As the saying goes, there are “lies, damn lies, and statistics”, and there appears to be a case for as much in cycling. We recently referenced a report through the Freakonomics blog that indicated that drivers were at fault in 90% of accidents that involve cyclists in Toronto. This finding seemed to contradict the widely held believe that cyclists are a bunch of lawless hooligans who are a threat to themselves and everyone on the road. This report certainly left a lot of cyclists (including this one) with a smug sense of superiority.
Well, it might be time for a reality check. The blog at OregonLive.com recently posted some of their own numbers based on police reports in America’s urban cycling mecca. Their study found motorists to be at fault in 53% of accidents. Now this was only a survey of accidents reported to police, but it is difficult to see how this number could get to 90% even factoring in unreported accidents and cases where the police officer’s conclusion was incorrect.
These numbers hardly leave drivers off the hook nor does it mean that the vast majority of cyclists are outlaws. Still, it does mean we have a responsibility to ride safe and not assume all motorists are out to get us.
on Sep 8th, 2009 at 12:58 pm
Major respect for posting this without being prompted. What I would presume based on little information so far is that it’s likely difficult to standardize methodology on who was at fault in accidents, and even if you managed, you’d run into problems with different participation in accident reports from city to city or from group to group.
on Sep 8th, 2009 at 2:06 pm
The portland statistics don’t have to get to 90%. The freakonomics blog reported on accidents in Toronto.
It would be interesting to see what the results of this study would be if you used Austin police reports.
on Sep 8th, 2009 at 2:17 pm
Tom,
I’d expect such a cycling hub as Portland to have similar numbers not that they’d have to be right at 90%.
Maybe the additional cyclists mean its an apples to oranges comparison. As you add more cyclists, you have more people who ride recklessly or in ignorance of their place in traffic/traffic laws? I’d bet our cycling population is more like Toronto in percentage than Portland. It would be interesting to see where we fall.
on Sep 8th, 2009 at 6:45 pm
There are fewer bike fatalities per cyclist in Portland than probably anywhere in the US due to cyclists being a regular part of traffic there. I’d guess that cyclist-caused accidents don’t change much relative to mode share increases, but driver-caused accidents decrease.
on Sep 9th, 2009 at 10:36 pm
Yeah, what’re the raw rates per thousand?
on Sep 9th, 2009 at 11:36 pm
I don’t think the real issue is who is at fault. Everybody makes mistakes, drivers and cyclists, but the issue is that the cyclist’s life is endangered, whereas the motorist is perfectly safe.
Every reasonable cyclist hopes that through diligence, they can ride safely.
Let’s suppose that *every* car/bike accident is the cyclist’s fault. What should be done? Do we just say “its those lawless cyclists’ fault, so we need not spend money on improvements or changes to infrastructure”?
What happens when its a car/car accident? In most cases, it is the fault of one or the other driver. Do we then assume that the roads do not need improvement, since it was the fault of one of them? Generally, if a pattern emerges, improvements are made, even though one of the drivers was at fault. Changes are made so that people are less likely to make mistakes.
The same process should be applied to car/bike accidents.
on Sep 10th, 2009 at 3:51 pm
Dave, when a lot of serious auto accidents were attributed to drunk driving, the logical course of action was to step up enforcement of DUI laws and tighten the penalties for DUI. Not to redesign the roads to make them safer to drive drunk (as if that were even possible).
on Sep 10th, 2009 at 5:17 pm
M1EK, you have a point, but consider also that they did try to make things safer for drunks, such as those barrel-collapsing cushions, guard rails, big bushes to slow down cars that go off the road, trees are generally not left within a certain distance of the road, etc.
The conversation about stop sign/traffic light enforcement is mostly about stemming the massive amount of irritation that bicyclists cause motorists. I don’t think bicyclists have the right to forcibly take the right of way in these situations, and I fear it raises the level of road rage that other cyclists will be subject to.
