[maxerickson]

And yet there is a private body that certifies helmets as meeting their private standards:

http://www.smf.org/cert

ETA: the article you link mentions ASTM maintaining a more stringent standard!

It also has this gem, given our discussion:

There may never be an improved government standard for bicycle helmets. Experts may never come to a consensus on a standard for testing the forces most closely associated with concussions. But one test can be administered now: the market test. After all, new technology costs more. "Adding that upcharge to a $50 helmet," Scott Sports designer John Thompson told me, "is a harder sell."

[msandford]

I might suggest that this is the "good Samaritan" principle at work. There is a standard that the government has made, and which it enforces. The government is supposed to do the right thing (at least so people tend to believe, all the evidence to the contrary notwithstanding) and so the existence of SOME governmental standard (even if it isn't great) really screws up attempts to improve upon it.

If there was no definitive standard and instead three competing standards bodies then people would have to get informed a bit and decide what to do. But there's a government standard, so anything that passes it must be OK and everything else is superfluous. I might not think that way when it comes to buying helmets, but how many moms or dads buying their child a helmet might think exactly that?

Once the complexity is hidden behind a good/bad (it passes the test or it doesn't) metric any kind of critical thinking tends to go out the window. The CSPC test is good in that it prevents people from making helmets that would fall apart if you dropped them accidentally.

But it's not good in that there are inherent tradeoffs between foam strength and shock absorption and it has in effect arbitrarily decided that one particular point on that line is the RIGHT tradeoff full stop.

I don't have the stats for lower versus higher energy crashes. But it seems likely to me (based on 30 years and at least 50,000 miles of cycling) that lower energy outweighs higher energy by perhaps some 100:1. And if the distribution is that skewed, then perhaps you could do more good by fixing the low energy crashes even if it costs on the high energy crashes.

But ultimately I'd prefer for people to have that choice by themselves. Right now you can buy high energy helmets and that's about it. There might be SOME that can also protect against lower energy crashes. But there aren't any low energy only helmets for sale ANYWHERE. I might prefer one of those for just tooling around town at speeds of less than 15mph. Thanks to the CSPC if I want one of those I have to make it myself. While I could probably do that it'd cost me a year and $20k and I wouldn't be able to recoup that investment by selling, because again that helmet wouldn't pass the mandatory CSPC muster.

I think the situation stinks, and telling me that it is actually good doesn't make it so.

[maxerickson]

Nowhere did I say the situation was good. I pointed out that manufacturers are free to develop their own standards and provided evidence to that effect.

Having skimmed that article, I end up expecting the CSPC to move sometime in the near future, as the science around concussions is moving forward enough to make it possible to actually arrive at a better set of tradeoffs.

[msandford]

> I pointed out that manufacturers are free to develop their own standards and provided evidence to that effect.

They can only develop their own standards in excess of what the CSPC already requires of them. And their standard already sets the bar very, very high. Too high to be useful to most people.

So while that statement isn't blatantly and grossly wrong, I would argue that it's not really correct either. A more honest statement might be "manufacturers are free to develop their own standards (so long as they also meet CSPC minimums) and provided evidence to that effect"

A large portion of the point of this debate is "are the CSPC minimums a good tradeoff" and so pointing out that people can do whatever they want in excess of the CSPC minimums isn't exactly great form.

[maxerickson]

You started here:

You've assumed constant foam density which I don't believe is a given. That's how all helmets are made right now, but nothing would prevent someone from making two separate pieces and bonding them together, except cost. And what's the incentive to increase cost (and thus price) unless you can use that to differentiate somehow?

I don't see how my beating the point (to death) that manufacturers can go ahead and do that is bad form. If you want it, go ahead and take the last word.

[msandford]

Yes manufacturers can do it. I agree 100%. The point is that they aren't able to get anything for it! The CSPC test has FALSELY COMMODITIZED bicycle helmets where all helmets are SEEN as having the same level of safety regardless of the truth. Which then makes it very tough to sell people on extra safety, because all helmets are supposed to be AT LEAST SAFE ENOUGH thanks to the test.

I'm not trying to get the last word here, I'm trying to actually convince you of something. I know I'm doing something wrong, but I can't tell what. To me it's incredibly obvious that this particular standard is causing undesirable outcomes in society. I'm doing my damnedest to try and convince you of that. I'm having no success whatsoever. Intuitively it must be my fault; only I can tell if you're getting the gist of what I'm trying to communicate. It would be folly for me to blame you for not understanding my bad explanation.

1. Helmets only help you when they crush or break or deform. If they don't do this, they haven't really protected you.

2. Helmets are very strong, this is so that they can survive a very bad, high energy accident. This is required as per CSPC.

3. EPS doesn't start to crush until you've exceeded its compressive strength. In helmets, this is dictated by the maximal impact that it has to withstand.

4. Lower than maximal impacts will cause very little of the helmet to get used, or perhaps none at all. That means that the helmet did little or nothing to keep the person safe.

5. Crash severity is probably distributed as 1/x or as Poission meaning that low energy events are far, far more frequent than high energy events. Probably by several orders of magnitude.

7. The false commoditization of helmets (due to the CSPC test) makes it very difficult to convince people that there is anything else even worth looking at for safety (since everything is definitely safe enough), and even if you can do so you still have to design your standard to be in excess of the CSPC standard.

8. This leads a great many people to suffer concussions where a better designed helmet might have prevented them, while still providing an excellent level of safety for truly severe events. It is entirely possible to make a helmet which wouldn't pass the CSPC test that can provide good protection against concussion and death at the same time.

[maxerickson]

I think the problem is that you are imputing a position I have not taken.

I'm not especially convinced that the density of foam used in current helmets is egregiously bad, but that's a little different than being opposed to regulators and industry taking steps to make better helmets (it perhaps makes me a bit blase on the issue, and I anyway tend to come off as irreverent, whether I intend to or not).

The conversation also shifted a bit here: https://news.ycombinator.com/item?id=9720536

and I didn't really come along for the ride. I see how dismissing the idea of CSPC fines might come across as argumentative, but I really don't think that is a risk (especially in the face of extant private standards), and it is a different issue than the good Samaritan and false commoditization statements you have followed up with.