Let's be careful out there

[Paul Smith]

On 17 Jan 2003 23:40:28 -0000, [email protected] (Geraint Jones) wrote:

>> You really are a complete ****** aren't you?

>I suppose that is not as bad-tempered as "bugger off", so it must be some sort of an improvement.
>How would "High blood pressure kills" do for a road safety slogan?

>I know it is not as catchy, but "Half mass times the square of speed kills."

So why are faster roads safer then? There must be more to it in practice.
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[James Annan]

Paul Smith wrote:

> So why are faster roads safer then? There must be more to it in practice.

Well it's a toss-up whether you are just so stupid you actually can't work it out for ourself, or
just pretending to be so stupid that you can't work it out for yourself.

But in either case, one very important factor is that the faster roads have many fewer junctions per
mile. Another is that they have wider carriageways too, generally speaking.

James

 

[Paul Smith]

On Sat, 18 Jan 2003 09:29:11 +0900, James Annan <[email protected]> wrote:

>Paul Smith wrote:

>> So why are faster roads safer then? There must be more to it in practice.

>Well it's a toss-up whether you are just so stupid you actually can't work it out for ourself, or
>just pretending to be so stupid that you can't work it out for yourself.

>But in either case, one very important factor is that the faster roads have many fewer junctions
>per mile. Another is that they have wider carriageways too, generally speaking.

Reduced to basics, faster roads are safer because they offer fewer opportunities for driver error.
This tells us that driver error is a far more important accident causation factor than mere speed.
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[Tony Raven]

James Annan <[email protected]> wrote:
>
> But in either case, one very important factor is that the faster roads have many fewer junctions
> per mile. Another is that they have wider carriageways too, generally speaking.
>

Are you saying its junctions, not speed that kills, James? ;-)

Out of curiousity I looked up the Japanese accident statistics.
http://www.npa.go.jp/toukei/koutuu1/01home/homee.htm A country where speed limits are mostly
around 30mph, slower in town, 60mph on motorways, there is very high use of public transport
and most cyclists cycle on the pavement. For a country with twice our population, there is
about the same number of vehicle fatalities but five times the number of cyclist fatalities.
Not sure what it means.

Tony ;-)

http://www.raven-family.com

"The single biggest problem in communication is the illusion that it has taken place" George
Bernard Shaw.

 

[James Annan]

Tony Raven wrote:

>
> Are you saying its junctions, not speed that kills, James? ;-)

That's right - it's not the speed that kills, it's the sudden stops...get rid of those pesky brake
pedals (levers) and we'd all be much safer.

> Out of curiousity I looked up the Japanese accident statistics.
> http://www.npa.go.jp/toukei/koutuu1/01home/homee.htm A country where speed limits are mostly
> around 30mph, slower in town, 60mph on motorways, there is very high use of public transport and
> most cyclists cycle on the pavement. For a country with twice our population, there is about the
> same number of vehicle fatalities

I'm very surprised by this, I would expect it to be much higher numerically at least if not
proportionally. And certainly I'd expect to see many more pedestrian casualties given the complete
lack of segregation in many areas. Obviously those slow speeds work after all...

> but five times the number of cyclist fatalities. Not sure what it means.

Well that almost certainly means that it's much safer here per mile cycled, despite the limited road
space and rather poor habits of most cyclists (wrong side of road seems very common).

I've just seen that HALF the bicycle fatalities are in the over-70 age range! And more than 2/3 are
60 and over...

Thanks for the link - very interesting.

James

 

[Tony Raven]

From TRL511 - http://www.trl.co.uk/1024/mainpage.asp?page=140 - The relationship between speed and
accidents on rural single-carriageway roads

"Accident frequency for all categories of accident increased rapidly with mean speed - the total
injury accident frequency increased with speed to the power of approximately 2.5 - thus indicating
that a 10% increase in mean speed results in a 26% increase in the frequency of all injury
accidents."

"The effect of mean speed was found to be particularly large for junction accidents; these accidents
were roughly proportional to the 5th power of speed, suggesting substantial potential for accident
reduction from strategies designed to reduce speeds at junctions."

Tony

http://www.raven-family.com

"The single biggest problem in communication is the illusion that it has taken place" George
Bernard Shaw.

 

[Paul Smith]

On Sat, 18 Jan 2003 15:30:35 -0000, "Michael MacClancy" <[email protected]> wrote:

>"Paul Smith" <[email protected]> wrote

>> Well, TRL 421 contains information supporting the assertion. They claim there's a relationship
>> between the accident rate and the number of "speeders". So that's easy then. Get rid of the
>> limits and we'll have no speeders and fewer accidents.

