The CycleChat Helmet Debate Thread

[martint235]

At least gin is useful

 

[jefmcg]

Cycle Helmets have just made it back into the basket of goods used by the ONS to measure inflation. So has gin.

They go well together. I know more people that got head injuries following gin than involving cycling.

 

[RoubaixCube]

Just a random question for those of us that actually do wear helmets...

Do any of you own more than one helmet and swap them around depending on your mood or the bike youre riding? Maybe you want to match the colors up with your bike etc etc.

Just curious as i bought a Kask Mojito in White/Black that ive worn for quite a few rides and i have to say that im so impressed with it im actually considering getting another in a different colour just to use on the commute or on days when theres going to be a chance of rain.

Generally speaking I know the whole concept sounds a little vain. but there are people in the world with more than 20 pairs of shoes

 

[martint235]

Just a random question for those of us that actually do wear helmets...

Do any of you own more than one helmet and swap them around depending on your mood or the bike youre riding? Maybe you want to match the colors up with your bike etc etc.

Just curious as i bought a Kask Mojito in White/Black that ive worn for quite a few rides and i have to say that im so impressed with it im actually considering getting another in a different colour just to use on the commute or on days when theres going to be a chance of rain.

Generally speaking I know the whole concept sounds a little vain. but there are people in the world with more than 20 pairs of shoes

As a non helmet wearer, I'd say that's actually a good reason. I may disagree with the basic premise but if you feel you must wear one then matching them to bikes or moods etc seems laudable

 

[Big Andy]

Just a random question for those of us that actually do wear helmets...

Do any of you own more than one helmet and swap them around depending on your mood or the bike youre riding? Maybe you want to match the colors up with your bike etc etc.

Just curious as i bought a Kask Mojito in White/Black that ive worn for quite a few rides and i have to say that im so impressed with it im actually considering getting another in a different colour just to use on the commute or on days when theres going to be a chance of rain.

Generally speaking I know the whole concept sounds a little vain. but there are people in the world with more than 20 pairs of shoes

No just the one for me, but then both my bikes are the same colour.

 

[mjr]

Just a random question for those of us that actually do wear helmets...

Do any of you own more than one helmet and swap them around depending on your mood or the bike youre riding? Maybe you want to match the colors up with your bike etc etc.

When I did use one, I only had one. The limited 3-5 year lifespan made it seem too wasteful to buy more. The colour choice was more coordinated with my usual clothing (dark blue or black) than the bikes anyway.

That said, I only had one bike at a time back then and rode less, partly because the stupid headgear made it uncomfortable and less convenient

 

[srw]

Peter Walker wades in...
https://www.theguardian.com/lifeandstyle/2017/mar/21/bike-helmet-cyclists-safe-urban-warfare-wheels

Walker used to be the Guardian's staff cycling writer. He's always presented as a bit of an evangelist for helmets, hiviz, bike lanes and portraying city cycling as scary, dangerous and unpleasant. But this article suggests he's done some reading. Nothing will be at all surprising or particularly controversial to anyone here, even if he's still a bit credulous about the individual benefit angle and hasn't addressed the denominator neglect point I bang on about.

The BTL comments are predictably dumb.

 

[mjr]

Transferring from another thread:

Would be interested to know if there was any punishment. But is it possible to know that it hasn't prevented further deaths? I think it's hard to know this given it's hypothetical?

That's about the UCI helmet rule, made more-or-less compulsory in 2003 (at least for senior men - I'm not sure if it came in at different times for women and U-23, so I'm not analysing them). It's hard to know for various reasons (lots of confounding factors, small numbers, and so on), but it doesn't look obvious that it's helped:

Deaths at senior men's pro races in the 14 years pre-compulsion, 1989-2003: 4 (1995 Tour de France, 1999 Volta a Catalunya, 2000 Tour of Argentina, 2003 Paris Nice)

Deaths at senior men's pro races in the 14 years post-compulsion, 2003-2017: 7 (2005 Subida al Naranco*, 2006 Six Days of Ghent, 2008 Classica de Amarante*, 2010 Giro del Friuli Venezia Giulia, 2011 Giro d'Italia, 2016 Gent-Wevelgem, 2016 Criterium International*).

