Speeding more dangerous than drink driving

[sstein]

According to some traffic police on bbc1 you are more likely to kill or seriously injure someone by driving over the speed limit then you are if you are drink driving.

This is shocking!! DO THESE PEOPLE ACTUALLY BELIVE THIS!!!!!

Cheers

Stuart Stein

[nonegreen]

Actually if you have had 1 unit of alcohol you are less likely to have an accident than if you have had none. From then on though it goes fairly quickly the other way and once you reach 5 units you are twice as likely to crash. In real terms though this is still not very likely. Statistics are a great tool for argueing about issues that are of little or no importance such as road deaths. In reality of course road deaths are affected by vehicle design, road design and user ability. All other issues are irrelavent.

[safespeed]

nonegreen said:

In reality of course road deaths are affected by vehicle design, road design and user ability. All other issues are irrelavent.

I don't quite think you've got the right words there when you say "road user ability". Some road users may have high levels of ability, yet be very poor at avoiding accidents because they use their skills to cut margins to the bare minimum.

So perhaps a better term is "road user quality", where we define "quality" in terms of their accident avoidance performance.

I also think this should lead us to consider the factors influencing average road user quality - many road safety ideas go to influence road user quality - and some of the modern ideas are clearly reducing road user quality.

edited to add: I'd like to make it clear that I'm in complete agreement with your three influences apart from the definition quibble.

Oh wait - we're forgetting another big issue and that's exposure. More vehicle use leads to more death if other things are held constant.

Best Regards,
Paul Smith
Safe Speed
www.safespeed.org.uk

>> Edited by safespeed on Monday 19th April 05:37

[safespeed]

Taking this whole "what influences the number of road deaths" thing to the next level, I've fairly recently defined three terms for predicting road deaths and suggested an equation for the relationship between them.

The terms are:

Exposure (expressed in billion vehicle kms)
Safety culture (effectively average driver quality)
Engineering (better every year for both roads and vehicles.)

More on:

www.safespeed.org.uk/smeed.html

Best Regards,
Paul Smith
Safe Speed
www.safespeed.org.uk

[V8 Archie]

Interesting stuff Paul. Looking at the final graph, what change happened in the mid-eighties? Is that seatbelt legislation?

[nonegreen]

safespeed said:

nonegreen said: In reality of course road deaths are affected by vehicle design, road design and user ability. All other issues are irrelavent. I don't quite think you've got the right words there when you say "road user ability". Some road users may have high levels of ability, yet be very poor at avoiding accidents because they use their skills to cut margins to the bare minimum. So perhaps a better term is "road user quality", where we define "quality" in terms of their accident avoidance performance. I also think this should lead us to consider the factors influencing average road user quality - many road safety ideas go to influence road user quality - and some of the modern ideas are clearly reducing road user quality. edited to add: I'd like to make it clear that I'm in complete agreement with your three influences apart from the definition quibble. Oh wait - we're forgetting another big issue and that's exposure. More vehicle use leads to more death if other things are held constant. Best Regards, Paul Smith Safe Speed www.safespeed.org.uk >> Edited by safespeed on Monday 19th April 05:37

No problem with your quibble Paul, but do you ever sleep?

[safespeed]

V8 Archie said:

Interesting stuff Paul. Looking at the final graph, what change happened in the mid-eighties? Is that seatbelt legislation?

The mid-80s flattening off in the fatality rate is exactly coincident with the economic boom, and is matched by a speeding up in the fatality rate fall in the early 90s slump.

See figure 4 on the page already linked.

Best Regards,
Paul Smith
Safe Speed
www.safespeed.org.uk

[V8 Archie]

safespeed said:

V8 Archie said: Interesting stuff Paul. Looking at the final graph, what change happened in the mid-eighties? Is that seatbelt legislation? The mid-80s flattening off in the fatality rate is exactly coincident with the economic boom, and is matched by a speeding up in the fatality rate fall in the early 90s slump.

Mmm... I'm not sure I see why this would affect the culture factor. Yet your graph 11 shows:

• an arresting of the increase of the culture factor in the late seventies; and

• a decrease of the culture factor the mid-eighties up to 1993.

In fact, now I think about it, it seems interesting that pre-1975 the culture factor is increasing in a period where road safety was constantly improving.

On a more directly (constructively) critical note:
Is the sentence "The Culture Factor is set up so that a value of 100 represents good road safety performance." there purely because 100 is a round number and 1993 is approximately 100? This claim looks a little arbitrary to me. I suspect that it could be used by the bad-hats as stick with which to beat you.

>> Edited by V8 Archie on Monday 19th April 18:05

[safespeed]

V8 Archie said:

Mmm... I'm not sure I see why this would affect the culture factor. Yet your graph 11 shows: an arresting of the increase of the culture factor in the late seventies; and a decrease of the culture factor the mid-eighties up to 1993. In fact, now I think about it, it seems interesting that pre-1975 the culture factor is increasing in a period where road safety was constantly improving.

This is really about the gradually improving rate of engineering vehicle safety - a factor that the simplified equations don't take into account. We're talking about broad approximation here not precision. Any answer should be treated as +/- about 20%, but the broad brush approach is none the worse for that. I have tuned it slightly to give greater precision in recent years. The fast drop in the early 90s seems to be a real improvement associated with the recession, possibly associated with lower leisure mileages in those years.

V8 Archie said:

On a more directly (constructively) critical note: Is the sentence "The Culture Factor is set up so that a value of 100 represents good road safety performance." there purely because 100 is a round number and 1993 is approximately 100? This claim looks a little arbitrary to me. I suspect that it could be used by the bad-hats as stick with which to beat you.

It is an arbitrary number - there are no units of culture - it's useful for trends and comparisons. I could have set it to 1 or 1 million and the resutls would be the same. I thought 100 was nice and round and easily imagined and reported.

Best Regards,
Paul Smith
Safe Speed
www.safespeed.org.uk