>B*llsh*t. Why then does the information about 421 on the TRL website say "The overall potential for
>accident reduction from measures to restrain speed is large." The word here is 'SPEED'. Stop
>misrepresenting other people's findings to meet your own ends.

From page 23 of TRL 421 (as a PDF)
===========================================================
"The practical consequences of influencing the proportion of speeders may be illustrated as follows.
If the proportion of speeders were to increase by one tenth, for example from 20% to 22%, the
accident frequency would be expected to increase by 1.1%, if all else is held constant. If on the
other hand the non-compliance level could be halved from 20% to 10% then the accident frequency
would be reduced by about 8%."
===========================================================

This is supposed to be serious research, but it's not. It's a joke.

These psudeo-scientists haven't noticed that their claim about the relationship between "number of
speeders" and accident rate could equally well be satisfied by removing the speed limits.

I'm very ashamed indeed of the TRL for publishing this utter rubbish. There's plenty more wrong with
the report too, especially the assumption of causality. I've email them asking for an explanation,
and promised to publish their defence of the report. Of course I've had no reply because there is no
defence. See:

http://www.safespeed.org.uk/trl421.html and http://www.safespeed.org.uk/onethirdemail.html
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[Paul Smith]

On Sat, 18 Jan 2003 17:44:50 +0000 (UTC), "W K" <[email protected]> wrote:

>"Paul Smith" <[email protected]> wrote in message
>news:[email protected]...

>> Reduced to basics, faster roads are safer because they offer fewer opportunities for driver
>> error. This tells us that driver error is a far more important accident causation factor than
>> mere speed.

>It might be, but as human beings apart from you make mistakes, its often better to make them drive
>more slowly than they believe they can.

I've seen no evidence to support that.

>Everyone thinks they are better drivers than they are, and everyone drives in the way they can get
>away with without accidents.

And largely they succeed. We still have the safest roads in the world.

>Remember that 1 sec gap bit? People make just as bad a choice about their choice of "safe speed".

No they don't. If they did we would have literally millions of accidents each day.

The system works quite well, and real improvements are likely to come from nudges which improve on
the aspects which already work well.
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[Paul Smith]

On Sat, 18 Jan 2003 18:12:48 -0000, "Tony Raven" <[email protected]> wrote:

>From TRL511 - http://www.trl.co.uk/1024/mainpage.asp?page=140 - The relationship between speed and
>accidents on rural single-carriageway roads

>"Accident frequency for all categories of accident increased rapidly with mean speed - the total
>injury accident frequency increased with speed to the power of approximately 2.5 - thus indicating
>that a 10% increase in mean speed results in a 26% increase in the frequency of all injury
>accidents."

>"The effect of mean speed was found to be particularly large for junction accidents; these
>accidents were roughly proportional to the 5th power of speed, suggesting substantial potential for
>accident reduction from strategies designed to reduce speeds at junctions."

Oh yes. I've read it time and again. Every time I read it I find another serious flaw. It's
propaganda, not science. I wish I knew why the TRL has allowed it to happen, but I don't.

I cannot trust a single word of it.

http://www.safespeed.org.uk/trl421.html

Exposes the most basic lie.
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[Just Zis Guy]

On Sat, 18 Jan 2003 18:12:48 -0000, "Tony Raven" <[email protected]> wrote:

>Accident frequency for all categories of accident increased rapidly with mean speed

No, Tony, that simply won't do. It's completely bogus - they indulge in all sorts of dreadful
sleight of hand, like comparing roads with similar speed limits instead of comparing everything with
those nice, safe motorways. Any rational analysis shows that the best way forward is to increase the
speed limit in built-up areas to 70, so they too will become as safe as motorways. Should also
reduce the incidence of speeding, which would be a useful collateral benefit.

Guy
===
** WARNING ** This posting may contain traces of irony. http://www.chapmancentral.com (BT ADSL and
dynamic DNS permitting)
NOTE: BT Openworld have now blocked port 25 (without notice), so old mail addresses may no longer
work. Apologies.

 

[Just Zis Guy]

On Sat, 18 Jan 2003 16:50:22 +0000, Paul Smith <[email protected]> wrote:

>We could still kill just as many if all speed limits were reduced to 12mph.

Dear oh dear. At 12mph many pedestrians and virtually all cyclists could easily outrun the cars. You
also need to get this schizophrenia seen to. "We," indeed.

Guy
===
** WARNING ** This posting may contain traces of irony. http://www.chapmancentral.com (BT ADSL and
dynamic DNS permitting)
NOTE: BT Openworld have now blocked port 25 (without notice), so old mail addresses may no longer
work. Apologies.

