Saturday, November 10, 2012

Bicycle helmets II

There's a lot of bicycle helmet activism going on these days ... it's a big issue, trying to force riders, by law or moral suasion, to wear a helmet.

I don't get it.  What's going on?  Why are my friends are so horrified when I ride bare-headed, especially when there are lots of other things that don't horrify them?  In fact, I bet they'd laugh at me if I decided to wear a helmet when driving, even though, to me, it doesn't seem that much different.

Last month, I asked for a "framework" for how you decided that helmets should be worn (for those of you who believe that).  I figured this was a good audience to ask, since we're a group that has repeatedly tried to set frameworks for other questions, like Hall of Fame credentials.  We criticize those who want Jack Morris in the Hall, because we don't see any principle by which Morris belongs but, say, Rick Reuschel doesn't.  If you consider that helmetless riding belongs in the societal "Hall of Shame," then, I asked, what's the principle by which helmetless driving doesn't? 

I didn't really get any serious answers, either here or at Tango's blog.  I did get a few responses that tried a justification, but that don't hold up.

One answer was, riders should be forced to wear helmets because, when they get injured, it imposes costs on the rest of society -- such as, for instance, the cost of medical care.  But ... almost all activities impose costs on others.  People get hurt riding even with a helmet, and people get head injuries in car accidents.  Why are those things OK for society to pay for, but not riding without a helmet? 

In fact, I bet just one baseball pitcher costs society more than ten helmetless riders.  Arm injuries are pretty much inevitable for pitchers.  How many extra medical appointments do you think the average pitcher uses up?  But we don't ban pitching.

At best, the "it imposes costs" explanation is just question begging.  At worst, it's an attempt to deflect the question by asserting a principle that we're not actually willing to follow.

Another answer I got was, "because that's what society determines is unacceptable."  Which, again, is begging the question.  No religious activist would accept that abortion should be legal because "society" has "determined" it's "acceptable".  And no gay person would accept those reasons for why he shouldn't be allowed to marry his partner.

This answer is actually indistinguishable from "because we say so."  It differs only in that it uses bigger words that make it sound like it means something meaningful and important.

A third answer I've heard is, "I had a really bad accident and a helmet saved my life."  A good friend here in Ottawa actually told me that -- he wiped out and broke his collarbone, among other injuries, and it took him a long time to recover.  The crash broke his helmet into two pieces, he said.

But ... that's not really a principle.  It's an anecdote.  I think everyone should understand that one person's experience shouldn't bind the rest of us.

Just to hammer home the point ... here's an actual news story about a child that was shot while riding in a car -- but she survived because a bible slowed the bullet down.  Her sister was spared because the bullet later lodged in a watermelon on the child's lap.

Should we legislate that we all carry watermelons on our laps from now on?


With respect, these rationalizations are pretty weak.  It does seem to me like many respondents started with the conclusion -- *of course* you should wear a helmet! -- and then tried to build a set of principles around it, which didn't test out.

I think it also goes to show that the issue is sensitive and emotional, when a community that's so used to objectively analyzing conventional wisdom still comes up with these kinds of responses.  If someone had made these arguments about wearing a helmet when *walking*, I think the flaws would be pointed out in seconds. 

What I was expecting was something to do with risk, with costs and benefits.  And that did come up.  A few people pointed out, reasonably, that the risk of helmetlessness was higher for bicycling than for driving.  From there, I expected the discussion to start talking about the actual risks.  Maybe someone would look up some numbers, and say, "helmets would save X lives for bicycling, but only Y lives for driving, and I believe the cutoff for forcing people to wear them should be between X and Y."  And then, we could go from there, trying to figure out what the threshold should be, just like we try to figure out the threshold for Hall of Fame admission.

But that didn't happen.  I don't think *anyone* talked about how big the risk has to be, even in broad terms.  I don't think there was any serious attempt to set a principle. 

As for benefits ... I think someone mentioned that helmets in cars were a bigger inconvenience -- and therefore bore a bigger cost -- than the inconvenience of helmets for cyclists.  I'm not sure about that ... why would it be different?  It's the same helmet!  Regardless, if you acknowledge that comfort and convenience are factors, then, shouldn't you try to figure out how high the costs are for cyclists?  Clearly, riders without helmets at least appear to have different preferences than other riders.  So the costs aren't obvious.

And people are different.  If I love cucumbers and you don't, and I want to pass a law forcing everyone to eat cucumbers because they have health benefits ... it'll take you about ten milliseconds to tell me, emphatically, that you hate cucumbers and who am I, who loves them, to force my preferences on you?  But, for bicycle helmets, the issue never even came up. 


As for the costs ... on almost any scale, the risk of helmetless cycling is small.  Here's an article that claims a helmet can save around 30 percent of cycling deaths.  (It's not clear from the article, but I believe that's comparing 0% helmets to 100% helmets.) 

