Safespeed lies and the Press Complaints Commission

[Clive George]

"Mr R@t (2.30 zulu-india)" <[email protected]> wrote in message
news:[email protected]...

> IME most local rags are not even worthy of litter tray lining material.
> They *thrive* on death, destruction, controversy and conflict. this comes
> from someone who once wanted to be a journalist BTW! In this day and age
> people don't believe in "good news" as its dismissed as spin; or it
> relates to someone doing something they don't agree with.

You live in the wrong place

We get our local rag, the Craven Herald, in part because the local village
news section is a reminder that nice people still exist and all is still
right with the world - it's populated by short paragraphs about the latest
fete, show, tombola, fete, talk, slideshow, WI event (normally with
clothes), whatever.

There was a murder trial for an incident in Settle a few years ago, but it
turned out not guilty (or case abandoned, can't remember) - insufficient
evidence that it was deliberate. (somebody fell off a balcony during a
fight).

Still, the photos of people in the paper do at least confirm to the local
press rules : there are two poses, one looking grumpy in front of something
(local protests about something) and one with everybody grinning cheesily,
often holding something (local gives to charity, wins something, opens shop,
etc).

cheers,
clive

 

[spindrift]

"That's only your personal opinion that is it 'obsure'."

Nope:

By and large, Safe Speed's claims have been ignored by the scientific
community, and there have been few, if any, formal studies evaluating
them. However, Safe Speed's interpretation of research has in some
cases been directly rebutted by the authors of that research, including
TRL and Hans Jocksch.

http://en.wikipedia.org/wiki/Safespeed#Opposition_and_Criticism



"As far as I can see Safe Speed is a road safety group. "

You think advising on ways to pervert justice is compatible with that
description?

Put it this way, would increasing the numbers of untraceable drivers
who know their crimes would go unpunished make the roads safer, or more
dangerous?

"Wrong. He claims inappopriate speed is dangerous."

I'm afraid not. Smith is the classic arrogant driver who over-estimates
his own abilities. He suggests that drivers always know the safest
speed to drive at, and he uses some laughable "research" to back up his
claims.

For instance:


"One third of fatalities are now caused by speed cameras" .

By extrapolating the change in number of fatalities between two
selected years to predict the fatality figure had that year-on-year
trend continued, Safe Speed reach this figure based on the advent of
speed cameras in the early 1990s.


The "one third" claim is disputed by the Parliamentary Advisory Council
for Transport Safety (PACTS) and the National Safety Camera Liason
(NSCL), which cite seatbelt and alcohol laws introduced prior to the
1990s, and recent increased road use and mobile phone use as better
explanations for the perceived increase in casualties[1]. NSCL also
points out that Safe Speed's figures are based entirely on accident
totals, rather than distinguishing roads with or without speed cameras.

Which? magazine reports[1] that NSCL cite three studies which do allow
for long-term trends, and which confirm the correlation between speed
cameras and accident reduction.

Ibid.

"Which means the Safe Speed view *must* be invalid?! "

Well, on the one hand we have accredited, peer-reviewed scientific
studies carried out by non-governmental agencies with no axe to grind.
(Unis of Liverpool, Bristol, TRL, ROSPA, ACPO etc etc)

On the other hand we have a man who refuses point blank to submit a
single page of his risible "research" to peer review and uses his
website to make very unpleasant personal attacks on anyone who dares
suggest that his reasoning is flawed.


One lone example:

His latest outburst has to be seen to be
believed.


http://news.bbc.co.uk/1/hi/uk/4646008.stm


Quote from above story: Paul Smith of road safety campaign group
Safespeed said:

"White van man poses less risk to pedestrians than even
cyclists and is one of the very safest road user groups."


When you`ve picked yourself up laughing at this terrible, dangerous
lie, take a look at the reasoning behind it.


Smith`s explanation:


http://www.safespeed.org.uk/forum/viewtopic.php?t=5737&highlight=


The usual Smith ploy of baffling with inappropriate statistical
comparisons...



Smith`s argument is that the risk of a van hitting a
pedestrian per van mile driven is less than the risk of a bus or
bicycle hitting a pedestrian per bus-mile driven or per mile cycled.