 

[Tim Woodall]

On Sat, 18 Jan 2003 16:50:22 +0000, Paul Smith <[email protected]> wrote:
>
> http://www.safespeed.org.uk/12mph.html
>
> We could still kill just as many if all speed limits were reduced to 12mph.

What planet are you living on?

Take your figure - average impact speed 21 mph for current deaths.

Now change all speeds by 1mph. It is reasonable to assume that this average impact speed figure
changes by at least the same.

So p(21.21) = 0.0079
p(20.21) = 0.0065
p(20.22) = 0.0095

p(20.23)*(132318 + 12555 + 1164) = 1154. (OK some rounding errors there)

p(20.24)*(132318 + 12555 + 1164) = 949.
p(20.25)*(132318 + 12555 + 1164) = 1387.

So a 1mph reduction in all speeds leads to an 18% reduction in fatalities.

And 1 mph increase in all speeds leads to an 20% increase in fatalities

What was it that TRL, Aussies etc came up with? 5mph-10mph speed reduction results in a halving of
injury accidents. OK we're talking about fatalities here rather than injury accidents but your
figures seem to agree very closely with the expert research.

Finally, your 12mph page is incoherent nonsense. Clearly if all speeds were limited to 30mph maximum
there could be no accidents at the 70mph [1] end but you don't even consider that. Merely that if
all speeds were limited to 30mph then all collisions would be at 30mph (why not 60mph it would look
so much better?)

Tim.

[1] actually it is theoretically possible that there could be but I don't think cars meet the
requirements for (almost) elastic collisions necessary[2]

[2] Problem for anyone who is interested - how can you drop a ball (without giving it any initial
velocity) and have it bounce 3 times higher than the height you dropped it from. (And once you
have solved that problem extrapolate it

<spoiler below> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . .

Consider:

. <---- light ball (e.g. ping pong ball) o <---- much heavier ball (e.g. tennis ball)

----- <--- ground.

Drop the two balls together but not quite touching one above the other. (You will probably find that
you won't usually get them quite lined up and the ping pong ball will ricochet off across the room -
the same trick works with a basket ball and a tennis ball but I suggest you don't try it indoors
In theory putting all three together should be quite spectacular - in practice you spend a lot of
time chasing tennis balls and ping pong balls that shoot off in strange directions

--
God said, "div D = rho, div B = 0, curl E = - @B/@t, curl H = J + @D/@t," and there was light.

http://tjw.hn.org/ http://www.locofungus.btinternet.co.uk/

 

[Just Zis Guy]

On Tue, 21 Jan 2003 13:59:11 -0000, "Tony Raven" <[email protected]> wrote:

>The things I find surprisingly effective are those flashing slow down signs.

There used to be one at the end of the A33 Winchester By-Pass - it was a point of pride to make it
light up, but that was before Gatsos.

Guy
===
** WARNING ** This posting may contain traces of irony. http://www.chapmancentral.com (BT ADSL and
dynamic DNS permitting)
NOTE: BT Openworld have now blocked port 25 (without notice), so old mail addresses may no longer
work. Apologies.

 

[Paul Smith]

On Sat, 18 Jan 2003 17:15:37 +0000, "Just zis Guy, you know?" <[email protected]> wrote:

>On Sat, 18 Jan 2003 16:47:14 +0000, Paul Smith <[email protected]> wrote:

>>Of course I've had no reply because there is no defence.

>Or maybe because they are all helpless with laughter at your re-assessment of Figure A2, discarding
>their obviously bogus criteria for grouping roads (by speed limit) in favour of an approach which
>ignores this and demonstrates that 70mph speed limits would increase safety on urban roads.

I have not re-assessed figure A2. That's their figure and their data.

And they are certainly not "helpless with laughter". See:
http://www.safespeed.org.uk/onethirdemailagb.html

The conclusions of the reports (TRL421 and TRL511) are absurd. Their road classification methods are
pure invention, and even slightly different classification criteria would yield a completely
different model.

Consider for a moment (if you've studied the reports) exactly how a road could have a higher speed
and a higher accident rate and yet be classified similarly to another road with a lower speed and a
lower accident rate. The whole model depends on finding conflicting road groupings like these for
its very existence. If such road pairings exist, we must consider why they are driven at different
speeds. Could it be because they are different in some important respect?

They even bend the model by including accident rate as a highly weighted road quality factor.
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[Paul Smith]

On Sat, 18 Jan 2003 17:23:23 +0000, "Just zis Guy, you know?" <[email protected]> wrote:

>>We could still kill just as many if all speed limits were reduced to 12mph.

