Support road.cc

Like this site? Help us to make it better.

news

Dad stops kid from crashing bike into parked car (+ link to video)

Footage goes viral - after soparking helmet debate

A video of a father dashing after his son to prevent him from crashing his bike into a parked car has been grabbing a l;ot of attention on Reddit - but not for the reason you might think.

 The footage, which you can watch here,  shows the father steadying his son's bike on a quiet suburban street before giving him a little push to help him on his way.

The father is jogging alongside his son as the youngster makes his first pedal strokes - then suddenly sprints into action as the nipper veers towards a parked car.

For many commenting on the video on Reddit, however, the quick-thinking father's prompt action to prevent a crash wasn't the most striking thing about the video, with the first commenter observing, "That kid needs a helmet" - an opinion that inevitably has sparked a debate on the subject.

Simon joined road.cc as news editor in 2009 and is now the site’s community editor, acting as a link between the team producing the content and our readers. A law and languages graduate, published translator and former retail analyst, he has reported on issues as diverse as cycling-related court cases, anti-doping investigations, the latest developments in the bike industry and the sport’s biggest races. Now back in London full-time after 15 years living in Oxford and Cambridge, he loves cycling along the Thames but misses having his former riding buddy, Elodie the miniature schnauzer, in the basket in front of him.

Add new comment

422 comments

Avatar
Bluebug replied to Rich_cb | 6 years ago
4 likes

Rich_cb wrote:
Bluebug wrote:

You are aware that you can do free online statistics courses?

https://www.coursera.org/courses?languages=en&query=statistics

https://www.khanacademy.org/math/statistics-probability

(Yes learning about probability  is relevant.)

It's probably worth you doing one.  

That way then you can understand why other posters are pulling your posts apart.

 

Good old deflection. Always there when you haven't got an argument to make.

I was trying to be polite instead of pointing out  like the other posters none of the data you have used supports your theories, and your statistical analysis of the data to draw your conclusions is terrible and is related to no modern scientific method taught at GCSE, A level, undergraduate degree level and post graduate degree level at Russell group universities in England and Wales.

Added to that some insurance companies do both basic and refresher courses in statistics for all workers on-site just to avoid people coming out with rubbish in front of customers.

Avatar
Rich_cb replied to Bluebug | 6 years ago
0 likes
Bluebug wrote:

I was trying to be polite instead of pointing out  like the other posters none of the data you have used supports your theories, and your statistical analysis of the data to draw your conclusions is terrible and is related to no modern scientific method taught at GCSE, A level, undergraduate degree level and post graduate degree level at Russell group universities in England and Wales.

Added to that some insurance companies do both basic and refresher courses in statistics for all workers on-site just to avoid people coming out with rubbish in front of customers.

So point out the actual flaws in my posts.

I'll save you some time.

At no point have I said that correlation proves causation.

Off you go.

Avatar
davel replied to Rich_cb | 6 years ago
3 likes
Rich_cb wrote:
Bluebug wrote:

You are aware that you can do free online statistics courses?

https://www.coursera.org/courses?languages=en&query=statistics

https://www.khanacademy.org/math/statistics-probability

(Yes learning about probability  is relevant.)

It's probably worth you doing one.  

That way then you can understand why other posters are pulling your posts apart.

 

Good old deflection.

Always there when you haven't got an argument to make.

Is denial and avoidance worse?

Denial as in "I never debated insurance Ts&Cs" 5 months after you did exactly that?

Avoidance as in ignoring the proof (you might want to Google-Fu the shit out of that word -
it's PROOF) that, actually, you did debate exactly that on this very site 5 months ago.

I think they're worse. Or evidence of insanity.

Avatar
Rich_cb replied to davel | 6 years ago
0 likes
davel wrote:

Is denial and avoidance worse?

Denial as in "I never debated insurance Ts&Cs" 5 months after you did exactly that?

Avoidance as in ignoring the proof (you might want to Google-Fu the shit out of that word -
it's PROOF) that, actually, you did debate exactly that on this very site 5 months ago.

I think they're worse. Or evidence of insanity.

I forgot a couple of posts that occurred 5 months ago.

I think it's far more worrying that you remembered them to be honest.

Avatar
davel replied to Rich_cb | 6 years ago
3 likes
Rich_cb wrote:
davel wrote:

Is denial and avoidance worse?

