). And we do that by getting more motorists off the roads, more bicyclists on them, and getting them to coexist. Your answer for that problem is as good as mine.

  • Rick says:


    Pete, the Dutch method (merge the sidepath into the road at intersections and have a bike-only light phase) sounds intriguing: do you have a link where I can find out more about it?

  • John Schubert says:


    — I’m much too masochistic to get money from someone who stands to gain from suppressing cycling safety. I’m motivated by analysis and conscience. See my web site, Google my name. See for yourself. And be an above-the-belt debater and don’t accuse me of being bribed unless you have facts, rather than speculation and innuendo.

    — The bikelane-and-sidepath industry regularly ignores safety information. I have, on my hard disc, studies showing serious increases in accidents resulting directly from these facilities in Copenhagen, Helsinki, Amsterdam and Berlin. Many bikelane and cycle track designs now popular were tried in Davis, CA, and abandoned when researchers Lott, Tardiff and Lott documented increased accidents from those facilities. When’s the last time a bikelane design company offered to share that information with its audience?

    — “Separation” is a false promise. What you really do with infrastructure is screw up intersections and put people in each other’s blind spots. Which you then ‘fix’ with the smog-increasing delay of additional bike-specific traffic light cycles. Noncompliance with these expensive traffic lights creates its own safety hazards.

    — It will NEVER be safe or comfortable to ride in a bike lane — or a cycle track — with motor traffic whizzing by inches from your left elbow, and approaching driveways and intersections from a position that puts you where you are ignored. And THAT is what separation advocates are pushing.

    — The problem with ‘my own little space’ is that cyclists get the leftovers — the gutter, the door zone, and the place where they are ignored by other traffic. It’s not safe or comfortable to ride there. That’s one reason why 95 percent of Portland residents said ‘no’ to the bicycle this morning. We’ve been told so many times that this crappy little space is good for us, and, sadly, it takes a real iconoclast to say, “no it’s more dangerous.”

    — Europe’s mode share? As you would say, follow the money. My sister in law in Germany pays $38 for a quart of oil for her car. It isn’t just the cost of gas — everything about auto ownership is punitively priced in Europe. Note that my sister in law has a car anyway. Culturally, Europeans enjoy their cars in a way few Americans do.

    — I think I understand your reservations about ‘vehicular cycling’ (I don’t call it that). But here’s the next level of understanding: you can both test and manipulate the behavior of other road users, including overtaking motorists, far more than you realize. See and

    — You mention your work as a critical care nurse. My work is in accident reconstruction and accident prevention. Both my personal experience and the many shelves of studies I have show a pattern: accident causes are almost always something the bicyclist could prevent. The cry for infrastructure gets in the way of accident-preventing, in ways you’ll better understand after seeing the “you lead the dance” video.

    — You write, “do you really think your mother would be safer having cars at speed pass within a couple of feet of her while she’s on her bike?” That’s exactly the drawback of separated facilities. You are unwittingly obeying a taboo imposed by the motoring society — that you don’t have the right to use a lane. So you want the leftovers. Neither a paint stripe nor a little barrier curb makes the leftover safe. (Note that four cyclists in cycletracks in Amsterdam died in right-hook collisions with trucks in a single year.)

    — Separated facilities don’t remove the need for others to watch for your safety. But they hide you from those others, and make that process more difficult.

    — Regarding ‘combinations’ of facilities: When the rules change from one intersection to the next (as they do in Cambridge and Portland) you can expect people to make mistakes, that, on some occasions, will be deadly. Consistency is essential.

    John Schubert

  • Rick says:

    John, before I was a nurse, I spent two years as a Legislative Assistant on Capitol Hill; to that end, I believe I’m pretty qualified to challenge anyone’s viewpoint when they say things like, ” I’m motivated by analysis and conscience”: I’m sure the flacks who represent big pharma, big tobacco, and the gun lobby all say the same thing, and I don’t see them creating a better world, ok? Questioning your motives when they’re so obviously one-sided seems to be a reasonable thing, and you still didn’t answer my question about where you get your money–although based on what you just said, I would imagine it’s mostly law firms who represent the interests of the auto insurance industry. And since you seem to represent those interests, don’t you think you should recuse yourself from this debate?

    Just askin’, you know?

  • peteathome says:

    Rick – shame on you. You bring nothing to the conversation by casting aspersions at people.

  • Alan says:


    John Schubert is a paid consultant and expert witness on this subject so I figured he can handle the heat, but I agree, it would be best if we keep it on topic and focused on the issues under discussion.


