The reason driving faster is more dangerous
Discussion
TurboHatchback said:
<snip>
I agree that if you take a busy pedestrianised area and plot a chart of the speed of a car driving through it vs the probability of a collision there will be a clear and unarguable correlation between the two. If however you take a four lane motorway at midnight and plot the same chart it will not show such a correlation until a certain threshold speed has been exceeded, i.e. driving down a motorway at 10mph is not measurably safer than driving down it at 20mph but driving down it at 100mph is clearly safer than at 200mph.
<snip>
Above selectively quoted as I think you're onto something here. Let's do the maths (all figures are estimates but could be collected reasonably accurately with some work)I agree that if you take a busy pedestrianised area and plot a chart of the speed of a car driving through it vs the probability of a collision there will be a clear and unarguable correlation between the two. If however you take a four lane motorway at midnight and plot the same chart it will not show such a correlation until a certain threshold speed has been exceeded, i.e. driving down a motorway at 10mph is not measurably safer than driving down it at 20mph but driving down it at 100mph is clearly safer than at 200mph.
<snip>
Edited by TurboHatchback on Thursday 12th May 18:19
Assumptions - light traffic, say 1030 at night on a weekday, clear weather, no roadworks in an Audi A4 capable of 140mph
1. A driver drives 10 miles at 10mph on a clear motorway - chances of crashing In any one second = 0.0001. Number of seconds = 3600. Chance of crashing on the trip = 0.36
2. A driver drives 10 miles on the same motorway at 20mph - chances of crashing in any one second = 0.0001. Number of seconds = 1800. Chances of crashing on the trip = 0.18
3. A driver drives 10 miles on the same motorway at 60 mph - chances of crashing in any one second = 0.0001 (same as before as the increased speed is offset by the closeness to the likely speed of surrounding traffic). Number of seconds = 600. Chances of crashing on the trip = 0.06
4. A driver drives 10 miles on the same motorway at 120 mph - chances of crashing in any one second = 0.0003 (triple chance of crashing compared to doing 60mph). Number of seconds = 300. Chance of crashing on the trip = 0.09
So, with those assumptions it's only 50% more likely that you'd crash at 120, than at 60 and half as likely than if you did 20...
Not sure what I really mean by this, and there are loads of factors, but inappropriate speed isn't the same as a ridiculous 'one size fits all' approach that he government seems to be taking.
0000 said:
I'm not reading the rest of that, but I assume the existence of this thread means you never did work out what a direct correlation is in that other thread.
The rest of the thread provides a supported argument to show there is a correlation. I'm sure you can come back with a random anecdote not addressing any of the arguments made. Esceptico said:
0000 said:
I'm not reading the rest of that, but I assume the existence of this thread means you never did work out what a direct correlation is in that other thread.
The rest of the thread provides a supported argument to show there is a correlation. I'm sure you can come back with a random anecdote not addressing any of the arguments made. LeftmostAardvark said:
Above selectively quoted as I think you're onto something here. Let's do the maths (all figures are estimates but could be collected reasonably accurately with some work)
Assumptions - light traffic, say 1030 at night on a weekday, clear weather, no roadworks in an Audi A4 capable of 140mph
1. A driver drives 10 miles at 10mph on a clear motorway - chances of crashing In any one second = 0.0001. Number of seconds = 3600. Chance of crashing on the trip = 0.36
2. A driver drives 10 miles on the same motorway at 20mph - chances of crashing in any one second = 0.0001. Number of seconds = 1800. Chances of crashing on the trip = 0.18
3. A driver drives 10 miles on the same motorway at 60 mph - chances of crashing in any one second = 0.0001 (same as before as the increased speed is offset by the closeness to the likely speed of surrounding traffic). Number of seconds = 600. Chances of crashing on the trip = 0.06
4. A driver drives 10 miles on the same motorway at 120 mph - chances of crashing in any one second = 0.0003 (triple chance of crashing compared to doing 60mph). Number of seconds = 300. Chance of crashing on the trip = 0.09
So, with those assumptions it's only 50% more likely that you'd crash at 120, than at 60 and half as likely than if you did 20...