Denial as in "I never debated insurance Ts&Cs" 5 months after you did exactly that?

Avoidance as in ignoring the proof (you might want to Google-Fu the shit out of that word -
it's PROOF) that, actually, you did debate exactly that on this very site 5 months ago.

I think they're worse. Or evidence of insanity.

I forgot a couple of posts that occurred 5 months ago.

I think it's far more worrying that you remembered them to be honest.

Slightly more than just a couple of posts: it's the same sort of nonsense as exhibited on here and other threads.

So you've either got a dismal memory or are just full of shit.

Avatar
Rich_cb replied to davel | 6 years ago
0 likes
davel wrote:

Slightly more than just a couple of posts: it's the same sort of nonsense as exhibited on here and other threads.

So you've either got a dismal memory or are just full of shit.

I've just got better things going on in my life to distract me from silly disagreements on road.cc.

Nice that you remembered it though.

Avatar
davel replied to Rich_cb | 6 years ago
4 likes

Rich_cb wrote:
davel wrote:

Slightly more than just a couple of posts: it's the same sort of nonsense as exhibited on here and other threads. So you've either got a dismal memory or are just full of shit.

I've just got better things going on in my life to distract me from silly disagreements on road.cc. Nice that you remembered it though.

Yeah, your posts suggest a well-balanced life.

But I do also copy all your posts into a date-sorted file. So when I read about someone in a helmet and graph-covered Tron suit going postal while raging inaccurately about evidence and hypotheses, I can hand it over to plod. You know, to do my bit.

Avatar
Rich_cb replied to davel | 6 years ago
0 likes
davel wrote:

Yeah, your posts suggest a well-balanced life.

But I do also copy all your posts into a date-sorted file. So when I read about someone in a helmet and graph-covered Tron suit going postal while raging inaccurately about evidence and hypotheses, I can hand it over to plod. You know, to do my bit.

It's good that you have something to keep you occupied I suppose.

Avatar
CygnusX1 replied to Rich_cb | 6 years ago
4 likes
Rich_cb wrote:
davel wrote:

Slightly more than just a couple of posts: it's the same sort of nonsense as exhibited on here and other threads.

So you've either got a dismal memory or are just full of shit.

I've just got better things going on in my life to distract me from silly disagreements on road.cc.

Nice that you remembered it though.

I suggest you get on with the better things in your life then, instead of continuing this disagreement on road.cc

Avatar
ClubSmed replied to Rich_cb | 6 years ago
2 likes
Rich_cb wrote:
ClubSmed wrote:

The issue is that you have not gathered relevant data because you don't have corresponding data. This results in you having no patterns or data that support your hypothesis (though you are now claiming it as a theory rather than a hypothesis)

Look at the diagram I provided.

Observation: There is a correlation in the UK between increased helmet wearing and decreased cycling fatalities.

Question: Could cycling helmets have been responsible for a fall in cycling fatalities? How would they do this?

Hypothesis: Cycle Helmets reduce deaths (from head injuries).

Testable predictions:
1: Overall injury rate will not fall when helmet use increases as helmets do not prevent accidents.
2: Head injury rates will fall as helmet use increases.
3: Deaths from head injuries will fall.

Data Gathered
Prediction 1: Proved correct
Prediction 2: Proved correct in adults.
Prediction 3: No data available to prove or disprove.

So as you can see I have actually followed the method/process.

It seems pointless me trying to explain why and how you're wrong, as so many others as well as myself have tried to do in the course of this thread.
If you do not possess the intelligence needed to take on board feedback and adapt with it then so be it.
The next step in the scientific method would be to publish your findings. Why don't you try those reputed journals that you admire and see how many of them are willing to publish your paper.
Questions:
How many journals would accept Rich_CB's research methods, analysis and findings?
What will Rich_CB do when the paper is rejected by all?

Hypothesis:
Rich_CB will not have his work taken seriously but that will not change his outlook.

Testable predictions:
Rich_CB's 'study' will not be accepted by any journal (respected or not)
Rich_CB will not accept this as any fault of his data or analysis.

It's now over to you Rich_CB to gather the data, good luck! I look forward to seeing the results.