However, I don’t think this behavior is a primary factor in car/bike accidents.
on Sep 10th, 2009 at 9:36 pm
Props for the follow-up. I just wanted to add that the statistics upon which Vehicular Cycling principles were developed also mirror the 50:50 statistics from the more recent Portland study.
on Sep 11th, 2009 at 10:04 am
Dave, those safety features were not put in for the benefit of drunk drivers; they don’t help reduce the accidents between drunk drivers and other road users to any noticeable degree.
And, yes, cyclist behavior is a primary factor in car/bike accidents – in older studies that show that, for instance, riding on the sidewalk or the wrong way on the road or at night without lights [...]. Including failure to yield right of way.
on Sep 11th, 2009 at 10:20 am
Mike,
I think we can all agree that unsafe behavior increases the risk of accidents. Seems like a no brainer. The safety features Dave mentioned are there to reduce the damage to life and property if an accident occurs, and we ought to be building that infrastructure where ever we can.
James Kunstler argues that the extremely wide roads and shallow arch turns of roads developed in the 1950s were designed to allow cars to move as fast as possible with as little skill as possible. He continues that part of this equation was the common behavior of the time of driving under the influence. There is no proof of this, but it seems like a compelling conclusion.
on Sep 11th, 2009 at 1:12 pm
I’d be wary of drawing any points from Kunstler – he may write something you like one day, and then go off into the weeds so far the next day that you feel like an idiot for referencing him. (In other words, I’ve been there, done that).
The point is that we handle motorist behavior issues with enforcement all the time. It is quite reasonable to do the same with the cyclist behaviors that cause accidents.
on Sep 14th, 2009 at 12:00 am
M1EK,
I don’t have too much to disagree with you about, really. I do believe infrastructure can make even lawless cycling safer, just like there are some gains for motorists I mentioned.
Enforcement efforts, however, seem to me to be mis-targeted. Police and motorist outrage is rarely targeted at cyclists without lights, or wrong-way riders, or riders on the sidewalk. I agree that most of these can be pretty dangerous.
I was reading an article about Amsterdam cycling, and traffic enforcement, and the official was not very concerned about stop signs, but was concerned about cyclists without lights at night.
Here in America, it is the reverse, because we are more concerned with minimizing inconvenience to motorists than for the safety of cyclists.
Again: we are more concerned with minimizing the inconvenience to motorists than we are worried about the safety of cyclists.
on Sep 14th, 2009 at 9:11 am
You make an assumption about motive that doesn’t even match the data. Failure to yield right-of-way has, in fact, been among the top reasons cyclists have died in every study I’ve ever seen on the topic.
on Sep 14th, 2009 at 11:07 am
Well, I am heartened to hear there *is* some data, but I’m not too confident in the analysis if they feel this is the top reason. It doesn’t jibe with my subjective experience and news I’ve read. Accident “causes” can be very different depending upon who is making the evaluation.
To be a little silly, in the last hour or two of surveying stuff I could find online, I’ve found the top reason cyclists have died has been head injuries.
It seems to me that many studies are simply devoted to helmet use and its impact on fatalities. Maybe I’m looking in the wrong places, do you have any pointers?
on Sep 14th, 2009 at 11:46 am
Dave,
The reason your probably coming up with those results is that the top reason cyclists die is their impact with the car. The behavior of the cyclist and driver as well as road conditions are contributing factors but the impact is the reason for injury or death.
on Sep 14th, 2009 at 4:58 pm
“among the top reasons” != “the top reason” – I don’t even remember the order of the reasons – this would inevitably come up in a helmet discussion in the “how dangerous is bicycling, really, if you actually obey the rules” sense (showing that if you ride with lights on at night, ride the right direction, and don’t run red lights or stop signs, you’re unlikely to be hurt or killed).
And elliott is correct in a pedantic sense.
on Sep 15th, 2009 at 12:35 am
LOL, if you actually obey the rules, etc. and think you are safe, you’re going to be seriously injured or dead in about a year. I’d say the incidence of left hooks, right hooks, and being driven into from a side street are about once every 3-4 months. Quite a lot depends upon where you are riding on a regular basis. Most of these are avoidable if you are very alert.