Not sure what I really mean by this, and there are loads of factors, but inappropriate speed isn't the same as a ridiculous 'one size fits all' approach that he government seems to be taking.
It means that if you start out trying to prove something you can invent numbers to make it work! Choosing to use risk based on risk per second but keeping that risk the same at higher speeds means that you are assuming it is safer the faster you go. Hardly an unbiased assumption. Risk per metre travelled would be unbiased. But even then making up numbers is pointless because you can make them up to prove whatever you want. Research of accident statistics show an increase is risk with absolute speed but also relative speed above the speed limit (ie the difference between the person speeding and the majority of traffic). Hardly surprising as speed differences mean passing more cars and a greater potential for errors of judgement by the person speeding or the person being passed. Assumptions - light traffic, say 1030 at night on a weekday, clear weather, no roadworks in an Audi A4 capable of 140mph
1. A driver drives 10 miles at 10mph on a clear motorway - chances of crashing In any one second = 0.0001. Number of seconds = 3600. Chance of crashing on the trip = 0.36
2. A driver drives 10 miles on the same motorway at 20mph - chances of crashing in any one second = 0.0001. Number of seconds = 1800. Chances of crashing on the trip = 0.18
3. A driver drives 10 miles on the same motorway at 60 mph - chances of crashing in any one second = 0.0001 (same as before as the increased speed is offset by the closeness to the likely speed of surrounding traffic). Number of seconds = 600. Chances of crashing on the trip = 0.06
4. A driver drives 10 miles on the same motorway at 120 mph - chances of crashing in any one second = 0.0003 (triple chance of crashing compared to doing 60mph). Number of seconds = 300. Chance of crashing on the trip = 0.09
So, with those assumptions it's only 50% more likely that you'd crash at 120, than at 60 and half as likely than if you did 20...
Not sure what I really mean by this, and there are loads of factors, but inappropriate speed isn't the same as a ridiculous 'one size fits all' approach that he government seems to be taking.
To me and I could be wrong but have been behind a lot of them in the past. The looking at your partner / wife / kids while driving at any speed results sometimes in an RTA.
Apart from a few the result is always the same, drift into the OS lane or the opposing lane or missing the ruddy bend in the road. Like today behind a Zafira (SP?) He was constant in his talking and looking at the passenger but failed to notice the bend. We were doing 30 - 50 in a NSL 60 limit as he kept slowing and then faster due to the inattention.
Unless you need to use sign language your passenger can hear you so why turn your head?
I have traveled most of Canada and Iceland and all the UK without ever looking at my passenger regardless.
I do agree speed may kill but going too slowly will also kill due to people taking chances when the weather and roads are good but frustration of a dwadling (SP?) driver especially one that blocks the overtake by turning into the opposing lane.
On the road WE have a responsibility for everything we do and everyone else, too slow and someone may overtake, to fast we may kill someone.
It is not hard to understand.
Apart from a few the result is always the same, drift into the OS lane or the opposing lane or missing the ruddy bend in the road. Like today behind a Zafira (SP?) He was constant in his talking and looking at the passenger but failed to notice the bend. We were doing 30 - 50 in a NSL 60 limit as he kept slowing and then faster due to the inattention.
Unless you need to use sign language your passenger can hear you so why turn your head?
I have traveled most of Canada and Iceland and all the UK without ever looking at my passenger regardless.
I do agree speed may kill but going too slowly will also kill due to people taking chances when the weather and roads are good but frustration of a dwadling (SP?) driver especially one that blocks the overtake by turning into the opposing lane.
On the road WE have a responsibility for everything we do and everyone else, too slow and someone may overtake, to fast we may kill someone.
It is not hard to understand.
Esceptico said:
I've only recently ventured onto this part of PH. I expected plenty of hate for cameras and defense of speeding but wasn't prepared for some of the bizarre claims being made, such as the speed limit not having any relationship to the dangers of the road. Such claims seem to defy basic physics and my own experience of driving/riding for 30 odd years. It would also make all advanced training (which focuses on identifying and managing hazards) wrong.
