Cycling UK Forum

axel_knutt wrote:You don't need any conscious decisions to salivate when you see food.

Ah - You've not seem my cooking!

EdinburghFixed wrote:...Maybe this weights the average, and we should look at the median instead (for example?)

I'm not a statistician either

...but no we should'nt look at the median. The median is a theoretical vehicle that gives us plenty of room. So is the average. We are interested in the small number of vehicles which come too close. This small number of vehicles is related, by way of standard deviation, to the mean. I'd guess that a mathematician will now give us the relationship of that small number of vehicles to the median, but I'm convinced my probability curves have average!

Just to re-iterate Dr. Watson's findings: Across the various road types, times of day and city locations drivers, on avarage, passed closer to cyclists with helmets. This means that the potential for accidents is greater for hemeted cyclists.

Edinburghfixed, I'm in complete agreement with you on this. It is a startling fact - not least because I guess most drivers don't even recognise they have passed a cyclist let alone noted them to be wearing a helmet.

Now I've looked even closer at Dr Watson's data! - I am meant to be working...

The Salisbury data shows 1.55m average for cyclists with helmets and 1.61m for cyclists without. That is 3.8% closer if you wear a helmet.

The Bristol data shows 1.48m average for cyclists with helmets and 1.58m for cyclists without. That is 6.3% closer if you wear a helmet.

Perhaps not surprising, there is a difference between the cities - for cyclists without helmets 1.5% closer in Bristol and 4.0% closer for helmeted cyclists.

So Bristol drivers are passing closer and are even closer if you wear a helmet.

After my previous post I'd looked at Watson's data a bit more:

The measurements were apparently taken in Salisbury, Bristol and Portsmouth, but the data set only records Salisbury and Bristol - I'm guessing that the third set of data should be Portsmouth and is incorrectly labelled.

He seems to have tried a good variety of road types and times of the day.

Still no Big city data or rural roads.

I'd also be interested to know how fast he was going...

It's worth checking his blog http://drianwalker.com/overtaking/ quite entertaining and a link to the raw data set! The man seems to know his stuff - just the kind of guy to do further research, assuming he could be suitably funded.

No prizes for guessing where I think at least some of that funding could come from...

Kirst wrote:What many people appear to be saying is "you have the right to choose whether to wear a helmet or not but if you do wear one you're making it more likely that I'll have to wear one and so I want you to choose not to because I don't want to wear one either" and that does not seem to be respecting their right to choose what they feel is right for them.

To be fair, going along with the right to choose is the right of others to ridicule you!

I wear a helmet because my wife says I should! There - I'm open to ridicule too.

My only input into the choice of helmet was to get a blue one just so it doesn't match my bikes and to make sure it fitted properly.

EdinburghFixed wrote:I find myself giving bare-headed cyclists more room, although I'm not sure exactly why. Perhaps they seem more fragile?

My girlfriend seems to do the same - unfortunately she's the only other driver I can observe regularly so that's as far as my anecdotes can go!

Really? - So do you not give helmeted cyclists enough room or do you give those without too much? It's a naughty question I know, but one or the other must be true. There must also be dangers associated with giving too much room...

While I would not wish to "pigeonhole" you, I had you down as anti-helmet, yet you have just expressed the view that sub-conciously you think of un-helmeted riders as more vulnerable.

I've not read Watson's experiments deeply - only the conclusions. By he lives in Salisbury and works in Bath so I'd guess most of the study was conducted in those areas. Both are small provincial cities, so I'd imagine low speeds and conjested roads. Such environments are probably the type of riding where low speed impacts are prevalent - that is - where the use of a helmet might be most beneficial. So we could draw the further conclusion that driver attitude (sub-concious or otherwise) acts against cyclists in such a way that collisions are more likely in circumstances a helmet may be of benefit.

Does anyone know if Watson's research is being continued? Having built a bike with the relevant equipment would there not be benefit in trying it in different locations? Not wishing to be London-centric but the must be an opportunity to gather vast amounts of data very quickly. Whenever I have to go to London, I'm always surprised at the number of bikes. Passing gaps seem to be very small, but then I've never ridden there...

CJ wrote:Interesting it isn't, predictable it is.

You are of course right! I didn't spell it out, but I agree entirely with your comment. Sorry if I gave the contrary impression.

CJ wrote:Motorists might not consciously make a note of whether you're wearing a helmet, but it'll be part of the general impression they form of you as they approach. At the closing speed you infer, they have only a few seconds to decide whether to wait behind or squeeze by.

My thoughts are that in the dark, on a rural A road, the motorist does not have ANY time to subconciously determine if I'm wearing a helmet. I'm also a driver on the same road in similar conditions. I've conciously looked to see if the few cyclists are wearing helmets as I've passed (giving plenty of room!), and usually its not possible to tell.

The vast majority of vehicles do give me plenty of room. Even those which come close (and some hurling abuse) seem to give much more than 1m. At 50-60mph closing speed (60mph rural trunk road with actual vehicle speeds up to 80mph and my average typically about 15mph) I'm grateful for that.

EdinburghFixed wrote:It still comes back to asking, what research would you expect this hypothetical organisation to conduct / sponsor?