>Dear oh dear. At 12mph many pedestrians and virtually all cyclists could easily outrun the cars.
>You also need to get this schizophrenia seen to. "We," indeed.

So find the flaw(s) in the research cited and the simple calculations on:

http://www.safespeed.org.uk/12mph.html

That "we" is "as a nation"; "we killed 107 child pedestrians in 2001" for example.
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[Just Zis Guy]

On Sat, 18 Jan 2003 18:39:38 +0000, Paul Smith <[email protected]> wrote:

>I thought you had integrity. Why snip the rest of my post in an attempt to distort my point? You
>should be ashamed of yourself.

Yet, strangely, I'm not. Maybe because your post contained nothing which hadn't been seen here
before and replied to elsewhere.

Guy
===
** WARNING ** This posting may contain traces of irony. http://www.chapmancentral.com (BT ADSL and
dynamic DNS permitting)
NOTE: BT Openworld have now blocked port 25 (without notice), so old mail addresses may no longer
work. Apologies.

 

[W K]

"Paul Smith" <[email protected]> wrote in message
news:[email protected]...
> On Sun, 19 Jan 2003 09:15:44 +0000 (UTC), "W K"

> >Errr. who says we have millions of accidents a day from the bad choice of minor tailgating?
>
> We don't. But there's no inevitable accident from tailgating.

Yes there is, anything unexpected happens and you hit the back of them. Just as if you drive at a
speed where a broken down car, a tractor spotted at precisely the wrong time etc. will make driving
at the limits of your "safe speed" will cause an accident.

>Driving too fast for the conditions is far far harder to get away with. In fact it's almost
>impossible. Anyone who consistently drives significantly too fast for the conditions is unlikely to
>last a week without crashing.

WOW thats the best one yet. So, the "britains worst driver" chaps [*] were not people who drove
significantly too fast?

SO, all drivers go at a reasonable speed as they seldom crash? Never mind that accident causing
situations are so rare that even if you drive at russian roulette speeds, there's a fair chance
you'll spend years between accidents. Just imagine the tractor scenario, its 1000000:1 that you meet
it at precisely the wrong place that would cause you to emergency brake. Hence, a little faster and
you have to go round a million corners before you hit one.

[* Young asian show off lad and goaty bearded 30-40 bloke that wrote off lots of cars]

 

[Paul Smith]

On Sat, 18 Jan 2003 18:30:36 +0000, "Just zis Guy, you know?" <[email protected]> wrote:

>On Sat, 18 Jan 2003 17:30:28 +0000, Paul Smith <[email protected]> wrote:

>>I have not re-assessed figure A2. That's their figure and their data.

>And your trend line.

No. That's their trend line. The figure is completely unchanged.

>>And they are certainly not "helpless with laughter". See:
>>http://www.safespeed.org.uk/onethirdemailagb.html

>I'm aghast! The DfT refuse to change policy in response to pleading from a motorist that speeding
>isn't dangerous! Whatever next? Maybe the Home Office will start refusing to agree with muggers who
>write about how safe mugging is?

>And I've read the reports and see nothing particularly controversial in their methods. This may be
>because I am not a statistician despite having achieved an A in maths A level, or it might be
>because I have absolutely no interest in opposing the enforcement of the law, which was the law
>when I passed my driving test and I've never been under any illusion that it was not the law.

But you want improved road safety. TRL421 represents what the government are offering, and it's so
full of holes that you could drive a bus through it. You should be worried.

>I have found that matching the black numbers in the red circles with the little pointer thingy on
>the dashboard prevents any of the cameras from flashing (most cars have this pointer thingy, it
>will be in the owners handbook under "speedometer").

Of course. If we all do that, the fluffy bunnies will be able to play in the fields again and all
will be well.
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[Paul Smith]

On Sat, 18 Jan 2003 18:32:36 +0000, "Just zis Guy, you know?" <[email protected]> wrote:

>Not that "we," this one (and others like it):

>"We use figures from official sources and well respected research to show......"

You think I do 100% of the work on SafeSpeed? Really? I get quite a bit of help.
--
Paul Smith Scotland, UK http://www.safespeed.org.uk please remove "XYZ" to reply by email Let's make
speed cameras as unacceptable as drink driving

 

[Tony W]

"Paul Smith" <[email protected]> wrote in message
news:[email protected]...

>
> Oh, very good. Why don't you tell us how you would relate the Joksch curve to the ratio of near
> misses to fatals. You might even like to infer an average impact speed.

It would be relatively easy to do though I am not convinced the Joksch curve would be the method I
would apply (see Tim's post for some of the reasons). As for inferring average impact speeds that
hardly represents rocket science.

The difference would be that I would know what I was doing in analysing the figures.

T