Thames Valley Speed Cameras

An independent report        

SPEEDING AND ROAD SAFETY

 

We hear a lot of claims about speeding (exceeding the speed limit) and accidents, so what are the facts?

 

Contents

 

1.1 Introduction

1.2 Results of police investigations

1.3 The speed camera puzzle

1.4 Can Police 'prove' after a collision that a vehicle had been "exceeding the speed limit"?

1.5 Can Police accurately determine after a collision whether a vehicle had been exceeding a speed limit?

1.6 Can speed cameras at least prevent 7.7% of KSI collisions?

1.7 Are pedestrians killed or injured due to vehicles exceeding the speed limit?

1.8 If Scotland had a thousand speed cameras, could the lives of 2 pedestrians be saved each year?

1.9 Motorcycles and mopeds

1.10 An example of "spinning" the statistics

 

1.1 Introduction

 

In order to prevent collisions, we need to know why they occur. When a collision results in someone being killed or injured, the Police investigate and fill in a report form (STATS19) accompanied by a contributory factors form. The officers select from the supplied list of possible contributory factors those that may have contributed to the collision they are investigating. The DfT collects these forms, adds up the factors and publishes the results online. Police don't investigate all collisions, and they don't always find suitable factors to assign when they do, so the following information contains only those collisions where the Police attended the scene, recorded the collision, investigated and assigned at least one contributory factor.

 

Despite the speed camera programme starting in 1993 and building to over 4,000 cameras by 2004, exceeding the speed limit was not investigated as a contributory factor by Police nationally until 2005. Also, just because a factor occurred, does not mean it caused the collision nor does it mean the information could be used to prevent other collisions. The real causes of a collision are not simply what happened, they are what went wrong ie what was different to normal in any particular circumstance. On it's own, exceeding the speed limit may only be of limited use for collision prevention because we would also need to know what speed other motorists were doing that didn't crash. The contributory factors do not contain any speed survey data that may have been taken at collision locations.

 

1.2 Results of police investigations can be downloaded from the DfT website: Road casualties in Great Britain.

 

2005 contributory factors: Road Casualties GB: 2005 download "Contributory factors to road accidents (Excel 113 kb)"

2006 contributory factors: Road Casualties GB: 2006 and download "Article 4 - Contributory factor statistics (Excel 229 kb)"

2007 contributory factors: Road Casualties GB: 2007 and download "Article 4 - Contributory factor statistics (Excel 243 kb)"

2008 contributory factors: Road Casualties GB: 2008 and download "Article 4 - Contributory factors (Excel 378 kb)"

2009 contributory factors: Road Casualties GB: 2009 and download "Article 4 - Contributory factors (Excel 378 kb)"

2010 contributory factors: Road Casualties GB: 2010 and download "RAS50001"

 

The contributory factors for collisions from each year are collated here contributoryfactors.xls (includes factors listed for PIA, FSC and fatal) and the results are compiled into Tables 1.1, 1.2 and 1.3.

 

Table 1.1 All collisions
Year	2005	2006	2007	2008	2009	2010	Average
Exceeding speed limit (inc "possible"), % collisions	5.0%	5.0%	5.5%	5.4%	5.1%	4.8%	5.1%
Exceeding speed limit (inc "possible"), number of collisions	7,314	7,258	7,725	7,103	6,594	5,836	6,972
All collisions investigated	147,509	145,798	140,361	131,582	128,185	120,827	135,710

 

Table 1.2 KSI collisions (Killed and seriously injured)
Year	2005	2006	2007	2008	2009	2010	Average
Exceeding speed limit (inc "possible"), % collisions	7.7%	7.7%	8.3%	8.1%	7.5%	7.1%	7.7%
Exceeding speed limit (inc "possible"), number of collisions	1,832	1,900	1,993	1,833	1,622	1,400	1,763
KSI collisions investigated	23,893	24,814	23,884	22,594	21,501	19,663	22,725

 

Table 1.3 Fatal collisions
Year	2005	2006	2007	2008	2009	2010	Average
Exceeding speed limit (inc "possible"), % collisions	12.4%	14.1%	13.5%	14.4%	15.6%	13.6%	13.9%
Exceeding speed limit (inc "possible"), number of collisions	325	381	342	313	301	221	314
Fatal collisions investigated	2,613	2,703	2,538	2,170	1,935	1,620	2,263

 

 

1.3 The speed camera puzzle

 

If the Police conclude that 7.7% of KSI collisions involved a vehicle exceeding a speed limit (Table 1.2) ...

... how can speed cameras reduce the number of those KSI collisions by 42% (national average at camera sites)?