Avatar
Rich_cb replied to ClubSmed | 6 years ago
0 likes
ClubSmed wrote:

It seems pointless me trying to explain why and how you're wrong, as so many others as well as myself have tried to do in the course of this thread.
If you do not possess the intelligence needed to take on board feedback and adapt with it then so be it.
The next step in the scientific method would be to publish your findings. Why don't you try those reputed journals that you admire and see how many of them are willing to publish your paper.
Questions:
How many journals would accept Rich_CB's research methods, analysis and findings?
What will Rich_CB do when the paper is rejected by all?

Hypothesis:
Rich_CB will not have his work taken seriously but that will not change his outlook.

Testable predictions:
Rich_CB's 'study' will not be accepted by any journal (respected or not)
Rich_CB will not accept this as any fault of his data or analysis.

It's now over to you Rich_CB to gather the data, good luck! I look forward to seeing the results.

Good old deflection.

Always there when you don't have an actual argument.

Avatar
ClubSmed replied to Rich_cb | 6 years ago
4 likes
Rich_cb wrote:
ClubSmed wrote:

It seems pointless me trying to explain why and how you're wrong, as so many others as well as myself have tried to do in the course of this thread.
If you do not possess the intelligence needed to take on board feedback and adapt with it then so be it.
The next step in the scientific method would be to publish your findings. Why don't you try those reputed journals that you admire and see how many of them are willing to publish your paper.
Questions:
How many journals would accept Rich_CB's research methods, analysis and findings?
What will Rich_CB do when the paper is rejected by all?

Hypothesis:
Rich_CB will not have his work taken seriously but that will not change his outlook.

Testable predictions:
Rich_CB's 'study' will not be accepted by any journal (respected or not)
Rich_CB will not accept this as any fault of his data or analysis.

It's now over to you Rich_CB to gather the data, good luck! I look forward to seeing the results.

Good old deflection.

Always there when you don't have an actual argument.

No, I am just laid up in bed with a very high fever and find it hard to go over the same ground again and again.

Rich_cb wrote:

Hypothesis: Cycle Helmets reduce deaths (from head injuries).

Testable predictions:
1: Overall injury rate will not fall when helmet use increases as helmets do not prevent accidents.

Only if helmets are the single factor, have you evidence that helmets are the only factor effecting injury rates?

Rich_cb wrote:

2: Head injury rates will fall as helmet use increases.

This does not have to be true for the hypothesis to be correct. It depends on helmet effectiveness on different severity impacts, the number of those types of impacts and how the numbers are moved down the line.

Rich_cb wrote:

3: Deaths from head injuries will fall.

Yes, if the hypothesis is true and the helmets were worn correctly in the environments where the majority of fatalities happen then you would expect to see this impact.

Rich_cb wrote:

Data Gathered
Prediction 1: Proved correct

This can not be proved correct unless you prove that there were definitely no factors that should have increased the number of fatalities, like increasing car numbers for example.

Rich_cb wrote:

Prediction 2: Proved correct in adults.

I don't believe that you have shown the data for all head injuries, only for serious (hospital submission) head injuries. With the data I did see you provide, it was disproved due to the sub group of cyclists with the largest decline not having an increased helmet wearing trend. It was also questionable due to the also significant drop in pedestrian head injuries over the same period.

Rich_cb wrote:

Prediction 3: No data available to prove or disprove.

So out of your three testable predictions two were pointless and the other not done due to lack of data.
That's just my take though, and I'm ill. Send it to the journals and see what they say.

Avatar
Rich_cb replied to ClubSmed | 6 years ago
0 likes
ClubSmed wrote:

So out of your three testable predictions two were pointless and the other not done due to lack of data.
That's just my take though, and I'm ill. Send it to the journals and see what they say.

Look at the diagram.

See the box that says 'develop testable predictions'?

'If my hypothesis is correct then I expect a, b, c...'

That's what I've done.

The data is behaving as expected if helmets were reducing deaths from head injury.

They don't have to be the only factor for that to be true.

The comparison with pedestrians helps to eliminate the effect of shared factors.

So there is evidence for a cyclist specific factor that reduces head injuries.

There is also evidence that during the time said factor was observed there was no change in the overall injury rate.

That evidence is exactly what you would expect to find if helmets were reducing head injuries.

