GreenShed wrote:
Your reasoning is poor and unsupported by any fact.
Fact: speed cameras have not been proven to be effective when accounting for the three obvious confounding factors (RTTM, long-term trend, bias on selection).
Fact: the well-established casualty trend significantly deviated when SCPs were brought into road safety policies.
GreenShed wrote:
Your yellow regression assumption
It is not an assumption.
The yellow bar is a 'least squares fit' on the curve from 1950 until 1994 with the width of the bar representing 95% confidence intervals.
http://www.safespeed.org.uk/sscw.htmlGreenShed wrote:
placed on the chart you have supplied is naive and shows, should you be able to understand the mechanisms behind road safety performance, that you would expect the number of deaths to accelerate in reduction as the numbers become lower.
No, it wouldn't have to anyone able to understand graphs
The Y-axis is logarythmic. Don’t you know it is proper practice to use a log axis when showing exponential curves?
I thought you were an engineer?

The graph indeed shows "
to anyone with a semblance of understanding of performance figures that there is
a diminished return in improvement as the numbers reduced, especially when you consider the rise in the volume of traffic that this chart takes into account.".

GreenShed wrote:
What can be seen in the later part ofthe chart is an exponential decay,
Wrong again!
The whole curve is exponential. The final part shows a much lower rate of exponential decay.
GreenShed wrote:
this is what I would expect to see from the imposition of a series of measures that cause a road safety imporvement, an exponential decay that has a series of marked steps as improvements are applied.
This exactly what has been demonstrated from 1950 to 2000, then to a lesser extent afterwards.
Next you'll be saying en extrapolation of that curve shows the fatality rate must become negative at some point in the future

GreenShed wrote:
If no improvements were made to control traffic collisions and casualties in 1997 to the present day then the chart would have reversed and the situation would have been worse.
Even you would be able to see that.
No I don't. I would expect that curve to level if no more improvements were effected (this is so obvious).
The fact is, we were seeing that relatively strong exponential decay until the SCPs showed up, then that curve tended to flatten out (rate of decay substantially reduce) even though all the other safety improvements continued unabated.
There has been a significant deviation from the well-established long-term trend at 2000, and only 1 factor changed during that time: the existence of the SCPs.
Only SCP support staff wouldn't be able to acknowledge that.
See, you really are dizzy