Is 30 percent big?  No way.  30 percent is nothing.  I'm pretty sure that riding at night increases risk by at least 30 percent.  I'd bet you that riding in rush hour traffic increases risk by at least 30 percent.  I'd bet you that riding fast increases risk by at least 30 percent.  In fact, I'd bet riding on the roads, instead of bicycle paths, increases risk by at least double. 

Do you want to ban those, too?  If you don't, then, since the cost higher, you must believe the benefit is higher, too.  Fine.  But it's different for different people.  I don't like riding at night, and I seldom do it.  Banning night riding costs me almost nothing.  Banning helmetless riding costs me a lot. 

If you want to allow night riding but not helmetless riding, logic forces you to admit that you are weighing someone else's benefit more than mine.  You're not tallying the actual cost to me, but, rather what you think the cost to me *should* be.  I *should* be willing to ride at night, because, that's worth the risk!  But not riding without a helmet, because that's not!

In fact, it's the very definition of unfair: banning something for no reason except that we believe you shouldn't want to do it. 

Now, you could come back with this response: OK, we acknowledge that you hate wearing helmets, and the risk isn't that big.  But, Phil, you're just being obstinate.  Once you wore a helmet, you'd get used to it, and your cost would drop to zero!  Long term, the benefit way outweighs the cost.  The law is our way of forcing you to give it a try!

That's the most reasonable argument I can think of, and I was expecting to get that from my original post.  And, you know, it may be true, that once I started wearing a helmet, I'd get used to it. 

But I might not.  And even if I do, who's to say that the benefit even outweighs the temporary unpleasantness? 

More importantly, do you want to live by that principle for everything?  Are you willing to submit to it yourself?

"Candy is unhealthy, so we're going to ban it.  You may not want us to ban it, but you'll get used to it and be glad you're not eating it."  "McDonald's is icky, so we're going to ban it.  You might not like eating at nicer restaurants at first, because they cost more and they're slower, but eventually you'll get used to it and not mind the extra time and money."  "PBS political specials may be boring, but they broaden your mind and make the population better informed, so we'll fine you if you don't watch them.  Don't worry, eventually you'll actually start to enjoy them."

There are literally hundreds of things you do in your life that increase your risk of something bad happening, and many of those you'd eventually get used to not doing.  That doesn't mean it's right for me to ban them.


And what about just riding less?  A safety factor of 30 percent means that seven trips without a helmet is just as risky as ten trips with a helmet. 

Suppose your spouse or child is the one commuting without a helmet, and that concerns you.  You want him to wear a helmet, so you don't worry.  You know he really, really hates helmets, and he'd never get used to one, but you want him to wear one anyway.

He things about it, and finally he says,

"You're right, the risk of riding without a helmet is higher, and I know you're concerned about my safety.  But I hate helmets, so I'm going to reduce the danger another way.  Instead of commuting to work by bike five days a week, I'm going to commute to work only three days a week, and take the bus the other two days.  So I'll reduce my risk, not just the 30 percent you wanted, but a whole 40 percent!  That's even though I'd rather ride all five days.  See how much I love you?"

Would you be happy?  Remember: the overall risk is LOWER than if he commuted daily with a helmet. 

I'm betting you wouldn't be much happier at all.  Can you say why?


Here's another test.  How fast do you drive?  Faster than the speed limit? 

If you drive 10 km/h (6 mph) over the limit, then, according to an Australian study, you increase your chances of having an accident -- not by 30 percent, but by over 400 percent.  Now, that sounds too high.  Another estimate (on the same web page) suggests it's only 30 percent, which sounds too low.  

I bet the real number is somewhere in between ... but never mind.  The point is, speeding is riskier than riding without a helmet, relatively speaking.

And, yes, I know speeding is already illegal.  But it doesn't carry the same social stigma.  I have had friends berate me for not wearing a helmet.  If I berated them for driving 10 km/h over the limit, they'd think I was a creep.  They'd tell me to mind my own business, and they'd probably avoid me.  And they'd be right ... nagging someone for speeding is socially unacceptable. 

That's true even though other people have a vested interest in your highway speed, since you're putting the rest of us at risk, not just yourself.  And, even though that risk is higher than for riding a bike without a helmet. 

My friend feels like he's doing a good deed by nagging me ... but I'm a creep if I nag back. 

Clearly, it's not about the risk.  It's something specific about the helmet.  It's arbitrary.


More in a future post.

Labels: ,


At Saturday, November 10, 2012 1:17:00 PM, Blogger David Pinto said...

I wear a helmet riding because when I was first riding as an adult in Boston, I had three close calls with other bikers and/or cars that convinced me I would be safer wearing a helmet. I have also gone on sponsored rides that required wearing a helmet to participate. I don't think anyone should be required to wear a helmet, however. I'm not a great person to judge by, however, since I tend to be a bit libertarian.