If you correct the paediatric head injury data for participation rate then the difference between pedestrians and cyclists disappears.

The difference in adults is maintained.

This is further evidence in favour of my hypothesis.

Avatar
ClubSmed replied to Rich_cb | 6 years ago
3 likes
Rich_cb wrote:
ClubSmed wrote:

So out of your three testable predictions two were pointless and the other not done due to lack of data.
That's just my take though, and I'm ill. Send it to the journals and see what they say.

Look at the diagram.

See the box that says 'develop testable predictions'?

'If my hypothesis is correct then I expect a, b, c...'

That's what I've done.

The data is behaving as expected if helmets were reducing deaths from head injury.

They don't have to be the only factor for that to be true.

The comparison with pedestrians helps to eliminate the effect of shared factors.

So there is evidence for a cyclist specific factor that reduces head injuries.

There is also evidence that during the time said factor was observed there was no change in the overall injury rate.

That evidence is exactly what you would expect to find if helmets were reducing head injuries.

If you correct the paediatric head injury data for participation rate then the difference between pedestrians and cyclists disappears.

The difference in adults is maintained.

This is further evidence in favour of my hypothesis.

This is exactly what I was talking about, you can't see the faults even when they are pointed out to you several times, by several different posters from differing helmet debate standpoints. Not one single person has come forward to support your workings out, many on this site believe in your hypothesis but no-one has agreed with your workings, does that not tell you anything?
I've done my best to show you the errors but as you won't listen, I am thinking that you are a lost cause. If you are so convinced by the accuracy of your process, data, analysis and conclusions then go try and get it published, as I previously suggested, and let us know what happens.

Avatar
Rich_cb replied to ClubSmed | 6 years ago
0 likes
ClubSmed wrote:

This is exactly what I was talking about, you can't see the faults even when they are pointed out to you several times, by several different posters from differing helmet debate standpoints. Not one single person has come forward to support your workings out, many on this site believe in your hypothesis but no-one has agreed with your workings, does that not tell you anything?
I've done my best to show you the errors but as you won't listen, I am thinking that you are a lost cause. If you are so convinced by the accuracy of your process, data, analysis and conclusions then go try and get it published, as I previously suggested, and let us know what happens.

The argument I've put across is pretty much the exact same one as proposed in the BMJ article I linked to.

Maybe you and all the other experts on this thread should write to the BMJ and alert them to their error?

Avatar
ClubSmed replied to Rich_cb | 6 years ago
6 likes

Rich_cb wrote:
ClubSmed wrote:

This is exactly what I was talking about, you can't see the faults even when they are pointed out to you several times, by several different posters from differing helmet debate standpoints. Not one single person has come forward to support your workings out, many on this site believe in your hypothesis but no-one has agreed with your workings, does that not tell you anything? I've done my best to show you the errors but as you won't listen, I am thinking that you are a lost cause. If you are so convinced by the accuracy of your process, data, analysis and conclusions then go try and get it published, as I previously suggested, and let us know what happens.

The argument I've put across is pretty much the exact same one as proposed in the BMJ article I linked to. Maybe you and all the other experts on this thread should write to the BMJ and alert them to their error?

We don't need to, the BMJ are already aware of the errors in the paper. Some of those I have already detailed, another key one is:

"The main conclusion of Cook and Sheikh, that a bicycle helmet prevents 60% of head injuries, is incorrect due to a fundamental error in the way they have treated their percentages. A correct analysis demonstrates unequivocally that there must be major confounding factors in their data set that they have failed to take into account, and therefore any estimate of helmet effectiveness is purely speculative.

Assuming that their basic analysis of the data is correct (although the numbers they quote in the text do not actually appear to match the figure plotted), they arrive at a figure of a 3.6% for the reduction in the head injury (HI) rate for cyclists, over and above the "background" reduction that pedestrians have also seen. They assume that this drop in HI is due to increased helmet-wearing. However, this reduction is presented in terms of the number of percentage points, and relative to the baseline value of 27.9% HI for cyclists in 1995-6 it actually represents a 3.6/27.9 = 13% drop in the HI rate.

