Safe Speed Forums

Nope because I haven't been on here much this week.

Thank you Pete317 for a direct response.



Well they would appear to considering that speed cameras have been put up.



I'm sure a range of things and I don't claim that speed cameras deal with anything other than speed. My argument on this thread has been that they are "much better than nothing", not that they are the sole answer to road safety.



I don't know how many of the total number of accidents they deal with.



But this is what regression-to-mean is about. Having an unusually large number of collisions which you would expect to drop back anyway. But this study which Smeggy mentioned deals with that.



Well actually they are placed where crashes occur, but they can't be put in until there has been at least 4 KSIs. Also, just because each camera doesn't save 'just' one life per year doesn't automatically mean that they aren't having any effect. And say if the 5,000 cameras between them saved 100 lives a year - would that not be worth it? Then each camera would only be saving one-fiftieth of a life.

Just when you thought it was over....

_________________
Paul Smith: a legend.

"The freedom provided by the motor vehicle is not universally applauded, however: there are those who resent the loss of state control over individual choice that the car represents. Such people rarely admit their prejudices openly; instead, they make false or exaggerated claims about the adverse effects of road transport in order to justify calls for higher taxation or restrictions on mobility." (Conservative Way Forward: Stop The War Against Drivers)

I imagine some roads can trick people into thinking they can see everything; but just because you think there is good visibility doesn't mean there is.

I think many people see a path they can take and think "that's clear, I'm going for it" but don't spend enough time checking the surroundings of the path they see to be clear.

Human nature, I see it all the time on a smaller scale, its like people see the path, and everything to the side of it it edited out of reality.

That's all well and good, and a bit of road engineering and\or better signs will improve things, however you are not answering the question, which was



Note it does not ask "How many How many accidents occur on straight roads with apparently good visibility, and what makes them happen? "

_________________
It is difficult to get a man to understand something when his job depends on not understanding it.

Upton Sinclair

So they stop more accidents than those caused by speed?



Nope I am willing to be convinced by evidence, not just someone making a noise.



It does no such thing, in fact in this report it attempts claim all sorts of imaginary benefits from the camera. The link has never been proven between installation of a fixed camera and reduction in accidents.

What makes you believe that, despite all the thousands of miles of less-than-ideal roads - with sharp and deceptive bends, narrow bridges, blind junctions, poor visibility, bad and slippery surfaces, etc etc etc - that a high proportion of accidents occur on good roads?



How on earth do they then make any meaningful measurement of camera effectiveness when they get, on average, one less fatality in 50 years? You'd have to monitor camera sites for at least 50 years, and at least four times that length of time before you could even begin to separate out RTTM effects.

That's unless you attended the 'Alice in Wonderland' school of statistics.

_________________
Only when ideology, prejudice and dogma are set aside does the truth emerge - Kepler

= headache!

Firstly, I'm not sure if the 5% figure is a fair one to use, but ignoring that for now, in different areas a different percentage of crashes will be down to speed. If it is the case that speed cameras tend to get put on straight roads, then it is quite plausible that more of the crashes on those roads were down to speed.

But also, if speed cameras are stopping more accidents than just those caused by speed than that is a good thing!



But what exactly would count as proof in your eyes? If there is a reduction in the number of accidents after a speed camera is put in, and there is still a reduction after you discount RTM and trend, then it is reasonable to believe that it is the speed camera that is responsible.

I'm not saying that it's a high proportion of accidents in total but crashes where there have been fatal or serious injuries. On roads with apparently good visibility people will drive faster, they still have the accidents but there are worse results because of the speed. If it was a narrow bridge for example I would expect people to be driving slower, so if there was a crash then it would be less serious.



That's not a fact by the way - I was just making a suggestion and I note that you didn't answer the question, "would that not be worth it?"



They spend 14 pages explaining how they estimate the effects of cameras in section 4 in the document.

http://www.dft.gov.uk/pgr/roadsafety/sp ... nscp/nscp/
thenationalsafetycameraprogr4597

http://www.dft.gov.uk/pgr/roadsafety/sp ... nscp/nscp/
coll_thenationalsafetycameraprog/thenationalsafetycameraprogr4598


Edited to break the above links in half.

That's the problem, firstly it is rare that a speed camera is installed without other real safety features. Secondly, if we follow the logic, if a garden gnome was installed and accidents reduced and stayed reduced afterwards, would it be reasonable to assume that the Gnome was responsible?

In order to proove that a speed camera reduces the accidents, one first has to proove that the accidents were caused solely by speed in excess of the limit.

1) Accidents tend to happen where there are hazards
2) Hazards tend to be few and far between on straight roads with good visibility
3) People tend to drive slower where hazards exist, and faster where no hazards exist
4) Hitting a pedestrian, hitting a tree, having a head-on collision, being T-boned by a lorry, overturning your car, driving into a lake or over a cliff are examples of accidents which are very likely to be serious or fatal even at relatively low speeds.

Of course, I'm willing to listen to any solid arguments to the contrary



Not fact, but within the ball-park. "Would that not be worth it?" would, of course, depend on there being any effect whatsoever. How confident are you that the current focus on speed enforcement isn't actually costing lives? If this is indeed te case, then it would be worth scrapping the cameras even if one life was saved.



