UK Report Shows Only 2% of Accidents Caused by Speeding

[Major Bloodnok]

s2art said:

Big Fat F'r said:

I've also said that it may account for a rise, as well as a fall.


BFF

Can you explain that, its not obvious. How can a return to the mean account for a rise unless the camera was put in when the incident statistics were unusually low. In which case why put the camera there?

Presumably he means that the rise that triggered the camera placement was RTTM, and that any preceding low rate was a statistical fluke. You'd have to take measurements over several years preceding the camera placement to show that, though.

[Major Bloodnok]

Of course, it's also irrelevant, since the argument is with the claim that speed cameras reduce accidents. Any rise in rate, whether before or after the camera, and whether due to RTTM or another factor has no bearing on the question of whether cameras genuinely reduce accidents. Or at least, it does nothing to support the pro-camera stance.

[Big Fat F'r]

s2art said:

Big Fat F'r said:

I've also said that it may account for a rise, as well as a fall.


BFF

Can you explain that, its not obvious. How can a return to the mean account for a rise unless the camera was put in when the incident statistics were unusually low. In which case why put the camera there?

I was under the impression that some people had suggested that cameras were sited for purely financial reasons. Therefore the camera will be located in areas where only normal variation is occurring.

The incident statistics don't have to be unusually low. If they are in an areas where they are low as part of the natural variation, then RTTM will acount for some of the increase. Just like it will acount for some of the reduction.

It's a point worth making, as some have appeared to believe that RTTM only means a reduction, which as you know is not true.

BFF

[Big Fat F'r]

GPSHead said:

Big Fat F'r said:

Guam said:

I think the question being posed now as to the nature of whom you may work has become relevant? I do concede however that this is a question no one should be pressured to answer.

That being said can you confirm that you are NOT connected in anyway to traffic enforcement?

That way people can decide whether your position may be skewed ornot.

Best wishes

Tom

But why would my position be skewed?

I've said that although I prsonally think there are uses for some cameras, I do not agree with all the placements.

I've also said that I think some limits are too low (and yes, some are too high).

I've said that although I blelieve in enforcement, I would prefer concentration on the real excesses, not the minor infringements.

I've said that although RTTM should be accounted for, I do not beleive that it explains all the claims in all cases.

I've also said that it may account for a rise, as well as a fall.

I've also said that not all SCP's are are corrupt (and not all BiB are corrupt either).

That seems pretty balanced to me.

BFF

I believe that the vast majority of people whose work is not connected with traffic enforcement would have taken the opportunity to say so there.

To quote someone-or-other, it's very difficult to get someone to understand something when their job depends on them not understanding it. Seems very likely to me. How about you, BFF? Presumably, whatever job you have, you're keen to keep it by any legal means?

Like it or not, until you confirm that your work is not so connected, many people (me included) will regard your opinions as quite possibly skewed and influenced by a vested interest. 98% of accidents by definition simply can't ever be stopped by speed cameras, and yet you refuse even to concede that we have too many. Many incidents have shown how low SCPs will stoop and yet you refuse to condemn any of them. You seem to claim to be entirely happy with the current camera situation, and yet you seem otherwise intelligent and rational (not a common combination at all). Something odd's going on.

Something odd is definitely going on.

When you say I refuse to concede that we have too many, do you mean accidents or cameras. I've said on many occasions that we have too many accidents on the road, and I've also said that I don't agree in the siting of all cameras. So where am I refusing to concede that we have too many.

I haven't refused to condemn any of the SCP, I've said that I certainly don't regard them all as corrupt. I'm sure some are incompetent, if not worse. I also think that some teachers are incompetent, some Doctors are dangerous, some entertainers are abusers, some of SS supporters are....whatever. But I will not condemn the whole group.

I have not claimed to be entirely happy with the current camera situation.

So, maybe there is a vested interest that is blinding the reader....but who is the reader? Just how high is that ladder of inference you are climbing.