The decrease in the number of helmetless cyclists over the same interval is 5.8 percentage points from a baseline of 84% unhelmeted, giving the percentage drop as 5.8/84 = 7%. Cook and Sheikh calculate helmet effectiveness to be given by the ratio 3.6/5.8 = 60%. However the correct expression to use is 13/7 = 186%. In other words, "helmet effectiveness" is so high that each helmet does not just save its wearer, but a non-wearer too. At this rate, head injuries would be eliminated completely if just a little over half of all cyclists wore them! This is clearly ludicrous.

A more reasonable conclusion to draw from this would be that there are some other factors that are responsible for the large drop in HI rate, and therefore any attempt to attribute some part of the total 30% (8.49/27.9) change to the provably marginal impact of a very small number of extra helmet wearers is at best highly speculative and fraught with inaccuracy."

The authors subsequently admitted to their mathematical error.

Avatar
Rich_cb replied to ClubSmed | 6 years ago
0 likes
ClubSmed wrote:

We don't need to, the BMJ are already aware of the errors in the paper. Some of those I have already detailed, another key one is:

"The main conclusion of Cook and Sheikh, that a bicycle helmet prevents 60% of head injuries, is incorrect due to a fundamental error in the way they have treated their percentages. A correct analysis demonstrates unequivocally that there must be major confounding factors in their data set that they have failed to take into account, and therefore any estimate of helmet effectiveness is purely speculative.

Assuming that their basic analysis of the data is correct (although the numbers they quote in the text do not actually appear to match the figure plotted), they arrive at a figure of a 3.6% for the reduction in the head injury (HI) rate for cyclists, over and above the "background" reduction that pedestrians have also seen. They assume that this drop in HI is due to increased helmet-wearing. However, this reduction is presented in terms of the number of percentage points, and relative to the baseline value of 27.9% HI for cyclists in 1995-6 it actually represents a 3.6/27.9 = 13% drop in the HI rate.

The decrease in the number of helmetless cyclists over the same interval is 5.8 percentage points from a baseline of 84% unhelmeted, giving the percentage drop as 5.8/84 = 7%. Cook and Sheikh calculate helmet effectiveness to be given by the ratio 3.6/5.8 = 60%. However the correct expression to use is 13/7 = 186%. In other words, "helmet effectiveness" is so high that each helmet does not just save its wearer, but a non-wearer too. At this rate, head injuries would be eliminated completely if just a little over half of all cyclists wore them! This is clearly ludicrous.

A more reasonable conclusion to draw from this would be that there are some other factors that are responsible for the large drop in HI rate, and therefore any attempt to attribute some part of the total 30% (8.49/27.9) change to the provably marginal impact of a very small number of extra helmet wearers is at best highly speculative and fraught with inaccuracy."

The authors subsequently admitted to their mathematical error.

That's an error based on the failure to correct for the decline in the rate of cycling amongst children which I've already mentioned.

It's not an error with calculating the difference between adult pedestrian and cyclist head injury rates which is what I referred to in my post detailing testing the hypothesis.

If you correct the data for participation rates you get a smaller overall benefit but one that actually mathematically makes sense.

It is also actually stronger evidence in favour of helmets having a benefit as once corrected you no longer see a larger benefit in the group that did not increase their wearing rate.

Avatar
davel replied to Rich_cb | 6 years ago
4 likes

Rich_cb wrote:

Here's a picture of the scientific process/method. As you can see I have collected relevant data from the literature and analysed it for patterns that would support my hypothesis. The patterns do support the hypothesis.

Where's the 'start with ideology and scratch around Google for shitty data' bit?

You don't test your hypothesis by stretching data that was compiled to do something different.

Your behaviour in this thread and others bears zero relationship to that diagram. Do you understand that? It isn't the process or that diagram that has the issue, it's you thinking that attempting to bend Google to fit your ideology is anything approaching scientific - that's the fucked-up bit. You've spent a lot of time on this: you could have done your own research by now.

"I'm pretty sure I've never debated Insurance T&C's with anyone."

Here's your warbling on insurance from all of 5 months ago: http://road.cc/content/news/226380-cycling-abroad-and-relying-nationwide...

You had a bang on the head or something?