  • Rick says:


    To me, the problem with having John’s opinions stated here is that he does this sort of thing for a living, can bring reams of one-sided data to back up his claims, and can–because, again, he does this for a living–over helm those of us who are interested and passionate, but complete amateurs.

    When Gary Fisher posts his perspectives here, or Grant Petersen, or another well known individual in the industry, it’s never to say “My way is the right way”; rather, it’s to act as a teacher or guide in bettering our understanding of the things they do everyday. But not John: oh, no–it’s “mathematical garbage” if he doesn’t agree with the topic, and I personally felt that if he couldn’t see both sides of the argument (and how could he? He makes a living going to court and testifying for that one side), then it was unfair for him to be partaking in this discussion.

    I may feel strongly about this subject, but I can’t back up my claims with data; what I’ve tried to offer is a perspective that says “if something doesn’t feel right, it probably isn’t”, and nothing more. Going toe-to-toe with an expert like John is like going up against Lance Armstrong in a race: he does this sort of thing for a living, and I could never compete with his strong opinions that are backed with a boatload of data–it’s just impossible.

    So, John, I’ll call you later on today to apologize–I’m sorry it became personal, and that is uncalled for.

  • peteathome says:

    I would point out that most of the advocates of facilities are people who receive money from the government, agencies and donations and are professional advocates. You might want to ask yourself what their motivations are. They need to get more people on bikes to justify their existence. Some of them are not actually interested in bicycling per se but in ecological issues and they only see bicycling as a way to reduce car dependence. Few that I have ever talked to care that much about actual safety, just more people on bikes. So you have to worry about THEIR motivation as much if not more than someone like Mr. Schubert. They justify their lack of concern about facility safety many ways – that the health benefits of bicycling outweigh any risk, that if more people bicycle it will get safer do to “safety in numbers”, etc. But they don’t necessarily have your best interest in mind when they promote facilities.

    Well, I am NOT a paid consultant. I receive nothing from any agency/group involved in bicycling infrastructure. I AM interested in bicycling is a safe and enjoyable manner for myself and I’d like more people to adopt what I find to be a very nice way of getting around. I also have a science and technology background. So MY self interest is being able to continue to bicycle in a safe and pleasant manner. I don’t do it for ecological or other motivations but because I like getting around by bicycle.

    Over the years I’ve found myself involved with people who push ideas to get more people bicycling. I like to think of myself as an ethical person, so before I jump on board with these people, I want to see if their ideas actually do what they say.

    My ethics don’t allow me to promote ideas that give the “illusion” of safety when they don’t actually, directly, achieve safety. I’m not going to tell people a facility will make them safer if it actually makes them less safe at first and then maybe, in the future will make them safer because it attracts bicyclists and in the long run makes people in general safer. That’s just not ethical by my standards.

    I’ve looked at the same studies that Mr. Schubert mention and I have to say I agree with what he says. And I’ve never seen technically good studies that challenge them. Only statistical studies that indicate that places where a lot of people bicycle tend to have lower injutr rates.

    Again, i recommend reading completely through the Copenhagen study. their results match many other studies yet were undertaken by a group I think any of you here can trust. You don’t need higher math to read and understand it. You can ignore the math and just read what they say. Their math is accurate as far as I can tell.

  • Rick says:


    Pete, I’m not challenging your (or really, for that matter, John’s) motivations on the subject–just the approach. I just viewed the video John asked me to look at, and I was struck by one thing: that’s the way I ride everyday. It’s hard to disagree with the techniques–but with one caveat: where are the moms and their kids? How in the world are we going to get them out of their SUV tanks for the ten or so local trips a day by asking them to ride like the folks in the video? I just don’t see that happening–they’ll never feel safe enough to go. I imagine that telling them to “just be confident” probably won’t cut it.

    So what’s your plan for passing the “Mom Test”? How would you get them out of their cars?

  • peteathome says:

    Mom Test – there are many solutions. What I won’t accept, personally, are facilities that are designed to make people feel safer when they are actually less safe.

    Moms and their kids already ride secondary streets with kids. Some with Cargo Bikes. See this website. Education can help in this area, as well as law enforcement.

    I personally don’t think there will be enough demand in the USA to justify the type of facilities that are “comfortable” for “moms” and are safe. Instead we put in half-baked systems that are neither safe nor useful for most bicyclists.

    Most of our cities just aren’t flat and dense enough, with reasonable weather ( not extremely hot in summer or arctic in winter) to attract a large modal share of year round cyclists. Some smaller university towns like Davis and some of the old, dense East Coast cities are major exceptions that have/can be developed.