This conveniently ignores the very different exposure to pedestrians of

each mode: both buses and bicycles do most of their mileage on roads
with comparatively high levels of pedestrian use; whereas total van
mileage includes lots of motorway and trunk road travel with very low
or zero pedestrian exposure.


No cyclists on motorways either.


Thus a totally meaningless conclusion is drawn from stats that look
convincing because they are taken from real data. The funny thing is
that Smith is so terribly pompous about his use of statistics: a quote
from the SS forum: "We can win the arguments on bare facts and logic"!


This is a schoolboy howler by a man hopelessly out of his depth.


What a laugh. If anyone would like to visit Safespeed and put this
point directly to Smith then please go right ahead. I tried
and my account was banned and the post deleted.


What's he so terrified of that he not only ignores dissent, he actually
bans it!

 

[Matt B]

spindrift wrote:
> One lone example:
>
> His latest outburst has to be seen to be
> believed.
>
> http://news.bbc.co.uk/1/hi/uk/4646008.stm
>
> Quote from above story: Paul Smith of road safety campaign group
> Safespeed said:
>
> "White van man poses less risk to pedestrians than even
> cyclists and is one of the very safest road user groups."
>
> When you`ve picked yourself up laughing at this terrible, dangerous
> lie, take a look at the reasoning behind it.
>
> Smith`s explanation:
>
> http://www.safespeed.org.uk/forum/viewtopic.php?t=5737&highlight=
>
> The usual Smith ploy of baffling with inappropriate statistical
> comparisons...
>
> Smith`s argument is that the risk of a van hitting a
> pedestrian per van mile driven is less than the risk of a bus or
> bicycle hitting a pedestrian per bus-mile driven or per mile cycled.
>
> This conveniently ignores the very different exposure to pedestrians of
>
> each mode: both buses and bicycles do most of their mileage on roads
> with comparatively high levels of pedestrian use; whereas total van
> mileage includes lots of motorway and trunk road travel with very low
> or zero pedestrian exposure.
>
> No cyclists on motorways either.
>
> Thus a totally meaningless conclusion is drawn from stats that look
> convincing because they are taken from real data. The funny thing is
> that Smith is so terribly pompous about his use of statistics: a quote
> from the SS forum: "We can win the arguments on bare facts and logic"!

Take a look at the DfT's statistics on road casualties - Road Casualties
Great Britain: 2004 (September 2005), chart 26 on page 77.

You will see that for /urban/ roads the KSI rate (killed and seriously
injured) of pedestrians per 100 million vehicle km is:
1.2 by LGVs (vans)
1.5 by Pedal cycles
2.8 by HGVs
3.1 by Cars
10.8 by Motorbikes
11.5 by Bus or coach

In urban situations per km travelled (ignoring distance travelled on
rural roads) vans appear to be less of a threat to pedestrians than any
other vehicle type, including pedal cycles.

Now lets look at rural roads, same units:
0.3 LGVs
0.5 HGVs
0.6 Car
0.7 Pedal cycles
1.6 Motorbikes
1.6 Bus or coach

Oh dear! Vans still pose the least threat per km travelled, and even
HGVs and cars pose less of a threat than pedal cycles, km for km.

--
Matt B

 

[Mark Thompson]

> So, cameras do work, after a fashion. But I'm not
> impressed.

Nor me. They only catch the twunts who fail to notice the warning signs,
white lines slap down the middle of the lane and the large 10 foot high
bright yellow box.

 

[Budstaff]