I suggest you look at the history of speed limits and they way limits are chosen for various roads. It is all very well using 'your own experience' but that's not necessarily correct. Chat to any group of drivers and you will get differing opinions.
Speed limits don't quite come into the definition of arbitrary, but it is close. Very close. In other words, they often bare no relationship to the number/type of hazards present. I am certain I am not unique in knowing of locations where speed limits have been imposed solely to save on signage.
There is also a pot of speed limits: 20, 30, 40, 50, 60 and national speed limit. So what about a road that is safe at 59?
Today I drove along a dead straight length of road, for over a mile, that had a 40mph speed limit. There were no turnings, no pedestrians, few hazards.
There are also speed limits imposed for social reasons, not to mention those that are imposed politically.
The claim that speed limits have a direct relationship to the number of hazards is truly bizarre. One only has to drive a few miles to realise this.
You suggest that all advanced driving focuses on hazards. I'm not sure that is correct. Police driver training, at least in my time, concentrated on systemised driving, with safety and smoothness following on from there.
However I'm not sure how speed limits being imposed for reasons other than safety would make any form of advanced training wrong.
KevinCamaroSS said:
Speed will obviously have an effect on the outcome, however, based on lots of professional analysis over the years, excess speed is only the cause of around 5% of all collisions. Should we not concentrate on other causes before blindly going down the speed kills route? Improve driver awareness and skills by 20% and you would cut down on more collisions than by cutting speeds by 20%.
This is a point that is often made in these discussions, and in itself is an entirely reasonable one. If it's going to be anything other than a "wouldn't in be nice if the weather was more like Southern California" type wishful thinking, it needs to be backed up by a explanation of how the awareness and skills are going to be improved, and the plan for that needs to be feasible and politically possible.
I expect you can see where I am going with this...
singlecoil said:
KevinCamaroSS said:
Speed will obviously have an effect on the outcome, however, based on lots of professional analysis over the years, excess speed is only the cause of around 5% of all collisions. Should we not concentrate on other causes before blindly going down the speed kills route? Improve driver awareness and skills by 20% and you would cut down on more collisions than by cutting speeds by 20%.
This is a point that is often made in these discussions, and in itself is an entirely reasonable one. If it's going to be anything other than a "wouldn't in be nice if the weather was more like Southern California" type wishful thinking, it needs to be backed up by a explanation of how the awareness and skills are going to be improved, and the plan for that needs to be feasible and politically possible.
I expect you can see where I am going with this...
0000 said:
singlecoil said:
KevinCamaroSS said:
Speed will obviously have an effect on the outcome, however, based on lots of professional analysis over the years, excess speed is only the cause of around 5% of all collisions. Should we not concentrate on other causes before blindly going down the speed kills route? Improve driver awareness and skills by 20% and you would cut down on more collisions than by cutting speeds by 20%.
This is a point that is often made in these discussions, and in itself is an entirely reasonable one. If it's going to be anything other than a "wouldn't in be nice if the weather was more like Southern California" type wishful thinking, it needs to be backed up by a explanation of how the awareness and skills are going to be improved, and the plan for that needs to be feasible and politically possible.
I expect you can see where I am going with this...
Read the first sentance and then realised poster must have brainwashed by the speed police into believing that ANYTHING BELOW the posted limit is safe in all conditions of weather and traffic. Another failing of the speed police propaganda is their ignorance of the difference between EXCESSIVE SPEED and Speed not appropriate for the conditions.
69.9 will be most possibly be safe on any motorway in light traffic in dry weather with sunlight causing no problems. It will be decidedly suicidal and most unsafe on the same road in icy weather with fog causing poor visibility.
Quite simple- drive within the speed limit at a speed that means you can stop in the distance you see to be clear without any harsh braking/evasive manoeuvres.
So perhaps when it comes to deciding on what to charge someone who drives into the rear of stationary traffic on an icy motorway in fog, perhaps the charge should be "driving under the influence of a poor road safety education system" .