BFF

[fluffnik]

Big Fat F'r said:

fluffnik said:

No, RTTM is necessarily massively significant with the datasets involved at the resolutions claimed.

No, its potentially massively significant. Where a genuine blackspot occurs, then a real reduction could be observed that would not be attributable to RTTM. So not every change can be put to RTTM.

There are no blackspots with sufficiently high accident rates for RTTM to be anything other than massively significant due to the intrinsic nature of the data.

Our roads are too safe for quick, high resolution accident stats to be possible...

Big Fat F'r said:

Not only that, RTTM can be seen as an increase, not just a decrease. Yet if a camera sight shows an increase in accident rate, not one of you starts spouting off about RTTM. Suddenly it's the camera that makes it worse. It can be RTTM both ways, yet your insistence that a reduction is always RTTM, but an increase is always the camera making it worse, only makes you the same as those you object to, and your intentions suspect.

I think you'll find that Paul and the numerate amongst us are more interested in statistically valid trends rather than meaningless snapshots swamped by noise artifacts amongst which RTTM is one of the most consistently significant.

Big Fat F'r said:

fluffnik said:

There is no open question to resolve, there is a blatant abuse and misrepresentation of statistics by supposed public servants. They are either incompetent or dishonest, in neither case should their employment continue.

Sacking people for incompetence! There'll be no one left in employment.

That might be the case in your industry (what was it again?), it sure ain't in mine...

[safespeed]

fluffnik said:

I think you'll find that Paul and the numerate amongst us are more interested in statistically valid trends rather than meaningless snapshots swamped by noise artifacts amongst which RTTM is one of the most consistently significant.

RTTM isn't really a 'noise artefact' - it's a bias.

The rules for speed camera placement - specifiying a minimum number of crashes before a camera could be sited - have caused a very substantial bias. The RTTM bias alone is of the same order as the total benefit claimed. Personally I believe that RTTM alone is actually larger than the average benefits claimed by DfT.

[Steven Toy]

Remember, it's not the siting of the cameras that really count. It's their accompaniment of (arbitrarily) lowered speed limits - especially those where they cover up the 40 repeaters before eventually removing them altogether and we have to play "spot the difference."

[WildCat]

They all position to make money. In France for example .. on the roads leading to all channel crossing ports . you have the hill. On this hill you see the 90 km lolly ,.. follow by a 70 km lolly... then you see sign which consist of )):.. with immortally ironic words FOR YOUR SAFETY. Then yards beyond .. the grey box with day-glo bordering.

SO .. .. safety never the issue in any currency .

But John Nichol was spot on when he pass comment on this story when reviewing the papers on some satellite news channel. He said that speed was not the cause of these accidents - poor driving skills cause a crunch at any speed ..even at zero reading in the car parks. He went on to say that more policemen, less scams und far more attention to improving actual skills und standards would be far better accident prevention measures.

Methinks John might just be a PH-er und a COAST fan

[Big Fat F'r]

WildCat said:

But John Nichol was spot on when he pass comment on this story when reviewing the papers on some satellite news channel. He said that speed was not the cause of these accidents - poor driving skills cause a crunch at any speed ..even at zero reading in the car parks. He went on to say that more policemen, less scams und far more attention to improving actual skills und standards would be far better accident prevention measures.

Having stated what many people already believe, I don't suppose he went on to suggest how to achieve it did he?

BFF

[andmole]

Guam said:

Big Fat F'r said:

Guam said:

I think the question being posed now as to the nature of whom you may work has become relevant? I do concede however that this is a question no one should be pressured to answer.

That being said can you confirm that you are NOT connected in anyway to traffic enforcement?

That way people can decide whether your position may be skewed ornot.

Best wishes

Tom

But why would my position be skewed?


BFF

No one but the most extreme on here would necessarily assume it automatically was, I for example have been involved in a debate on another section of the site, because I have never hidden who I am and what I do I answered in what I believe was an unbiased way and believed I was resisting "self Promotion" however given the knowledge of who I was and what I do some members believed I had overstepped the mark and was showing some biase and the thread was subsequently closed at member request.