"Matt B" <[email protected]> wrote in message
news:[email protected]...
> Take a look at the DfT's statistics on road casualties - Road Casualties
> Great Britain: 2004 (September 2005), chart 26 on page 77.
>
> You will see that for /urban/ roads the KSI rate (killed and seriously
> injured) of pedestrians per 100 million vehicle km is:
> 1.2 by LGVs (vans)
> 1.5 by Pedal cycles
> 2.8 by HGVs
> 3.1 by Cars
> 10.8 by Motorbikes
> 11.5 by Bus or coach
>
> In urban situations per km travelled (ignoring distance travelled on rural
> roads) vans appear to be less of a threat to pedestrians than any other
> vehicle type, including pedal cycles.
>
> Now lets look at rural roads, same units:
> 0.3 LGVs
> 0.5 HGVs
> 0.6 Car
> 0.7 Pedal cycles
> 1.6 Motorbikes
> 1.6 Bus or coach
>
> Oh dear! Vans still pose the least threat per km travelled, and even HGVs
> and cars pose less of a threat than pedal cycles, km for km.
>
> --
> Matt B
Perhaps you could tell us how many KSI's _in total_ were inflicted by
cyclists and white vans on pedestrians, so that we can see if the comparison
is statistically valid. After all, if the DfT's sample rate is per 100
million klicks, then the sample size for bikes must be vanishingly small by
comparison with vans. And while you're at it perhaps you have to hand the
figures for cycling casualties inflicted by vans and vice versa?

 

[calum]

"Matt B" <[email protected]> wrote in message

> Your out-of-context selective quote doesn't show that he did 'explicitly
> encourage breaking the law'. Any reasonable person who has read the whole
> thread will see that he wasn't.

I'm a reasonable person.

In a thread where the discussion was on how to avoid accruing points on a
driving licence by naming another as the driver, Paul Smith wrote "brilliant
idea".

That's encouragement.

 

[Matt B]

Budstaff wrote:
> Perhaps you could tell us how many KSI's _in total_ were inflicted by
> cyclists and white vans on pedestrians, so that we can see if the comparison
> is statistically valid. After all, if the DfT's sample rate is per 100
> million klicks, then the sample size for bikes must be vanishingly small by
> comparison with vans.

Traffic by vehicle type (100 million vehicle km):
39 Pedal cycles
52 Motorbikes
52 Buses and coaches
294 HGVs
608 LGVs
3981 Cars and taxis

Total pedestrian KSI by vehicle type:
47 Pedal cycle
182 HGVs
306 Motorbike
325 LGVs
409 Buses and coaches
5465 Cars and taxis

> And while you're at it perhaps you have to hand the
> figures for cycling casualties inflicted by vans and vice versa?

LGV/Pedal cycle two-vehicle accidents KSI:
1 LGV user
121 Pedal cycle users

No mention though of which 'user' was to blame for any of the incidents
resulting in these casualties.

--
Matt B

 

[Just zis Guy, you know?]

On 29 Mar 2006 06:50:23 -0800, "spindrift" <[email protected]> said
in <[email protected]>:

>By extrapolating the change in number of fatalities between two
>selected years to predict the fatality figure had that year-on-year
>trend continued, Safe Speed reach this figure based on the advent of
>speed cameras in the early 1990s.

Indeed. Despite the change in reporting mechanisms around that time
which means that trends are not directly comparable, the exponential
nature of the decline over time, the fact that due to stochastic
variation individual years can't be compared, the fact that the number
of cameras only grew by a few hundreds over that time, the fact that
the "loss of trend" is not visible in the much larger (hence more
reliable) KSI trend, the fact that it applies only to motorcyclists,
the fact that no comparable trend is visible when cameras increased
very substantially in number some years later and the fact that it is
only visible on those roads least likely to have cameras. And that's
before you've even started to consider what other potential causes
there might be, starting with something which *is* proven to reduce
road safety, namely mobile phone use.

http://www.chapmancentral.co.uk/Web/public.nsf/documents/One_In_Three

Incidentally, is your mail address valid?

Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk

85% of helmet statistics are made up, 69% of them at CHS, Puget Sound

 

[Budstaff]