I'd suggest another place the OP might like to visit is Safespeed.org.uk, where a lot of the myths on speed and danger and the current road safety driver education system are debated.
Just seen where this is and the question arises- is this best forum ?
69.9 will be most possibly be safe on any motorway in light traffic in dry weather with sunlight causing no problems. It will be decidedly suicidal and most unsafe on the same road in icy weather with fog causing poor visibility.
Quite simple- drive within the speed limit at a speed that means you can stop in the distance you see to be clear without any harsh braking/evasive manoeuvres.
So perhaps when it comes to deciding on what to charge someone who drives into the rear of stationary traffic on an icy motorway in fog, perhaps the charge should be "driving under the influence of a poor road safety education system" .
I'd suggest another place the OP might like to visit is Safespeed.org.uk, where a lot of the myths on speed and danger and the current road safety driver education system are debated.
Just seen where this is and the question arises- is this best forum ?
Esceptico said:
Punishing individuals speeders is not because their actions were necessary dangerous but rather to try to stop a large number of people speeding.
I cherry picked this because I felt like it, and it's true. At this time there is an entirely disproportionate drive to limit speed, and in the last two years the proliferation of cameras to do this is almost incredible, in view of the insignificant number of road casualties we are dealing with (and the fact the massive increase of cameras is on motorways where hardly any of these casualties occur anyway).
The absolute problem with speed is individual driver ability, provided the vehicle is roadworthy. Near enough all drivers think they are good at it. I think I'm pretty handy, and near enough everyone I know tells me I'm pretty handy, so maybe I am. I am a good boy in urban areas but I don't give a damn about the other limits. On Motorways I tend to stick to 100-ish on the clock, although North of 150 isn't a rarity. 150+ on a B or C road doesn't phase me much dependent on circumstance, I'm not Rossi on the bike but sliding anything on 4 wheels about is child's play.
However, I have to share the roads with other people, and this is where caution is required. They don't look, they don't indicate, they have poor lane discipline, and poor hazard perception. My driving has to accomodate the poor driving of others, and it is genuinely poor in many cases.
Then again, no doubt the fact I am speeding absolves all the incompetents of any blame.
If all drivers were forced to drive faster there would be carnage, I have no doubt about that. But ultimately, as the jobsworths say, "It's a limit, not a target". So nobody is being forced to do anything, other than slow down apparently on current form.
Unnecessarily in my opinion.
Serious question - what makes you think that there's any similarity between driving a car on a public road, and playing a computer game where you constantly have hazards flying at you from all directions at a higher rate than the human mind is capable of processing?
Have a think about that the next time you're driving down the road.
Have a think about that the next time you're driving down the road.
Esceptico said:
I've only recently ventured onto this part of PH. I expected plenty of hate for cameras and defense of speeding but wasn't prepared for some of the bizarre claims being made, such as the speed limit not having any relationship to the dangers of the road. Such claims seem to defy basic physics and my own experience of driving/riding for 30 odd years. It would also make all advanced training (which focuses on identifying and managing hazards) wrong.
I've set out what I see as the argument and evidence for the relationship between speed and risk of crashing.
1. First point is that calling crashes "accidents" is very misleading. Crashes are not generally caused by acts of God - almost all crashes are caused by human error. Yes there will be the odd crash because of an unforeseen blow out and the like but the vast majority of crashes are because of driving mistakes. I would split driver error into three main categories:
a) Observation / anticipation errors - failure to read the road and anticipate what is going to happen next (either through not paying attention and/or poor skills and/or slow reactions) - eg not seeing someone coming from the right when turning left and pulling into their path
b) Errors of judgement - failure to make the right decision eg continuing to drive at 70 on the motorway in a thunderstorm
and aqua planning as a result
c) Errors in controlling the car eg not braking hard enough, not swerving around an obstacle, target fixation
Of course, poor driving can exhibit one, two or all three of these at once.
