Safe Speed Forums

from http://slower-speeds.org.uk/

How do people write this and believe it? Even if at 70mph the chances of a crash on my journey are 100%, that means that at 50mph there is a 0% chance? It doesn't make sense! Or am I missing something here?

That's why they say "in general", it's a way of covering their backs when somebody picks at their claims like you just have and proven them wrong.

http://www.safespeed.org.uk/dtlr_spin.html


They likely mean a geometric accumulation:

70 = 100
69 = 100 * 0.95 = 95
68 = (69) * 0.95 = 100 * 0.95 * 0.95 = 90.25
67 = (68) * 0.95 = 100 * 0.95 * 0.95 * 0.95 = 85.74
66 = 81.45
65 = 77.38
60 = 59.87
40 = 21.46
20 = 7.69

I've always taken this to mean that the 5% reduction is applied anew to the current crash risk, rather than being cumulatively applied to the original risk...

i.e.

If at speed X your crash risk is Y, then at speed X-1 your risk drops to 0.95Y.
=> at speed X-2 your risk drops to 0.95*0.95Y = 0.9025Y
=> at speed X-3 your risk drops to 0.95*0.9025Y = 0.857Y
etc.


So whilst the per-step drop in risk IS 5%, the overall drop in risk is NOT 5% * reduction in mph - as you can see above, after dropping 2mph the overall drop is 9.75% rather than 10%, after 3mph it's 14.3% rather than 15%, and so on. Note also that with each subsequent step, the overall reduction diminishes - the first 1mph drop gets you the full 5% reduction, the second drop gets you an additional 4.75%, the third drop gets you 4.55%... thus unless you start with zero risk, you can never reach zero risk using this risk-reduction model.

_________________
Chris

Thank you Twister, well explained. I "get it" now. It's still bobbins though!

IIRC this piece of statistical garbage comes from the cobbling together of several completely dissimilar studies on speed in certain situations. Paul has it somewhere on this site.

On other words, it's complete bobbins, as the previous poster said.

It's probably true based on a simplistic model which assumes every vehical goes the same speed and frustration and tiredness doesn't count.

Not to mention assuming an equal degree of attentiveness, competence and anticipation from all drivers. Can anyone spot a huge fatal flaw there?