Avatar
ClubSmed replied to Rich_cb | 6 years ago
1 like

Rich_cb wrote:

The TRL acknowledged that the data on childrens' helmet wearing was not as thorough as the data on adult wearing rates. In their later surveys they also looked at segregated leisure routes. They found higher rates of children cycling and far higher rates of children using helmets on those routes. If that data is representative then it would be very difficult to determine whether it was the infrastructure or the helmets which caused the decline in head injuries in children. That caveat does not apply to the adult data.

The TRL study states that:
"Both adults and children were more likely to wear a helmet when cycling on a cycle path than on the road or pavement"
Does this not point to the increase in cycling infrastructure and helmet wearing being inextricably interlinked as to make it impossible to determine whether just one is the cause of the injury reduction? Personaly I am surprised by this finding though as I would have thought that cycle paths are the least dangerous enviroment and therefor the has the least need for the wearing of a helment. Are cycle paths of a recreational route included in KSI data or is it just roads? If not then that would skew the data sets quite a lot wouldn't it?

Rich_cb wrote:

 The paper I linked to looked at serious head injuries, if helmets do reduce fatalities the most obvious way they would do so is by reducing serious head injuries. Implying that I somehow moved the goalposts is a bit disingenuous. If helmets reduce death from head injury then they would also reduce rates of serious head injury. The data shows a reduction in serious head injury. Helmets would not reduce other minor injuries and may increase the number of minor head injuries (by converting serious head injuries to minor ones). The data shows a static number of overall injuries indicating an increase in minor injuries. The data is behaving as you would expect if helmets were reducing death from serious head injuries. It doesn't prove causation but it is evidence that supports the hypothesis.

Have I missed your posting of the details of the number of minor injuries over the period or is this another made up statement that you have mistakenly written in such as way as to look like a fact

Avatar
BehindTheBikesheds replied to alansmurphy | 6 years ago
2 likes

alansmurphy wrote:

Are you not sleeping well, is this really complex data keeping you up at night? How do you hypothesise One Directions influence? They weren't even born (probably) in 1995, is there some kind of greater power at work? But as their record sales increased so did fatalities, it's all there in the data...

Avatar
davel replied to ClubSmed | 6 years ago
3 likes

ClubSmed wrote:

For any that are new to this thread and finding over 200 posts to be daunting, here is a summary:

Rich_CB: Cycle Helmets prevent Cyclist Deaths!

Other Thread posters: There is no proof of that

Rich_CB: Yes there is, here are two graphs

Other Thread posters: but pedestrian fatalities drop too and at a similar rate

Rich_CB: They start at different times so it must be different factors

Other Thread posters: It could be the same factor that affects cyclist more slowly or later

Rich_CB: No, it has to be cycling specific

Other Thread posters: No it doesn't and your graphs do nothing to prove causation

Rich_CB: I don't have to, it shows correlation and that's all I need

Other Thread posters: But they are not even the same data set

Rich_CB: The same data sets don't exist so they don't have to be the same

Other Thread posters: Yes they do

Rich_CB: No they don't, and here is a paper that proves Cycle Helmets prevent Cyclist Injuries

Other Thread posters: Wait, you said cyclist deaths initially

Rich_CB: Same thing

Other Thread posters: No it isn't, one being true does not make the other true. One could be true and make the other worse.

Rich_CB: Whatever, it's clear proof, I win

Other Thread posters: But pedestrian head injuries drop significantly too and they don't wear helmets

Rich_CB: That doesn't matter as cyclist head injuries drop more

Other Thread posters: But cyclist head injuries drop more for children who your graph shows not to have had an increase in wearing helmets...

Rich_CB: There is not the matching child pedestrian data so we have to discount that

Other Thread posters: But when you have not had matching data, that did not stop you. Anyway, here is the matching child pedestrian data and it doesn't support your hypothesis

Rich_CB: Helmets effect children differently

Other Thread posters: Really, I didn’t know that. Where is that information?

Rich_CB: I didn't say it was a fact!

Other Thread posters: Uh, OK?!?!

Rich_CB: Anyway, I have proven correlation

Other Thread posters: No you haven't, they are different data sets! But OK, have some proof of correlation with Cyclist Fatalities and One Direction/ iPhones/ Age of Miss America/

Rich_CB: Yawn! You are being silly

Other Thread posters: No more so than you.

 

Fair play Smeds: I've called you wardenny before but this made me properly crease up (Rich_cb's response made me chuckle too, but with the caveat that he was working from tidy material).