I wish there were more people how thought like you, Phil. Sometimes it seems that all people do is try to pass laws that seem like good ideas, but cause us to lose some liberty. The large soda laws in NYC are another example of this. I'm almost to the point that anyone who says, "There should be a law," gets slapped with a large fish.

At Saturday, November 10, 2012 2:31:00 PM, Anonymous Anonymous said...

It seems like the decreased rate of survival in an accident from not wearing a helmet is being compared to whether risky behavior causes the accident itself. Wouldn't a more apt comparison be to compare to something like seat belt use? It seems like it is hard to compare accident prevention to the effectiveness of something meant to reduce the severity of injury if the accident actually occurs.

At Saturday, November 10, 2012 10:07:00 PM, Anonymous Alex said...

Two quick points - I think people responded to your first post with a 'work backwards from helmets' approach because that's what you asked them to do: you said to come up with a framework that should make me wear a helmet. Also, I think it's unfair to compare percentages for a decrease and an increase. Something that must hit zero, like deaths, can only decrease 100% from where it is. On the other hand, things are rarely bounded above, and things like accidents can potentially increase by thousands or more percent. Maybe a 400% increase in chance of accident is in some way larger than a 30% decrease in death, but simply comparing the two numbers is unfair.

On a different point, I thought I had a pretty reasonable framework that you skipped over: People who travel via street must have safety protection. Cars wouldn't need helmets because drivers/passengers are already protected with air bags and seat belts. Bike and motorcycle riders would need helmets and preferably protective garments (like leathers). The only flaw in the framework I can think of offhand is that buses would need significant safety upgrades, but that seems somewhat reasonable anyway. So, would you like to argue why street transportation shouldn't/couldn't require protection?

At Sunday, November 11, 2012 1:48:00 AM, Blogger Kyle - Driveline Baseball said...


"People who travel via street must have safety protection."

Says who? Just because people driving in cars are forced to wear seatbelts? People in cars are forced to wear seatbelts therefore people on bikes should be forced to wear helmets isn't a very logically valid argument; it is begging the question.

At Sunday, November 11, 2012 1:53:00 AM, Anonymous Anonymous said...

Ok, I missed the original discussion, but here's my framework.

Start with the basic American values: "All men are created, liberty and the pursuit of happiness."

Most of the difficult political questions arise when we are asked to trade one or more of these values in exchange for another.

In the case of bike helmets, we are being asked for a decrease in liberty and the pursuit of happiness in exchange for an increase in life.

It's virtually impossible to quantify many of these increases and decreases in our American values objectively. Well, yes we can quantify the increase in life from helmet laws, but we can't really quantify the decrease in liberty, except in as a sort of subjective bayesian inference. That's what makes it hard.

When should we make these tradeoffs? Well, that's why we have political disagreements. Different people, um, value these values differently. To generalize an example, left-wingers place a higher value on equality than right-wingers do. Right-wingers place a higher value on liberty than left-wingers do. Extremists on either side like their preferred value so much that they would not exchange it for anything under any circumstances.

I am not one of these extremists. I like all my American values somewhat equally, and can imagine some circumstances, or dare I say a framework, in which I'd be willing to exchange one value for another.

Because it's so subjective, my personal framework is to avoid any tradeoff that is roughly equal. We should only consider a tradeoff where we give up a small amount of one value (whatever that means) in exchange for a large amount of another (whatever that means).

Seat belt laws create (IMO) a minor decrease of liberty, in exchange for (IMO) a large increase in life. Bike helmet laws (IMO) are of a similar scale of magnitude, but probably yield a slightly greater decrease in liberty, and a slightly smaller increase in life. The NYC soda law is much bigger decrease in liberty in exchange a much smaller (if any) increase in life.

So I'd rather have seat belt laws than bike helmet laws. I'd rather have bike helmet laws than soda laws. Where would I draw the line where the tradeoff is worth it? Well, I can't quantify that, but my gut feeling is probably somewhere close to where bike helmet laws currently exist -- require them for kids, but not for adults.

At Sunday, November 11, 2012 2:28:00 AM, Anonymous Anonymous said...

I suppose I could clarify that equality, life, liberty and the pursuit of happiness are not necessarily the only relevant values. There is a whole list of others in the preamble to the Constitution. There's the phrase "God and Family". There's also the small of a woman's back, the hanging curve ball, high fiber, good scotch, opening your presents Christmas morning rather than Christmas Eve and long, slow, deep, soft, wet kisses that last three days.

At Sunday, November 11, 2012 11:19:00 AM, Anonymous dmick89 said...

I have a few things to add, but they aren't a framework so I apologize.

1. I don't like the government legislating what people can or cannot do so I would not not support a law demanding cyclists wear helmets.