That is actually right and proper as the members could make their own assessment of the posts and either disregard, or reassess and act accordingly.

If that holds true elsewhere on the forum then for TRUE percieved balance, there needs to be some disclosure of interest to allow other posters to make their own decision and render a verdict that we all have to live with (whether you or I agree with the assessment or not)

Cheers

Tom

Well said that man

There's still a question hanging around waiting to be answered on the subject of "interest" when it comes to putting across points of view.

[Dan Friel]

I'm a little uncertain why someone's background affects the status of their postings? There's plenty of posters here who are clearly anti-camera and work towards their removal, but this doesn't devalue their contributions.. I'd suggest that having someone with knowledge of cameras is more benefical than a random punter.

My view on the 2% issue is this. It's probably not something to get too hung up on. When recording causation, the police have to been certain about each factor. In the vast majority of incidents it will be impossible to "prove" that the speed limit has been broken. It'll be easier to record "inappropriate speed" (unless detailed surveys of the scene are made or the guilty party admits to speeding). Considering that a sizeable proportion of UK motorists speed everyday (the details are readily available), then it's likely that speeding motorists contribute to the accident stats in a far more significant way than the 2% stat suggests.

[andmole]

Dan Friel said:

I'm a little uncertain why someone's background affects the status of their postings? There's plenty of posters here who are clearly anti-camera and work towards their removal, but this doesn't devalue their contributions.. I'd suggest that having someone with knowledge of cameras is more benefical than a random punter.

In some circumstances your background can actually legally prevent you from participating in discussions, let alone trying to influence the outcome of those discussions. So yes, a persons background is relevant.

[mr motor]

Safety cameras are for tax revenue not safety, if further proof is required, just look at where they are positioned?

Having spoken to several safety camera operators over the years (and note that most safety camera operators are NOT police officers as perceived by the general public) I was informed that they position themselves for the maximum number of results which I take to mean number of motorists nicked for speeding rather than for reducing accidents.

Have stopping distances increased over the past few years? well I would say yes because we, the motorist behind the wheel are spending far too much time looking at our speedometer and the sides of roads looking out for those 'high visibility' safety camera vans and static cameras that are hidden behind trees, road signs or over the brows of hills, and speed limit signs that go up and down like a yoyo!

As to only 2% of accidents caused by speeding, I don't have access to hospitalisation figures so can't really comment on this figure, but from my own experiences I would say that a total lack of any sort of police presence on our roads is probably the real reason for accidents occuring. Rebuilding the traffic divisions and putting properly trained traffic police back on the roads, and the bobbie back on the beat will do more to reduce accidents, injuries and crime in the UK than any amount of this very silly CCTV paranoia, unless of course the governmental priorities aren't really about reducing accidents at all - OK the last bit said in jest over our current party political goons!

[safespeed]

Dan Friel said:

My view on the 2% issue is this. It's probably not something to get too hung up on. When recording causation, the police have to been certain about each factor.

No. Absolutely not. The system for coding contributory facts requires Police to record a 'confidence' rating with each factor. This can be 'possible' or 'very likely'. In this way the excellent contributory factor system - probably the best in the world - tends to overcome that particular problem.

The 2% figure includes both those factors rated as 'possible' and 'very likely'.

Dan Friel said:

Considering that a sizeable proportion of UK motorists speed everyday (the details are readily available), then it's likely that speeding motorists contribute to the accident stats in a far more significant way than the 2% stat suggests.

No it isn't. We slow down where crashes are most likely (i.e. where hazards threaten) and speed up on clear roads. Most speeding is taking place where hazards are fewer and crashes are less likely.

And, hey, that's EXACTLY the behaviour we need from our drivers.

And don't forget that the 2% figure INCLUDES the nutters, the stolen cars, the drunks and the Police officers on emergency business.