 

Chapeau. 

Avatar
ClubSmed replied to ClubSmed | 6 years ago
4 likes

ClubSmed wrote:

For any that are new to this thread and finding over 200 posts to be daunting, here is a summary:

Rich_CB: Cycle Helmets prevent Cyclist Deaths!

Other Thread posters: There is no proof of that

Rich_CB: Yes there is, here are two graphs

Other Thread posters: but pedestrian fatalities drop too and at a similar rate

Rich_CB: They start at different times so it must be different factors

Other Thread posters: It could be the same factor that affects cyclist more slowly or later

Rich_CB: No, it has to be cycling specific

Other Thread posters: No it doesn't and your graphs do nothing to prove causation

Rich_CB: I don't have to, it shows correlation and that's all I need

Other Thread posters: But they are not even the same data set

Rich_CB: The same data sets don't exist so they don't have to be the same

Other Thread posters: Yes they do

Rich_CB: No they don't, and here is a paper that proves Cycle Helmets prevent Cyclist Injuries

Other Thread posters: Wait, you said cyclist deaths initially

Rich_CB: Same thing

Other Thread posters: No it isn't, one being true does not make the other true. One could be true and make the other worse.

Rich_CB: Whatever, it's clear proof, I win

Other Thread posters: But pedestrian head injuries drop significantly too and they don't wear helmets

Rich_CB: That doesn't matter as cyclist head injuries drop more

Other Thread posters: But cyclist head injuries drop more for children who your graph shows not to have had an increase in wearing helmets...

Rich_CB: There is not the matching child pedestrian data so we have to discount that

Other Thread posters: But when you have not had matching data, that did not stop you. Anyway, here is the matching child pedestrian data and it doesn't support your hypothesis

Rich_CB: Helmets effect children differently

Other Thread posters: Really, I didn’t know that. Where is that information?

Rich_CB: I didn't say it was a fact!

Other Thread posters: Uh, OK?!?!

Rich_CB: Anyway, I have proven correlation

Other Thread posters: No you haven't, they are different data sets! But OK, have some proof of correlation with Cyclist Fatalities and One Direction/ iPhones/ Age of Miss America/

Rich_CB: Yawn! You are being silly

Other Thread posters: No more so than you.

Time for the 300+ post update amendment:

Rich_CB: As nobody can prove my hypothesis wrong then it must be right!

Other Thread posters: That's not how it works, stop abusing science

Rich_CB: 'The Scientific Process' is clearly another thing I know everything about and you know nothing

Other Thread posters: You mean 'The Scientific METHOD'?

Rich_CB: Same thing, anyway here is a graphic from  Wikipedia to prove that I know what I am talking about.

Other Thread posters: Uh, OK?!?!

Rich_CB: And here are all my thoughts and data and how they fit the model

Other Thread posters: Why can't you see that if everyone else is saying that you are wrong (from both sides of the helmet debate), then you should take another look?

Rich_CB: The fact that posters here are resorting to insults means that I am right!

Other Thread posters: You clearly aren't learning anything here. The next step is to publish your results, why don't you see if one of the journals will do that?

Rich_CB: Classic deflection when you have no argument!

Other Thread posters: No argument, what have the last 300 posts been about?

Rich_CB: My thoughts fit perfectly into the framework from Wikipedia so I win!

Other Thread posters: Your thoughts do not fit the scientific model, that was such a crude attempt. The only question that does support your hypothesis is the one you have no data for

Rich_CB: Then the BMJ should be informed as my paper is pretty much the same as theirs

Other Thread posters: So you are now admitting no fresh ideas only plagiarism? Anyway, the BMJ are already aware of the issues including the gaping mathematical error.

Avatar
alansmurphy | 6 years ago
1 like

If you're arguing coincidence for one then argue it for both...

Avatar
davel | 6 years ago
2 likes

Something happened to ped deaths around the same time as the downward trend in cyclist deaths to make them fall at about the same rate.

What's more likely:

Scenario A: an increase in helmet use from 15% to 30% resulted, completely independently, in cyclist death rates dropping around the same as pedestrians death rates did, for completely different reasons, around the same time?

Scenario B: a multitude of factors resulted in roads becoming increasingly safer for cyclists and peds around the same time, but because they're different types of user the trends don't follow exactly the same pattern?