"Matt B" <[email protected]> wrote in message
news:[email protected]...
> Budstaff wrote:
>> Perhaps you could tell us how many KSI's _in total_ were inflicted by
>> cyclists and white vans on pedestrians, so that we can see if the
>> comparison is statistically valid. After all, if the DfT's sample rate is
>> per 100 million klicks, then the sample size for bikes must be
>> vanishingly small by comparison with vans.
>
> Traffic by vehicle type (100 million vehicle km):
> 39 Pedal cycles
> 52 Motorbikes
> 52 Buses and coaches
> 294 HGVs
> 608 LGVs
> 3981 Cars and taxis
>
> Total pedestrian KSI by vehicle type:
> 47 Pedal cycle
> 182 HGVs
> 306 Motorbike
> 325 LGVs
> 409 Buses and coaches
> 5465 Cars and taxis
>
>> And while you're at it perhaps you have to hand the figures for cycling
>> casualties inflicted by vans and vice versa?
>
> LGV/Pedal cycle two-vehicle accidents KSI:
> 1 LGV user
> 121 Pedal cycle users
>
> No mention though of which 'user' was to blame for any of the incidents
> resulting in these casualties.
>
> --
> Matt B
Thanks. From this it appears that although we (cyclists) are not entirely
blameless we're not exactly the main problem either.

 

[John B]

Matt B wrote:

> spindrift wrote:
> > Letter to the editor:
>
> You believe that a couple of dozen cameras is a better way to make our
> roads safer than increased and comprehensive policing by real people?
> Amazing.
>
> I'm sure that all you will achieve with this post/letter is more
> publicity for Safe Speed, because much of what you say about Paul Smith
> is, at best, inaccurate, and some possibly verging on the libellous.

What is inaccurate?
What do you think is libellous?
And why?


john B

 

[mb]

spindrift wrote:

> Letter to the editor:
>


I'm beginning to think you are using this newsgroup to launch attacks
on someone who has attacked you.

Do you think you could take it elsewhere, it seems to have very little
to do with cycling.

--
Mike

 

[mb]

spindrift wrote:

> "You believe that a couple of dozen cameras is a better way to make
> our
>
> roads safer than increased and comprehensive policing by real people?
> "
>
> £50 to a charity of your choice if you can find a shred of proof for
> this daft claim.
>
> Alternatively I shall take the advice of the other posters on here who
> refrain from lying about what people have posted and ignore your
> pointless trolling.


To whom are you replying? Please learn to post properly and include
some context.

Or better still, don't bother.

--
Mike

 

[Rich]

"Matt B" <[email protected]> wrote in message
news:[email protected]...
> spindrift wrote:
snip

The discussion was exploring
> /legal/ ways of avoiding driving licence points for what might be
> characterised as a technical victimless offence.

Technical victimless offence? the hospitals and graveyards around here are
full of those "non-victims"
>
> --
> Matt B

 

[David Martin]

Matt B wrote:
> Budstaff wrote:
> > Perhaps you could tell us how many KSI's _in total_ were inflicted by
> > cyclists and white vans on pedestrians, so that we can see if the comparison
> > is statistically valid. After all, if the DfT's sample rate is per 100
> > million klicks, then the sample size for bikes must be vanishingly small by
> > comparison with vans.
>
> Traffic by vehicle type (100 million vehicle km):
> 39 Pedal cycles
> 52 Motorbikes
> 52 Buses and coaches
> 294 HGVs
> 608 LGVs
> 3981 Cars and taxis
>
> Total pedestrian KSI by vehicle type:
> 47 Pedal cycle
> 182 HGVs
> 306 Motorbike
> 325 LGVs
> 409 Buses and coaches
> 5465 Cars and taxis
>
> > And while you're at it perhaps you have to hand the
> > figures for cycling casualties inflicted by vans and vice versa?
>
> LGV/Pedal cycle two-vehicle accidents KSI:
> 1 LGV user
> 121 Pedal cycle users
>
> No mention though of which 'user' was to blame for any of the incidents
> resulting in these casualties.

Are those urban figures or total figures?

...d

 

[Matt B]

Rich wrote:
> "Matt B" <[email protected]> wrote in message
> news:[email protected]...
>> spindrift wrote:
> snip
>
> The discussion was exploring
>> /legal/ ways of avoiding driving licence points for what might be
>> characterised as a technical victimless offence.
>
> Technical victimless offence?

How many of the 2 million or so speeding offences automatically dealt
with by the use of speed cameras in 2004 do you suppose also resulted in
a collision with a victim?