Suppose you start with the observation that in every country around the world, an increase in helmet wearing has not been observed to reduce the rate of cyclist head injuries. The mandate of the organisation might be to support research that explains this mystery.

I don't know what research should be conducted. I think it's clear that the research to date is either dismissed as flawed (and a lot of it does seem to be) or it's considered to be not relevant to the UK... I'd think there was scope for a cycling organisation to be able to be neutral in that I't should be able to argue "in the interests of its membership".

I think the second quoted paradox is at the heart of the problem. If helmets had anything more than a trivial effect this should show up in the nature of the injuries. Please note that I am not commenting on numbers here - just types of injuries. If helmets made matters worse that too should be evident.

There seems to be good evidence that various other factors (risk compenstation of both cyclists and motorists being significant) influence accident rates.

Now speaking only for me! - I don't believe that motorists know if I'm wearing a helmet - I cycle a lot in the dark. The biggest perceived risk is for me to be rear ended by a motorist not paying attention. With a potential differential speed of 50-60mph a helmet would not be much use. The alternative accident would be a side swipe with me ending up either on the verge or in the roadway. It is for those potential incidents that I'd like ot know how the helmet changes my chances.

I've not back and checked but I think it was Si who posted earlier on the relative unimportance of helmets in the whole debate on safe cycling. That is a view I'm comfortable with, but helmets seem to be one thing we do (or don't do) as riders which is clear and unambiguous. Yet we seem to choose on the basis of personal choice which we back up by the mistaken and unjustified belief that we are properly informed through internet searches.

I suspect we all know that the government cannot be relied on to make this kind of decision on our behalf. Us being cyclists actually concerned about possible helmet legislation - a minority of a small minority. It is intersting to note those countries with compulsory helmet legislation have not changed their policies on publication of the research showing that this has not reduced accident rates.

It may be of no surprise to anyone that in the last seven years I've bought from cycle shops - three cycle helmets, probably two tyres, one full set of cables, two chains and NO NEW BIKES. Granted, I'm not a high milage cyclist but in the future I'd expect to buy another helmet as the current one will go through its poorly defined life span before I will need another chain...

360fix wrote:15,000 person-years of cycling means one person cycling constantly for 15,000 years before meeting with a fatal accident.

Or to put it another way it is 15,000 cyclists all cycling constantly for one year without a single one of them having a fatal accident.

Sorry, that is not true. The link says there are 3 to 5 deaths a year (due to cycling?) of the 60,000 club members. There is no correlation given between the amount of cycling and the chance of an accident. The 15,000 number is simply 60,000 members divided by four deaths.

Lets say on average each club member cycles 100 miles a week. That is 6,000,000 miles for four fatalities or 1,500,000 miles per death. That is still a rather meaningless number if you are trying to predict your own demise!

pjclinch wrote:But ultimately the bottom line is if you review the available literature what you'll find is that you won't find truly copper-bottomed evidence one way or another, and the nature of the thing is more the reason than we're just not trying hard enough

I think that is what prompted me to start this thread...

I thought it would be beneficial if an organisation like CTC took a more definite position in the debate. We have had lots of comment on why they should not. As I am not a member I have to respect that position.

However - I don't accept that further research would be pointless and would still be inclined to support (in my very small way) an organisation that set out to provide better evidence. Of course the research would need to be conducted to very rigourous standards and would need to be able to demonstrate impartiality.

I don't know what form such research would need to take, but there are people much cleverer than me who might. I really don't accept that the problem is too difficult for further investigation.

So the big question is - Do we want all these government employees geting fit?

kwackers wrote:A walk for example is often cited as 'good exercise', by my reckoning that's only true if you're pretty unfit for me it barely registers an increase in HR.

If I recall - It's a brisk walk. The idea (I guess) is that you define brisk by how out of puff you get.

There is a ratio between effort expended walking to cycling. You probably wouldn't need to walk (briskly) for very far before you'd be doing more benefit than cycling 2.5 miles

Thanks Edinburgh fixed.

-And of course 2.5 miles assumes only five daily commutes. Once they are used to it, they'll be cycling to McDonalds and KFC...

Agreed 2.5 miles each way is very achievable. I was actually surprised it was that low. There was no reference that I saw in the article as to where that figure came from - any ideas?

Tony,

Thanks for that link. I found it very interesting.

Mr de Jong has a mathematical model equating health benefits for cycling to cost of head injuries. He has set it up so that the relative costs can be varied. He shows that helmet law costs money if there is decrease in cycling as a result - His results are disputed!

I'd love for this study to help move the debate - maybe it will. What it does not do however is contribute any new fact. I think that what is (as you might have guessed) missing is the real evidence on which to base this kind of study.

I'm guessing now but I believe that Mr de Jong has assigned a positive benefit to wearing a helmet in the event of an accident - How? What benefit?

Of course - if helmet wearing actually made the consequences of accidents worse, and if compulsion reduces the number of cyclists, it could be shown that compulsion saved money by stopping people cycling...

There were two nice bits of data in the article: Mr. de Jong has set the benefit of cycling at $1 per km... UK Government emplyees need to cycle 40km a week to reduce heart disease rates. - Presumably they will still be able to drive their cars to the continent in order to clock up those kilometres...