2. A reduction in deaths of 30% is also improving the quality of life of the drivers who would have previously killed these cyclists.

3. We reduce the speed limits in residential neighborhoods to protect the civilians. We reduce it in school zones to protect the children. We reduce it in work zones to protect workers. Why have I never heard anyone suggest we reduce the speed limit in heavily biked areas to protect those on bikes? A 5mph reduction in speed limit would have, I'd think, a significantly higher reduction in bicycle deaths than mandating helmets.

At Sunday, November 11, 2012 11:47:00 AM, Anonymous Alex said...

Kyle - Phil's original post asked for readers to create a framework that the reader would be willing to follow and requires bike riders to wear a helmet. That was the framework I made up. If the higher-level concern is reducing injuries or lowering health care costs, it also seems perfectly reasonable.

At Sunday, November 11, 2012 6:46:00 PM, Anonymous Anonymous said...

Here’s my attempt:

Helmet-less bike riders place an undue burden of concern on motorists due to the dramatic power differential between a bike and a car. A slight careless move by a motorist can easily result in severe injury or death for a bike-rider. Yes, free born citizens have the right to risk their own necks but you don’t have the right to make me a participant—an unwilling participant due to a minor transgression on my part.

On the road, the penalties of poor driving tend to be symmetrical to the offense. For example, if you swerve too fast, you can dent your fender and if you drive at 90 mph, you risk serious injury. The hemlet-less rider violates this naturally governing principle.

At Monday, November 12, 2012 1:02:00 PM, Anonymous StLhawk said...

Kenarson is basically where I'm at.

But honestly, the framework is really simple.
1) I want to live in a society that subsidizes health care for everyone, since it's a significant collective good and falls victim to all sorts of distorted incentives (it's really hard to shop around). 2) One consequence of this is society as a whole has some investment in health care outcomes, since your bad decisions cost me money.
3) There is significant evidence that helmets reduce risk of death and traumatic brain injury while bicycling

3) thus helmets should be mandatory, particularly for children.
4) Other safety regulations like seatbelt laws and traffic lights are appropriate presuming proper data justifying their enactment.

At Thursday, November 15, 2012 6:37:00 PM, Blogger Swoods said...

The Hand Formula is the framework.

Let B = cost of your having to wear a helmet plus, theoretically, the costs anybody else bears from your wearing it, like say someone close to you who's annoyed by the fact that you have to wear it too. If you consider that latter part to be negligible, this is how much you would pay per ride to not wear your helmet if The State (or some authority) made you wear a helmet without paying.

Let P = the probability of your having an accident that kills you.

Let P'= the probability that you'll die in this accident if you're wearing a helmet (if a helmet "increases safety" by 30%, then P'=1/(1+.3)=.769, right?)

Let L = the cost to all of society that the loss of your life would impose

If B < P'L, then it is utilitarian public policy for us to force you to wear a helmet.

A select term life insurance policy at provides $250K in coverage, so let that be L.

Suppose B = $10. Solving, P needs to be less than .0000676 (14792 to 1) for the cost of your wearing the helmet to be less than the benefit.

For B = $20, P < .0001352 (7395 to 1)

For B = $50, P < .000338 (2958)

Obviously these numbers are pretty arbitrary. Maybe your life is worth much more than $250,000, but that doesn't help your argument. I doubt your burden is greater than $50. What you'd need to complete the analysis is the frequency of deadly motorcycle crashes. In the absence of that information (or in the general case of insufficient information), you presume liberty because it's the best default position. For now, there's no reason for us to make you wear a helmet.

At Saturday, November 17, 2012 7:15:00 AM, Blogger Scott Segrin said...

Excellent article.

I would add to it by saying that I've always found it incredibly hypocritical that society says that we must wear helmets when we ride bicycles, but then turns around and claims that bicycles ought to be able to share the same roads, paths, and every other conceivable right of way with motor vehicles.

Riding a bike in the traffic lane of a hilly two-lane highway with a 45 mph speed limit seems to far more increase the risk of injury - to both the bicycle rider and the vehicle driver - than if the bicycle rider weren't wearing a helmet. But every time I mention this to a bike rider, I am scorned. I get the "we have a right to be there too and you better just accept it, abide by it, and shut up" treatment.

At Tuesday, March 08, 2016 6:00:00 PM, Anonymous Daniel B said...

"Another estimate (on the same web page) suggests it's only 30 percent, which sounds too low."

Sounds high to me. Speed, in a vacuum, isn't a problem. Driving fast in the fast lane, on a straight road, in good conditions, not swerving nor tailgating, isn't dangerous (yet it's over-enforced drastically compared to almost every other type of driving violation out there). But the way reports work, any accident in which the person was over the limit counts as a speed-caused wreck.


Post a Comment

<< Home