Will Bicycling to Work Get You Killed?

That's a question the Freakonomics blog asked recently. My short answer: probably not. They disagree. They quote a study that states that:

Per kilometer, cyclists are 12 times more likely than car drivers to suffer a fatal accident

But that doesn't answer the question they asked. If you remove children from consideration that ratio goes way down (the study considered all cyclists). In addition I suspect it will go down if you remove inexperienced or recreational cyclists. If you were to compare bike commuters to automobile commuters I think you might find a fatalities-per-kilometer ratio for cyclist that is as good as, or even better than, the one for drivers.

Some numbers. In 2005, cyclists made up 6.3% of all traffic fatalities in the District. In 2004 it was 7%. In 2003 it was 0%. In 2002 it was 2.1%. In 2001 it was 2.9%. I believe in 2006 it was also 0%. That makes a 6-year average of 3.05%. Without knowing who constituted that group of cyclists you can't eliminate any for being too young or inexperienced so let's stick with the number. According to the 2000 census 1.21% of cyclists biked to work. Of course that number is out of date and poorly gathered (those are people for who biking is the primary means of getting to work). So, while it seems cyclists are over-represented in the numbers by a little under 3 times, it's certainly nothing like 12. And I think with better data, you see results that are even less scary. Of course, I'm not considering it per kilometer either.

One thing they do point out is that the added risk of injury (if there is one) is somewhat counterbalanced by the added health benefits of biking.

On the other hand, a Danish study found that people who do not bike to work suffer a 39 percent higher mortality rate than those who do.

Partly because biking is good for you, partly because driving is bad for you.

The post has some good links One of the links was to a Times article that mentioned

A journalist who regularly bicycled to work in Washington was killed when he rode headlong into the door of a truck as the driver opened it.

It also has a nice graphic, but it also has some things you might disagree with or find overstated.

Further injuries and deaths result from cyclists who ride illegally on the sidewalk and mow down pedestrians.

But it includes this nice stat (though as a way, oddly, of showing that cyclists are sometimes to blame)

One study attributed 60 percent of bicycle-motor vehicle accidents to the drivers and 17 percent to cyclists. But another study of crashes involving children on bikes found that 80 percent of those accidents were the fault of the bicyclists.

Which backs up what I was saying about kids above. Another great article is this one from 2006

“With the subway,” Mr. Ford said, “you’re literally in a tunnel. When I bike, I see faces. I see storefronts. I’ll stop to visit someone. I’m engaged in the city.”

Comments

[Contrarian]

I'm always suspicious of any statistics regarding bicycle accidents, the data collection is just not rigorous. I'm particularly suspicious of any statistics that purport to assign fault for collisions. In all but the simplest collisions there are multiple contributing factors -- either party could have avoided the collision by doing something differently. When there are multiple contributing factors, the only way to establish fault is through litigation, and relatively few collisions go to trial.

A simple but crude measure is to just add up the contributing factors. So if a motorist who is speeding hits a cyclist riding without lights at night, that's one contributing factor for each. I know of two studies done this way, one by the city of Toronto and one by the NHTSA. The Toronto study found that overall, motorists had slightly more than half of the contributing factors to automobile/bicycle collisions. The NHTSA study looked solely at right-of-way violations, and found that about 20% more collisions involved a motorist violating the right-of-way of a cyclist than vice-versa.

Local police departments are notorious for just making numbers up about the proportion of collisions where cyclists are at fault, to the point where it's best to assume in the absense of supporting research that any such statistic is made up. These made-up numbers serve an important purpose, however: they provide a guide on how the department in general views cyclists, and thus are the best available indicator of a city's overall cycle-friendliness.

[Wonger]

As I've argued before, comparing autos and bikes on a distance standard is awkward.

The automobile's ability to travel so much farther during a given amount of time gives the auto a huge advantage in statistical comparison.

For example, take a car driving 60 miles in one hour before having an accident. Compare this to a cyclist riding 12 miles in one hour before having an accident.

Is the cyclist truly having accidents at five times the rate of the auto? Clearly, I think not.

I believe a better standard would be to compare by time in motion, or perhaps by trip count. That is the number of accidents per transit outing.

I seem to remember that Washcycle had linked to some data earlier that did offer alternative standards, and the cycling fared much better, statistically speaking.

[Taleswapper]

And they're always ignoring this one: You're much more likely to kill someone when driving than riding a bike, right? How can you choose to drive with that on your conscience? :)

[Dan]

I usually commute to work by bike, but circumstances have dictated me driving my car the last couple of days. This has given me a bit more insight into why drivers can sometimes lose their tempers with bikers.

Picture this: it's dark except for the headlights and street lamps glaring off your windshield, cars are backed up a quarter mile, finally there's a space opening up in the next lane, the one you need to be in anyway, so you start to nudge over. Suddenly, a couple of bike commuters shoot thru the gap that you were heading for. They seemed to come out of nowhere because they didn't have any lighting whatsoever, and you almost hit them.

What you're probably thinking is not "There goes a couple of guys helping the environment by not riding polluting automobiles to work." No, what goes through your mind is this: "What a couple of morons."

Perhaps these are the future biking fatality statistics the study is referring to.

[Mike doan]

I don't care what the statistics say: I feel safer in my car than on my bike. I've had two bike accidents in the last six months (one my fault, the other not). During rush hour, in D.C. and Arlington, where I ride, traffic is too slow to cause many serious injuries in an automobile but you can still get killed on your bike. Still, knowing the risks, I'd rather bike.

[washcycle]

Sorry about your accidents. I hope you're OK.

It's normal to feel safer in your car. There are probably a lot of people who FEEL safer in car. It's like a little fortress. But we know there are people who feel safer in their car than in a plane, despite the statistics that show one is significantly safer in a plane.

Now I'm not arguing that cycling is as safe as flying - or even as safe as driving. Just that it probably isn't 12 times as dangerous.

Speaking of feeling safe. I always feel less safe when it's rained. Partly because I am, but partly because the wet roads make cars louder and sound closer. So fear is not entirely rational.

[bawler]

Thanks! Good to know the stats at least might be on our side.

But the fact remains that most roads are intimidating to bikers, regardless of perception or reality. Many of my coworkers countlessly say they will not bike because its not safe and they literally fear for their life. I understand them. I feel the same way biking on Mass Ave everyday.

All the more reason for more bike lanes, bike trails (that means you Met Branch), and bike boulevards & cycle tracks (get some innovation DC)!

[Paul]

The cyclist pictured above is certainly more likely to die en route--carrying multiple pieces of luggage, wearing dress shoes, not wearing a helmet, with no lights or reflective markings anywhere on her kit. At least she'll go the hospital or morgue in style...

[Richard Layman]

I had a bike-car altercation last night. My front wheel lost. Other than a skinned right knee, I am fortunately fine. I guess it's the fourth time. 2x I was hit with no damage, none. Once I fractured my ribs (that was partly my fault).