    Instead, I think bike facility people need to jump on board with other groups that might head in the same direction. For instance, with an aging population more and more older people really shouldn’t be driving large, high speed automobiles. But mobility is incredibly important in our society. So maybe we should be pushing for fully-connected, low speed secondary streets for Neighborhood Electric Vehicles, bikes, etc. Any vehicle could use these streets but the speed limit is set to around 20 mph. The NEVs are light and generally won’t do a lot of damage in a collision and don’t look threatening to fearful cyclists. More like golf carts.

    Most cars would use the regular streets since they are faster. The NEVs can’t.

    Fully connected mean you can get from any point A to B in a reasonably straight line, using traffic lights, ramps, etc. to cross busy roads.

    Speed reduction really does increase safety for bicycles. But enforcement would be important to control the speed.

    LOTS of older people who, as a group, are politically powerful. So there is enough political capital for the NEV idea, I think.

    So that’s one idea. I’m interested in hearing other ideas that are politically possible, result in safe and useful networks and don';t limit where bicyclists can go.

  • John Schubert says:


    My phone hasn’t rung yet. My phone number is on my web site.
    I’m amazed by your reaction to the fact that I mention data. Instead of looking at, or exploring the links on, you clamor that I’m heavy handed and accuse me of being on someone’s payroll.
    I don’t apologize for knowing my stuff, any more than you should when a patient codes and you need to know what to do to keep the patient alive.
    Moms and their kids? It’s linked to on this very page: And as Pete says, it is unethical to offer the illusion of safety while actually endangering people. Poorly designed facilities do just that.
    You seem unable to process the notion that many of us found a different way of looking at the bicyclist’s place on the roadway. You’re looking for the catch, hence your accusation of intellectual prostitution. Those accusations, by the way, are beneath contempt. It’s a laughable concept that there are law firms “who represent the interests of the auto insurance industry” that would benefit from suppressing information about reducing car/bike collisions.
    The real intellectual prostitutes are the design firms who build accident causes into their designs. I would be far wealthier if I were to go over to their side. As you would say, follow the money.
    Our way is lower stress, safer, available to anyone anywhere in the U.S., regardless of pork barrel spending or absence thereof. If you can accept the notion that this exists, that we find it a superior alternative to the pipe dream of shoehorning door zone bike lanes onto every street, then you’ll have less motive to attack the people who have found it attractive.
    Our way — I favor the term “bicycle driving” — is movement-based. Traffic moves. Bicyclists are traffic. We refer to bicyclists as vehicle operators because a bicycle has more in common with other vehicles than with pedestrians — the bicycle has a large turning radius, compared with a pedestrian; it has a much longer stopping distance; it can’t shift sideways; the bicyclist can’t look in one direction while traveling in another the way a pedestrian can. These operating limitations mean that some designs have built-in accident causes.
    You dropped the names Gary Fisher and Grant Petersen. I’ve known both of them for about 30 years; Gary lived at my house for two months, and he and I have ridden hundreds of miles together. I wouldn’t consider either Gary or Grant shy about expressing an opinion and citing facts to back up that opinion.
    So…. let’s start with data.
    The links on my web site will keep you busy for a long time.
    Here’s a study that shows accidents of almost all types going UP in Copenhagen, as a result of their facility design:
    … which leads to the question: assuming their accident rate really is low (it’s hard to learn about data collection methods in my own back yard, let alone in Denmark), why is it low? I attribute that to the Danish cyclists’ advanced age, slow speed, lifelong instruction in bike handling skills. But the data clearly show that the facilities aren’t causing a low accident rate. Rather, the facilities increase accidents. The lowest accident rate of all would come with facilities designed around the concept of bicycle movements, rather than bicyclists as a politically protected group, combined with better skills, awareness, and a culture of civility among all road users.
    John Schubert

  • Alan says:

    John and Rick, since this has heated up to the point of nearly boiling over, I’m going to ask that you take this offline and contact one another directly.

    Thanks all, for your participation in this discussion.


  • John Schubert says:

    Believe it or not, Rick and I spent 53 minutes on the phone, and he charmed me more than I would have thought possible. We are working hard to better understand each other.

    In particular, I express admiration for the fact that Rick has identified one of the most accident-prone groups of bicyclists in the nation (Spanish-speaking people who work on farms, in restaurants, and such jobs, who do a lot of night riding) and has begun work on interventions to get them to use safe cycling practices.

    Rick, don’t forget to send me your e-mail address. You have mine, but I don’t have yours.

    John Schubert

© 2011 EcoVelo™