2. The rate of error depends upon a number of factors eg if you are tired you are likely to make more mistakes. However, looking at the first two categories: errors of observation and judgement, how often you make a mistake will be related to how many times you need to make an important observation or make an important decision. This is where the complexity of the road and number of hazards plays a key role. On a straight, empty motorway in the day then there are few if any hazards (something that demands your attention and a correct decision). Consequently the risk of driving on such roads even for poor drivers is low. Increasing speed should also not increase the risk (at least to a measurable amount). On the other hand, driving in town mostly demands constant attention and decisions eg people pulling out at junctions, on coming traffic on your side of the road as they go around parked cars, etc. Increasing the number of hazards, increases the number of opportunities for humans to screw up and given the law of large numbers, eventually someone will make an error that leads to a crash. Where does speed play a role? The faster you drive the greater the demands put on your powers of observation and ability to make the correct judgements. Increasing the difficulty of the task increases the number of errors. This is evident in every other area of life eg look at computer games that rely on reaction speed - as you progress through the game it gets harder as you have to deal with more obstacles at a quicker pace, the early screens at the slow tempo are easy but you die when the pace hots up. Or take tennis. Watch Wimbledon. Do servers get lots of aces on their first serve hit at 140 or on the second serve hit at 100? It is the first serve because the returner has less time to act correctly and get his racquet on the ball. Increasing the speed limit in town from 30 to 40 would inevitably lead to more errors and more errors will lead to more crashes.
3. As well as increasing the number of errors, additional speed will also lead to more crashes because of the third category of error - driver input. Put simply, the faster you are travelling when you are called upon to take action to avoid a crash (brake, swerve, etc) the greater the skill and the faster the action required to do so successfully. Clearly for a given level of skill the higher the speed then the more likely a "situation" will lead to a crash, rather than a near miss.
4. Lastly, the faster you are travelling then the more kinetic energy you have and the greater destructive potential your vehicle has. All else being equal a crash from a higher initial speed will have more serious consequences for yourself and third parties.
In summary, crashes are mainly caused by driver error. The rate of error increases as the difficulty of the task increases.
Roads with more hazards put greater demands on drivers and so they make more errors. Increasing speed also increases difficulty and increases the error rate. Increasing speed on roads with more hazards should show a steeper increase in errors than on roads with few hazards (this is borne out by research). Increasing speed increases driver input error and also increases the severely of crashes. Roads with lots of hazards (eg urban roads) are more inherently dangerous than roads with few hazards eg motorways,
dual carriageways (in terms of consequences, although your risk of a crash on the motorway is lower but the consequences could be more severe as you are travelling faster). A direct correlation exists between speed and risk for crashes for all roads. The strength of that correlation is directly related to the number of hazards. To reduce the number of crashes to an acceptable level whilst trying to keep traffic moving it is therefore appropriate to have different speed levels (via limits) for different roads. None of the preceding is dependent upon what limits society chooses. It only states the obvious that higher limits will inevitably lead to more crashes. How many more crashes and the impact of such crashes are dependent on lots of other factors. Whether additional crashes are acceptable for the benefit of higher limits is a political decision.
No doubt the speeding apologists will come up with nonsense objections rather than trying to deal with each point above.
This reads like a competition to use the words "hazard", "correlation" and "crashes" as many times as possible.I've set out what I see as the argument and evidence for the relationship between speed and risk of crashing.
1. First point is that calling crashes "accidents" is very misleading. Crashes are not generally caused by acts of God - almost all crashes are caused by human error. Yes there will be the odd crash because of an unforeseen blow out and the like but the vast majority of crashes are because of driving mistakes. I would split driver error into three main categories:
a) Observation / anticipation errors - failure to read the road and anticipate what is going to happen next (either through not paying attention and/or poor skills and/or slow reactions) - eg not seeing someone coming from the right when turning left and pulling into their path
b) Errors of judgement - failure to make the right decision eg continuing to drive at 70 on the motorway in a thunderstorm
and aqua planning as a result
c) Errors in controlling the car eg not braking hard enough, not swerving around an obstacle, target fixation
Of course, poor driving can exhibit one, two or all three of these at once.