I am still looking at reports. One of the interesting things I noticed is that in NYC, pedestrians are the largest fatality group, outnumbering motorists, bicyclists, etc. by a large margin. I am sure they are blasting thru those stop signs and traffic lights…
Another interesting thing is the disparity in cause factors between fatal and non-fatal accidents. In the NYC study, “bicyclists factors” only were cited in 42% of the accidents, “vehicle only” 20%, and both 36%. Yet in non-fatal crashes, “bicyclists factors” only were cited in 13%, “vehicle only” 35%, and both, 6%, no factors documented, 45%. They go on to say something like: because both parties are available to recount the accident, the non-fatal crash factors assignment may be more accurate. Well, as they say, history is written by the winners…
In the Toronto study, which got presented as “90% of bike/car accidents are the fault of the car driver” they specifically say: “Finally, any attempt to quantify such
factors as driver and cyclist error, inattention, negligence, or culpability must be carefully thought through. Assessing these always involves many layers of subjective judgements, in the statements from the individuals, the evaluation of the investigating officer, and the eventual interpretation by the researcher.” The Toronto study does not explicitly present its data in terms of culpability.
Another interesting item is the relatively high incidence of trucks and other large vehicles in fatalities. This was a sub-focus of the London study, and also turned up in a few others.
“Failure to yield right of way” is not always used in the manner in which I think this discussion is using it, but I’d say that generally about 6-13% of crashes involve cyclists running stop signs and stop lights.
A heartening item (to me) is that several studies don’t show significant levels of drunk driving as a factor in cyclists deaths. By this I mean perhaps 2-4% involve DUI for the motorist. The percentage is higher for cyclists BUI, I think.
on Sep 15th, 2009 at 3:02 pm
Dave, your first paragraph is objectively false. I commuted thousands of miles on my bike over a decade or so before the arthritis kicked me off, and had precisely one accident – and no motor vehicle was involved (went over handlebars after I had to brake suddenly on an oily patch south of San Marcos). I had a handful of near-misses, of course, so your advice to pay attention is well given, but your advice that “if you actually obey the rules” you’re somehow unsafe is, well, strikingly unsafe.
on Sep 15th, 2009 at 8:27 pm
Let me be perfectly clear (TM)
“but your advice that “if you actually obey the rules” you’re somehow unsafe is, well, strikingly unsafe.”
My position is that just obeying the rules will not ensure your safety. It is certainly better than not obeying the rules (in a large majority of cases…).
Every road user is subject to accidents, no matter how alert or mindful of the law they are. When you are riding a bicycle, the consequences of an accident are much more serious, and it is impossible to ensure your safety by your actions alone.
on Sep 16th, 2009 at 9:16 am
Your statement is still objectively false. You said, basically, that if you follow the traffic laws but don’t pay much attention to anything else, you’ll be seriously injured or dead within a year – this is absurd on its face. It’s the kind of attitude that underlies the “well, I’m safer if I run this red light to get out of the way” idiocy.
on Sep 16th, 2009 at 4:59 pm
Perhaps it is an exaggeration, certaintly it depends quite a lot on the circumstances under which you are riding. Say you live in the university neighborhood, and you don’t own a car. The Texas Driver’s handbook is not going to really tell you all that you need to know. It is not for nothing that there are courses specifically to help people learn to deal with city riding.
“It’s the kind of attitude that underlies…”
I don’t believe you know what kind of attitude it is, thus your spurious remarks.
on Sep 17th, 2009 at 2:41 pm
Dave, I live near UT; and, again, commuted thousands of miles over ten years.
In my experience, those who claim they are constantly getting into accidents or nearly so are exactly the ones who drive like lunatics.
on Sep 17th, 2009 at 2:51 pm
c /drive like lunatics/bike like lunatics. Ironic, huh?