--
Matt B

 

[Matt B]

David Martin wrote:
> Matt B wrote:
>> Budstaff wrote:
>>> Perhaps you could tell us how many KSI's _in total_ were inflicted by
>>> cyclists and white vans on pedestrians, so that we can see if the comparison
>>> is statistically valid. After all, if the DfT's sample rate is per 100
>>> million klicks, then the sample size for bikes must be vanishingly small by
>>> comparison with vans.
>> Traffic by vehicle type (100 million vehicle km):
>> 39 Pedal cycles
>> 52 Motorbikes
>> 52 Buses and coaches
>> 294 HGVs
>> 608 LGVs
>> 3981 Cars and taxis
>>
>> Total pedestrian KSI by vehicle type:
>> 47 Pedal cycle
>> 182 HGVs
>> 306 Motorbike
>> 325 LGVs
>> 409 Buses and coaches
>> 5465 Cars and taxis
>>
>>> And while you're at it perhaps you have to hand the
>>> figures for cycling casualties inflicted by vans and vice versa?
>> LGV/Pedal cycle two-vehicle accidents KSI:
>> 1 LGV user
>> 121 Pedal cycle users
>>
>> No mention though of which 'user' was to blame for any of the incidents
>> resulting in these casualties.
>
> Are those urban figures or total figures?

Total. The stats don't break those ones down.

--
Matt B

 

[Matt B]

calum wrote:
> "Matt B" <[email protected]> wrote in message
>
>> Your out-of-context selective quote doesn't show that he did 'explicitly
>> encourage breaking the law'. Any reasonable person who has read the whole
>> thread will see that he wasn't.
>
> I'm a reasonable person.
>
> In a thread where the discussion was on how to avoid accruing points on a
> driving licence by naming another as the driver, Paul Smith wrote "brilliant
> idea".
>
> That's encouragement.

Given that they were discussing how it could be done /legally/ then even
if he thought something might be a "brilliant idea" that could hardly be
characterised an act to "explicitly encourage breaking the law".

--
Matt B

 

[Simon Proven]

Matt B wrote:
> spindrift wrote:
> > One lone example:
> >
> > His latest outburst has to be seen to be
> > believed.
> >
> > http://news.bbc.co.uk/1/hi/uk/4646008.stm
> >
> > Quote from above story: Paul Smith of road safety campaign group
> > Safespeed said:
> >
> > "White van man poses less risk to pedestrians than even
> > cyclists and is one of the very safest road user groups."
> >
> > When you`ve picked yourself up laughing at this terrible, dangerous
> > lie, take a look at the reasoning behind it.
> >
> > Smith`s explanation:
> >
> > http://www.safespeed.org.uk/forum/viewtopic.php?t=5737&highlight=
> >
> > The usual Smith ploy of baffling with inappropriate statistical
> > comparisons...
> >
> > Smith`s argument is that the risk of a van hitting a
> > pedestrian per van mile driven is less than the risk of a bus or
> > bicycle hitting a pedestrian per bus-mile driven or per mile cycled.
> >
> > This conveniently ignores the very different exposure to pedestrians of
> >
> > each mode: both buses and bicycles do most of their mileage on roads
> > with comparatively high levels of pedestrian use; whereas total van
> > mileage includes lots of motorway and trunk road travel with very low
> > or zero pedestrian exposure.
> >
> > No cyclists on motorways either.
> >
> > Thus a totally meaningless conclusion is drawn from stats that look
> > convincing because they are taken from real data. The funny thing is
> > that Smith is so terribly pompous about his use of statistics: a quote
> > from the SS forum: "We can win the arguments on bare facts and logic"!
>
> Take a look at the DfT's statistics on road casualties - Road Casualties
> Great Britain: 2004 (September 2005), chart 26 on page 77.
>
> You will see that for /urban/ roads the KSI rate (killed and seriously
> injured) of pedestrians per 100 million vehicle km is:
> 1.2 by LGVs (vans)
> 1.5 by Pedal cycles
> 2.8 by HGVs
> 3.1 by Cars
> 10.8 by Motorbikes
> 11.5 by Bus or coach
>
> In urban situations per km travelled (ignoring distance travelled on
> rural roads) vans appear to be less of a threat to pedestrians than any
> other vehicle type, including pedal cycles.
>
> Now lets look at rural roads, same units:
> 0.3 LGVs
> 0.5 HGVs
> 0.6 Car
> 0.7 Pedal cycles
> 1.6 Motorbikes
> 1.6 Bus or coach
>
> Oh dear! Vans still pose the least threat per km travelled, and even
> HGVs and cars pose less of a threat than pedal cycles, km for km.