2. The rate of error depends upon a number of factors eg if you are tired you are likely to make more mistakes. However, looking at the first two categories: errors of observation and judgement, how often you make a mistake will be related to how many times you need to make an important observation or make an important decision. This is where the complexity of the road and number of hazards plays a key role. On a straight, empty motorway in the day then there are few if any hazards (something that demands your attention and a correct decision). Consequently the risk of driving on such roads even for poor drivers is low. Increasing speed should also not increase the risk (at least to a measurable amount). On the other hand, driving in town mostly demands constant attention and decisions eg people pulling out at junctions, on coming traffic on your side of the road as they go around parked cars, etc. Increasing the number of hazards, increases the number of opportunities for humans to screw up and given the law of large numbers, eventually someone will make an error that leads to a crash. Where does speed play a role? The faster you drive the greater the demands put on your powers of observation and ability to make the correct judgements. Increasing the difficulty of the task increases the number of errors. This is evident in every other area of life eg look at computer games that rely on reaction speed - as you progress through the game it gets harder as you have to deal with more obstacles at a quicker pace, the early screens at the slow tempo are easy but you die when the pace hots up. Or take tennis. Watch Wimbledon. Do servers get lots of aces on their first serve hit at 140 or on the second serve hit at 100? It is the first serve because the returner has less time to act correctly and get his racquet on the ball. Increasing the speed limit in town from 30 to 40 would inevitably lead to more errors and more errors will lead to more crashes.
3. As well as increasing the number of errors, additional speed will also lead to more crashes because of the third category of error - driver input. Put simply, the faster you are travelling when you are called upon to take action to avoid a crash (brake, swerve, etc) the greater the skill and the faster the action required to do so successfully. Clearly for a given level of skill the higher the speed then the more likely a "situation" will lead to a crash, rather than a near miss.
4. Lastly, the faster you are travelling then the more kinetic energy you have and the greater destructive potential your vehicle has. All else being equal a crash from a higher initial speed will have more serious consequences for yourself and third parties.
In summary, crashes are mainly caused by driver error. The rate of error increases as the difficulty of the task increases.
Roads with more hazards put greater demands on drivers and so they make more errors. Increasing speed also increases difficulty and increases the error rate. Increasing speed on roads with more hazards should show a steeper increase in errors than on roads with few hazards (this is borne out by research). Increasing speed increases driver input error and also increases the severely of crashes. Roads with lots of hazards (eg urban roads) are more inherently dangerous than roads with few hazards eg motorways,
dual carriageways (in terms of consequences, although your risk of a crash on the motorway is lower but the consequences could be more severe as you are travelling faster). A direct correlation exists between speed and risk for crashes for all roads. The strength of that correlation is directly related to the number of hazards. To reduce the number of crashes to an acceptable level whilst trying to keep traffic moving it is therefore appropriate to have different speed levels (via limits) for different roads. None of the preceding is dependent upon what limits society chooses. It only states the obvious that higher limits will inevitably lead to more crashes. How many more crashes and the impact of such crashes are dependent on lots of other factors. Whether additional crashes are acceptable for the benefit of higher limits is a political decision.
No doubt the speeding apologists will come up with nonsense objections rather than trying to deal with each point above.
Most of this is stuff is obvious anyway - it hardly needs each point to be laboured so heavily, unless you want an "obvious stuff is obvious" type thread to argue about?
Ian Geary said:
This reads like a competition to use the words "hazard", "correlation" and "crashes" as many times as possible.
Most of this is stuff is obvious anyway - it hardly needs each point to be laboured so heavily, unless you want an "obvious stuff is obvious" type thread to argue about?
It should be obvious but judging by many of the comments it doesn't seem that way. Most of this is stuff is obvious anyway - it hardly needs each point to be laboured so heavily, unless you want an "obvious stuff is obvious" type thread to argue about?
Gassing Station | Speed, Plod & the Law | Top of Page | What's New | My Stuff