[alphadog]

safespeed said:

Dan Friel said:

My view on the 2% issue is this. It's probably not something to get too hung up on. When recording causation, the police have to been certain about each factor.

No. Absolutely not. The system for coding contributory facts requires Police to record a 'confidence' rating with each factor. This can be 'possible' or 'very likely'. In this way the excellent contributory factor system - probably the best in the world - tends to overcome that particular problem.

The 2% figure includes both those factors rated as 'possible' and 'very likely'.

Dan Friel said:

Considering that a sizeable proportion of UK motorists speed everyday (the details are readily available), then it's likely that speeding motorists contribute to the accident stats in a far more significant way than the 2% stat suggests.

No it isn't. We slow down where crashes are most likely (i.e. where hazards threaten) and speed up on clear roads. Most speeding is taking place where hazards are fewer and crashes are less likely.

And, hey, that's EXACTLY the behaviour we need from our drivers.

And don't forget that the 2% figure INCLUDES the nutters, the stolen cars, the drunks and the Police officers on emergency business.

I think that you also have to compare statistics like this with what you as a driver experience out there each day. Based on my personal experiences over the years (30000 - 80000 miles annually), I think the 2% figure is about right.

[Richard C]

Big Fat F'r said:

fluffnik said:

Big Fat F'r said:

There is no open question to resolve, there is a blatant abuse and misrepresentation of statistics by supposed public servants. They are either incompetent or dishonest, in neither case should their employment continue.

Sacking people for incompetence! There'll be no one left in employment.

That might be the case in your industry (what was it again?), it sure ain't in mine...

BFF clearly earns his keep from the public sector

[CommanderJameson]

Big Fat F'r said:

I’m not claiming its linear. I’m claiming that it’s riskier when you increase speed. For many complex reasons. But still riskier.
BFF

It is indeed riskier. And given that 70%-odd of motorists on the motorway clearly accept that risk (and, given my drive on the A1M yesterday, I'd say it's >80% in the bits where elephant racing doesn't artificially limit the cars) why not give the people what they're taking anyway, which is an increased limit of around 80-90MPH?

It's all very well saying "it increases risk" but the practical reality is that, on motorways at least, millions and millions of people routinely exceed the speed limit by a good 10MPH or more every single minute of every single day and do not crash.

[Dan Friel]

The casusation data subject is an interesting one, and I'd like to get a police view on the data. How does the average traffic cop interpret this?

Speeding locations - I'd argue that most speeding (i.e. highest percentage of total moves) occurs in urban envrionments, exactly where traffic needs to be slowed. The total disappearance of police speed traps in urban environments is worrying.

[Peter Ward]

Big Fat F'r said:

WildCat said:

But John Nichol was spot on when he pass comment on this story when reviewing the papers on some satellite news channel. He said that speed was not the cause of these accidents - poor driving skills cause a crunch at any speed ..even at zero reading in the car parks. He went on to say that more policemen, less scams und far more attention to improving actual skills und standards would be far better accident prevention measures.

Having stated what many people already believe, I don't suppose he went on to suggest how to achieve it did he?

BFF

Once the overall goal is agreed then we can define the means to achieve it. Therefore once we as a country decide that we're really going to address road safety, and that we're going to do it by (1) improving driving skills and (2) having more trafpols, then we can define the means to achieve them.

Perhaps at the same time the country might decide to include pedestrian skills, cycle-riding skills....

How long will it be before the country reaches this point? 5 years? 10 years? "Speed kills" is just so much easier for most people to swallow than "bad driving kills", not least because they only have 1 thing to concentrate on in order to claim "it wa'n't my fault guv, I wa'n't speedin'" when the inevitable accidents occur.

[Nick_F]

The biggest weakness in the 'bad driving kills' approach is that two thirds of us think we're better than average drivers.

Trying to get someone to accept that their driving is substandard is like trying to get them to admit that they're bad at sex...