IMV, KSI is a poor metric when comparing different KSI rates with
wildly different severity ratios (as for these accidents). K:KSI ratio
for pedal cycles KSI peds is 3:47 (0.064). For LGV it's 39:325
(0.12). This is a huge disparity. It's like comparing apples
with oranges. For K only (K is much better defined than SI,
not trivially open to fiddling, and is the maximum severity):

Urban:

0.0 Pedal Cycles
0.1 Light goods vehicle
0.2 Car
0.8 Motor Cycles
0.8 Heavy Goods Vehicle
1.5 Bus or coach

Rural

0.0 Pedal cycles
0.0 LGV
0.1 Car
0.2 HGV
0.2 Bus/coach
0.3 Motor cycles

This is of course casualties to pedestrians only. The difference
between LGV and cyclists is larger still if you include all other
classes of road users (subtract table 24 from table 41 to work
this out) you find that cyclists still come bottom of the K pile,
unsurprisingly.

Part of the reason for this being the case is that only ONE
pedestrian was killed in an accident involving them being hit
by a pedal cycle in the whole of 2004. So we can actually
work out the value to more sig. figs given the information
about miles travelled.

1/39 = 0.025
39/608 = 0.064

Ratio: 2.56

This means that LGV are actually 2.56 times more likely
to be involved in a ped. death, per veh. mile, than cyclists.
Give or take a huge margin of error since only one ped was
killed by being hit by a cycle in the entire year.

But Paul Smith says they're safer.

 

[Simon Proven]

Matt B wrote:
> Budstaff wrote:
> > Perhaps you could tell us how many KSI's _in total_ were inflicted by
> > cyclists and white vans on pedestrians, so that we can see if the comparison
> > is statistically valid. After all, if the DfT's sample rate is per 100
> > million klicks, then the sample size for bikes must be vanishingly small by
> > comparison with vans.
>
> Traffic by vehicle type (100 million vehicle km):
> 39 Pedal cycles
> 52 Motorbikes
> 52 Buses and coaches
> 294 HGVs
> 608 LGVs
> 3981 Cars and taxis
>
> Total pedestrian KSI by vehicle type:
> 47 Pedal cycle

This is the value for pedestrians KSI in accidents
/involving/ cyclists. There are separate tables
breaking down accidents by single road user
type involvement with peds. The actual figure
for pedal cycle is 1K, 42SI hit by pedal cycles,
and 3 where peds where SI when hit by other
road user types but a pedal cycle was somehow
involved in the accident. Your table thus double
counts some accidents. See RAGB table
23 for the breakdown. By comparison 2 cyclists
were K by pedestrians. This probably makes
pedestrians a more dangerous road user group
per mile travelled than cyclists. Discuss.

> No mention though of which 'user' was to blame for any of the incidents
> resulting in these casualties.

It has been reported that pedestrians are at
fault some 70% of the time. It's particularly
noticable how less careful they are when
I approach them on my bike vs when I
approach them in my car. I suggest that
pedestrians not hearing the approach of
a cyclist is a significant factor.

 

[Budstaff]

"Matt B" <[email protected]> wrote in message
news:[email protected]...
> Rich wrote:
>> "Matt B" <[email protected]> wrote in message
>> news:[email protected]...
>>> spindrift wrote:
>> snip
>>
>> The discussion was exploring
>>> /legal/ ways of avoiding driving licence points for what might be
>>> characterised as a technical victimless offence.
>>
>> Technical victimless offence?
>
> How many of the 2 million or so speeding offences automatically dealt with
> by the use of speed cameras in 2004 do you suppose also resulted in a
> collision with a victim?
>
> --
> Matt B

So drunk driving (if no collision involved) is also a 'victimless offence'?