 

Clearly something is wrong so there are 2 possibilities:

 

1.3.1) cameras are not causing those reductions.

1.3.2) the Police are under-reporting "exceeding the speed limit" as a contributory factor.

 

The governments 4-year evaluation report does show (if read VERY carefully) that the cameras did NOT cause the 42% reduction but do Police under-report?

 

The contributory factors tables state: "Includes only accidents where a police officer attended the scene and in which a contributory factor was reported." Therefore all the collisions included in the tables above were investigated by the Police at the scene of the collision and the Police were confident enough in their conclusions that they officially reported the factors that contributed. The tables do not include collisions where Police did not attend or where they did not investigate or where they determined that no particular factor contributed so this may exaggerate all factors, including speeding (ie the real percentages are likely to be lower than shown in Tables 1.1, 1.2, and 1.3).

 

1.4 Can Police 'prove' after a collision that a vehicle had been "exceeding the speed limit"?

 

They don't have to. Obviously if police had to "prove" every contributory factor, then very few would be reported and their investigations would be of little value. The contributory factors form has been very well designed such that Police officers are only requested to give their "opinion", proof is NOT required. This should allow full reporting of all contributory factors.

 

The instructions on how police officers are to report accidents are on the main page (above) (download the STATS20 form).

 

This "STATS20" form states (p10): "The Contributory Factors are largely subjective and depend on the skill and experience of the investigating officer to reconstruct the events which led directly to the accident".

 

and (still p10): "The Contributory Factors reflect the Reporting Officer's opinion at the time of reporting and are not necessarily the result of extensive investigation. Furthermore, it is recognised that subsequent enquiries could lead to the reporting officer changing his opinion. This is not a problem."

 

This clearly states that the contributory factors are the "opinion" of the Police officer, proof is not required.

 

While slight injury accidents may not always be subject to "extensive investigation", we expect serious injury accidents to be investigated more thoroughly and we expect that fatal accidents WILL be subject to "extensive investigation".

 

Also, every Police officer will be aware of current road safety strategies and may pay particular attention to factors such as exceeding the speed limit, drink driving, mobile phone use etc when investigating a collision. As a result, these particular factors would be very unlikely to be "under-reported". Furthermore each "contributory factor" has 2 boxes, the first called "very likely" and the second called "possible" (p10). The figures quoted in all the tables on this page include all the "very likely" and all the "possible" factors. In all accidents where "exceeding the speed limit" was not reported, the opinion of the officer at the scene was that it was not even "possible" that any vehicle was "exceeding the speed limit". As a result, the figures for "exceeding the speed limit" should be the maximum that occurred.

 

1.5 Can Police accurately determine after a collision whether a vehicle had been exceeding a speed limit?

 

If Police were not able to do this, then we might expect to see large variations in the frequency of "exceeding speed limit" as a contributory factor for different Police areas or for different time periods, but this is not the case. All reports seem to show remarkable consistency both over time (6 years of data so far) and in different areas (see 2009 tables above, pages "Country" and "English Regions"). For instance, in Thames Valley, the percentages of collisions that involved speeding were very similar to the national average:

 

Table 1.4	% speeding, all collisions	% speeding, all collisions
Year	Thames Valley	National
2005	4.8%	5.0%
2006	5.7%	5.0%
2007	5.8%	5.5%
2008	5.7%	5.4%
2009	5.5%	5.2%
Average	5.5%	5.2%

Table 1.5	% speeding, KSI collisions	% speeding, KSIcollisions
Year	Thames Valley	National
2005	7.5%	7.7%
2006	6.8%	7.7%
2007	8.0%	8.3%
2008	8.5%	8.1%
2009	7.7%	7.9%
Average	7.7%	7.9%

 

 

 

 

 

 

 

 

 

 

 

Note: Thames Valley figures, Latest Casualty Statistics, P3 for area wide totals and P4 for exceeding speed limit.

Note: Thames Valley figures may be slightly higher than stated because Police may not have investigated every collision.

 

We can see that in the vast majority (92%) of KSI collisions (Killed or Seriously Injured) the professional opinion of the Police officer actually at the scene was that they did not involve a vehicle exceeding a speed limit, not even "possible".

 

1.6 Can speed cameras at least prevent 7.7% of KSI collisions?

 

Unfortunately, probably not. Even in collisions where vehicles WERE exceeding the speed limit, many of these collisions may include stolen cars, criminals, the emergency services, joy riders, foreign drivers, drivers who have not registered their vehicles correctly, illegal / illegible or cloned number plates etc and, in such circumstances, speed cameras may have little or no beneficial effect. Unfortunately the police reports do not separate out from the collisions that involved speeding, those that may have been influenced by cameras.