I compiled the above from various lists already mentioned in this discussion, including http://www.velonews.com/2011/05/news/road/noted-fatalities-in-professional-cycling_172004 and https://en.wikipedia.org/wiki/List_...related_death#Cyclists_who_died_due_to_a_race - you can argue that the ones marked with * were not primarily crashes, but it's arguable that insulating one's head may provoke other health problems.

Searching this subforum for keywords like UCI and deaths will find plenty of old discussion, but the list mentioned in https://www.cyclechat.net/threads/pro-cyclists-wearing-helmets.82649/ seems to be missing so I figured it's worth resummarising. There seems to be surprisingly little formal analysis online about the effects of the UCI helmet rule, or maybe I'm just stinking at searching this lunchtime!

 

[ufkacbln]

Realistically the Peleton is the best arbiter for helmet effectiveness

Hundreds of riders, riding in the most dangerous conditions and taking the most risks

Massive crashes and individual accidents recorded in high quality by TV. Networks

Combine that with the fact that they are using the most up to date helmets

Surely the data pre and post compulsion would prove the efficacy of helmets... if they were as effective as claimed

 

[benb]

The numbers are way too small to draw any conclusions.

 

[mjr]

The numbers are way too small to draw any conclusions.

Only if you insist on a particular significance level. There's various calculations we could do, but the conclusions will be rather weak. There might be enough incidents to work with if considering all head impacts, but it's difficult enough to verify even the list of deaths, so I suspect checking 150ish days of racing a year for crashes may take quite a bit of work. Has anyone seen it done?

 

[srw]

@srw is the sample big enough to show anything worthwhile?

The short answer is no.

 

[srw]

@srw is the sample big enough to show anything worthwhile?

The long answer is that you can draw some conclusions from the data, with a lot of caveats.

The first is that pro-racing cycling is a lot riskier than what the rest of us do. I haven't checked @mjr's data, but 150 days of UCI world tour cycling looks about right from their website. Let's say an average peloton of 150 riders, for 22,500 racing days per year. Fourteen years gives you 315,000 racing days, for a death risk of about 1 in 30,000 per day's racing. Compare that with the roughly 1 in 10,000,000 journeys in my most-often-cited paper, and even allowing for the increased journey time when you're racing over 150km or more compared with pootling through London the pros expose themselves to a significantly greater risk of death. Even though they're riding over largely closed roads, with police escorts and supposedly predictable motor traffic.

So drawing a conclusion from pro racing for the rest of us is always going to be iffy.

Now I'm beginning to get into harder maths at which I'm rather rusty, so someone might need to correct me. On the face of it, the death risk on any day pre-helmets was about 1 in 40,000; post-helmets it's (roughly) twice as much.

The appropriate model to use is probably the Poisson distribution, with a mean of 0.3 deaths per year. If I've looked up the right statistic, the standard deviation of the Poisson distribution is the square root of the mean, or 0.5 deaths per year. So the post-helmet observed mean of 0.6 deaths per year is less than even one standard deviation from the pre-helmet observed mean, and you'd have to make some really heroic assumptions to deduce any difference.

That's even before thinking about normalising for different numbers of race days, different peloton sizes, different racing conditions (more motorbikes and commissaire cars), different race lengths, and so on and so forth.

So it's certainly not possible to say that helmets have improved death risk. It's also not possible to say that helmets have worsened death risk. And death isn't the only risk to which pro bike riders are exposed.

(@mjr - I did the same search as you a while ago, and equally found nothing. Which, at one level, is odd. Here is a neat dataset for which exposure data and incident data are publicly available, and where there is a definitive intervention point.)

 

[growingvegetables]

Sorry! An irreverent and disrespectful like. Indicates a profound "Whoosh!", but loving the Douglas Adams reference in the relevance of the fish distribution.

 

[srw]

(@mjr - I did the same search as you a while ago, and equally found nothing. Which, at one level, is odd. Here is a neat dataset for which exposure data and incident data are publicly available, and where there is a definitive intervention point.)

One reason for that might be publication bias. Someone's done the same sort of back-of-the-fag-packet calculations as I've just done, and realised that however hard they work the data nothing will come of it, so it will be a "boring" paper (from the perspective of a publisher).