Have you actually read that entire report, including the appendices (especially appendix H)?
If you give it a critical reading - or ask someone schooled in statistical analysis to do this for you, I'm sure you won't remain so confident about the report's conclusions.

_________________
Only when ideology, prejudice and dogma are set aside does the truth emerge - Kepler

Ha, I've seen you deleting an important thing I said!

It was right in front of that above paragraph. Here it is again: But what exactly would count as proof in your eyes?



So you think you know better than the people who wrote that 160-page report then?

1. Is that actually a fact though?
2. I may be wrong when I say that speed cameras tend to get put in on straight roads.
3. So when they do crash the results are worse.
4. But then you are generally less likely to have those accidents if you are going slower, for example overturning your car.

Now here's a fact:
On urban roads, injury accidents reduce by between 2 and 7 per cent per 1mph reduction in average speed. TRL 421



I certainly haven't read it and I have no intention of doing so either. But appendix H is one part that I have had a look at.



But we then get back to what I said to Odin: But what exactly would count as proof in your eyes?

OK I have no idea what you are talking about here, the only person who can delete something you've said is you. Or are you referring to the edit where I corrected my grammar, would you like me to put the doubled up 'the' back in? As for your bold, you have quoted my answer.

But just to make it simpler.

Answer = In order to proove that a speed camera reduces the accidents, one first has to proove that the accidents were caused solely by speed in excess of the limit.



How do you come to this conclusion, the report does not seek to make a link it is simply reporting on the statistics, the number of pages does not make it any more compelling. I do not think I know better, I am asking a question, which hasn't been answered by anyone, how is this showing that I know better?

I am sorry HJeg, I would normally have edited my previous post, but you seem to object to that. So please don't feel I am bombarding you.

Just saw this:


That figure is only true of the very specific model that was used, it cannot be applied in the blanket way that it is batted around. Using a similar model we could argue that accidents reduce when speed increases, take for example a residential road and a motorway. Statistically more accidents occur on residential roads than on motorways, thus we can establish from these 2 pieces of information that it is safer to drive at 70 than 30, thus increasing speed reduces accidents.

As you can see that statement is clearly garbage, but a similar logic was used in TRL 421, don't believe me check it. Also read Height Kills on this website, it is a good analogy of that statistic.

Fundamentally so. The greater the density and/or degree of hazards, the greater the inherent risk of accidents



All the ones I've ever seen have been on straight roads. Of course that doesn't mean that they're all on straight roads, but it certainly seems to be the case. Anyone know differently?
.


That does not necessarily follow. More to the point though, you haven't shown that there are not far fewer accidents on such roads and, if so, how this could possibly be - seeing that the inherent risk is very much lower.



Not necessarily. I've seen cars overturning at 40mph or less, when they are laterally destabilised by things like clipping a kerb or being bumped by another vehicle. You can run off the road at relatively low speeds for the same reasons, or if the road surface is slippery or you've misjudged the severity of the bend (for example, if it unexpectedly tightens up) Head-on collisions can happen at any speed. Imagine coming around a bend only to find another car on the wrong side of the road and right on top of you. And if you drive over a cliff it doesn't make much difference whether you're doing 10mph or 100mph



Already answered by Odin



Then you'll know that, with all the caveats given about paucity of suitable data, systemic problems, assumptions that have to be made, etc, that the results are inconclusive at best.
From reading this report, I got the distinct impression that the people who wrote it only did so because they were paid to do so, despite that they knew full well that it's a load of old cobblers.
As an aside, did you read the bit (page 111) about figures from 600+ red light cameras being lumped together with those from fixed speed cameras?



What would count as proof that Santa Claus exists?
What would count as proof in your eyes that speed cameras don't live up to their promise?

_________________
Only when ideology, prejudice and dogma are set aside does the truth emerge - Kepler

In my post of Sun Jan 13, 2008 11:08 pm, I wrote this paragraph: "But what exactly would count as proof in your eyes? If there is a reduction in the number of accidents after a speed camera is put in, and there is still a reduction after you discount RTM and trend, then it is reasonable to believe that it is the speed camera that is responsible."

In your post of Mon Jan 14, 2008 1:53 pm, the question had mysteriously vanished.



To change slightly what I was asking, what would count as proof in your eyes that the collisions were caused solely by speed in excess of the limit?

The report is basically saying that speed cameras have cut the number of fatal and serious collisions. What they are saying is that the number of FSCs is lower than would have been had the case had the cameras not gone in.



As far as I can see, the statistics have been put together in the process of writing the report. And clearly they do make the link between speed cameras and reduced collisions. What exactly was your question again? The report shows that the number of collisions is lower than had the speed cameras not gone in. Of course, without having two parallel universes we can never know for sure, but they do go into it some detail.

Don't worry about it.



But I'm not batting this round saying that it applies on all roads. It says "on urban roads" and that's what I'm using to it to back up - my point about the reduction in FSCs by speed cameras in urban areas.