 

And many of the collisions that had exceeding the speed limit as a factor may have been caused by a driver under the influence of alcohol. Do speed cameras prevent drunk drivers from crashing? Are drunk drivers safe within the speed limit?

 

1.7 Are pedestrians killed or injured due to vehicles exceeding the speed limit?

 

"If you hit a child when driving at:

• 40mph, you will probably kill the child  

• 30mph, the child has an 80% chance of survival

• 20mph, the child is likely to survive with minor injuries"

 

The above is often quoted to justify why speed limits must be rigidly enforced, so how many pedestrians are killed or injured by drivers who exceeded speed limits? The answer to this is known, so why is this information not included in the publicity? This DfT report shows "exceeding the speed limit" as a factor in collisions with pedestrians in GB (P15, Table 7):

 

Injury ............. Total ..... Maximum percentage where exceeding the speed limit may have been a factor

Fatal ................333 ....................... 6%

Serious ........... 3,956 ...................... 2%

Slight ............ 13,627 ...................... 1%

 

We can see that, of all pedestrians killed in collisions with cars, the driver was not exceeding the speed limit in 94% of cases (note: why does that report show figures only for cars? Would speeding be a lower factor for all vehicles?). I cannot find contributory factors for pedestrian deaths for all of Great Britain for other years, but figures for Scotland are published here:

 

In 2005 there were 65 pedestrians killed where 1 involved exceeding the speed limit: ... Scotland2005_0041882
In 2006 there were 61 pedestrians killed where 3 involved exceeding the speed limit: ... Scotland2006_0054406
In 2007 there were 60 pedestrians killed where 2 involved exceeding the speed limit: ... Scotland2007_0078805
In 2008 they seem to have stopped publishing this information.

 

We can see that, in Scotland from 2005 to 2007, (the only years for which figures have been published) there were an average of 2 pedestrians killed each year where "exceeding the speed limit" was a factor (or a possible factor). That's an average of 3% of all pedestrian fatalities in Scotland that involved vehicles speeding (that's all vehicles, not just cars). We see that over 96% of all pedestrians killed in Scotland did not involve any vehicle exceeding any speed limits.

 

1.8 If Scotland had a thousand speed cameras, could the lives of 2 pedestrians be saved each year?

 

1.8.1) The chances that the cameras would be in just the right location (even using a thousand of them) is very small

 

1.8.2) The data doesn't tell us whether those collisions involved stolen cars, criminals, the emergency services, joy riders, foreign drivers, drivers who have not registered their vehicles correctly or illegal / illegible or cloned number plates etc

 

1.8.3) The data doesn't tell us if the drivers were drunk and cameras may be unlikely to prevent drunks killing pedestrians

 

1.8.4) Speed cameras may have negative side effects that are larger than any benefit they provide (effects of cameras)

 

1.9 Motorcycles and mopeds make up less than 1% of vehicle traffic, but their riders suffer 14% of the total deaths and serious injuries on Britain's roads. Is this because motorcyclists exceed speed limits?

 

Report on mototcycle safety at DfT and download "In Depth Study of Motorcycle Accidents".

 

On p35: “Speeding was found to be common among the respondents, with 58% (n ¼ 143) admitting to always, or frequently, breaking the speed limit. The remaining respondents admitted to ‘occasionally’ breaking the speed limit but only when they thought it was safe to do so. Travelling in excess of the speed limit was considered to be a contributory factor in just 3.5% accidents on the motorcycle accident database”.

All of these Motorcyclists admitted to speeding, yet over 96% of motorcycle accidents occurred when no-one was speeding.

 

1.10 An example of "spinning" the statistics

 

Where does "SPEED IS A FACTOR IN 28% OF FATAL CRASHES" come from?

 

1.10.1) This means that the majority of fatal crashes (72%), do not involve "SPEED" as a factor

1.10.2) "SPEED" has a wide definition including excessive or inappropriate speed, much of which is within the speed limit

 

So how many fatal crashes actually involved a vehicle "exceeding the speed limit"? On average 13.9% of all fatal crashes that occurred in the last 6 years involved (or possibly involved) a vehicle exceeding the speed limit (Table 1.3).

 

In other words, in over 86% of fatal crashes, the opinion of the investigating officer at the scene was that it was not even "possible" that the deaths involved a vehicle exceeding a speed limit. In the end, speed cameras may be attempting to solve a problem that is actually much smaller than we are being told, and may also have significant negative side effects.