Avatar
Rich_cb replied to davel | 6 years ago
0 likes
davel wrote:

Something happened to ped deaths around the same time as the downward trend in cyclist deaths to make them fall at about the same rate.

What's more likely:

Scenario A: an increase in helmet use from 15% to 30% resulted, completely independently, in cyclist death rates dropping around the same as pedestrians death rates did, for completely different reasons, around the same time?

Scenario B: a multitude of factors resulted in roads becoming increasingly safer for cyclists and peds around the same time, but because they're different types of user the trends don't follow exactly the same pattern?

If your explanation is that the same factor is affecting both groups how do you explain the rapid fall in pedestrian fatalities prior to 1995 which occurred while there was no significant change in the cycling fatality rate whatsoever?

The most logical explanation is a pedestrian specific factor.

Avatar
davel replied to Rich_cb | 6 years ago
2 likes

Rich_cb wrote:
davel wrote:

Something happened to ped deaths around the same time as the downward trend in cyclist deaths to make them fall at about the same rate. What's more likely: Scenario A: an increase in helmet use from 15% to 30% resulted, completely independently, in cyclist death rates dropping around the same as pedestrians death rates did, for completely different reasons, around the same time? Scenario B: a multitude of factors resulted in roads becoming increasingly safer for cyclists and peds around the same time, but because they're different types of user the trends don't follow exactly the same pattern?

If your explanation is that the same factor is affecting both groups how do you explain the rapid fall in pedestrian fatalities prior to 1995 which occurred while there was no significant change in the cycling fatality rate whatsoever? The most logical explanation is a pedestrian specific factor.

No: I'm saying there are multiple factors that will have affected different road users differently.

You're trying to simplify this into single factors, and are arguing for Scenario A.

Avatar
Rich_cb replied to davel | 6 years ago
0 likes
davel wrote:

No: I'm saying there are multiple factors that will have affected different road users differently.

You're trying to simplify this into single factors, and are arguing for Scenario A.

There are obviously multiple factors.

What I'm arguing is that the pre 1995 decline in pedestrian fatalities was due to an additional pedestrian specific factor.

How else do you explain the huge decline in pedestrian fatalities (pre 1995) while cyclist fatalities remained unchanged?

Avatar
zanf | 6 years ago
0 likes

Visit any of the US dominant cycling subReddits and they are a cesspit of supidity, especially with regards to helmets.

 

Avatar
ClubSmed | 6 years ago
0 likes

Davel, you just got me questioning why I started wearing a helmet, I now remember that it was because of pressure from my (now ex) wife due to an accident that her father had during a commute by bicycle. I am glad I do now as most of my commute is along a canal, a river and through a park where the helmet saves my head from multiple branch lashings.

I also have to travel around 500 metres each side of this by road, and have come to believe that the helmet will probably give me decent protection during this time given the speed I am likely going.

I agree that the onus should be on the driver to look out for cyclists (that ideal state is certainly not going to happen in the near future) but I don't think that striving for this negates the need for a helmet. There are always going to be issues that cannot be controlled like black ice, falling trees/branches, animals running out etc.

I also agree that the helmet projects an element of danger that may put any would be cyclists off, but I don't think it creates as much as a barrier as the perceived need for lycra (I also wear lycra on my commute by the way).

Avatar
davel replied to ClubSmed | 6 years ago
0 likes

ClubSmed wrote:

I don't think that striving for this negates the need for a helmet. There are always going to be issues that cannot be controlled like black ice, falling trees/branches, animals running out etc.

I also agree that the helmet projects an element of danger that may put any would be cyclists off, but I don't think it creates as much as a barrier as the perceived need for lycra (I also wear lycra on my commute by the way).

So, to me, the risk of the first (I see the risks of having an off in which a helmet would protect me as pretty insignificant) doesn't outweigh the potential damage. But, again, in something that's impossible to measure, it probably comes down to ideology.

I'd sooner we stop apologising for existing alongside cars, and I suppose I see helmets as feeding that, while it isn't encouraged in any other road user (who happen to die in greater numbers).

But yes: I also used to have a section of commute where my head would get beaten up by branches... if I hadn't worn a helmet. It was pretty useful for that.

Pages

Latest Comments