15.2% increase in fatal + SI crashes at Welsh speed cameras

[Dave Finney]

New report by Professor Maher finds:

OVERALL: a 15.2% increase in fatal and serious collisions at 238 speed camera sites all across Wales.

the above overall result can be separated into:
1) a 13.6% reduction at 61 fixed speed camera sites
2) a 29.9% increase at 177 mobile speed camera sites

The report uses the FTP (Four Time Periods) method.
http://speedcamerareport.co.uk/04_rtm.htm

The FTP method is the most accurate available because it is the only method capable of completely excluding the effect of site-selection (also known as RTM) from the final results.

http://discovery.ucl.ac.uk/1474743/1/Alternative%2...

This new report therefore questions the report by GoSafe (the speed camera operators).

Both reports (Maher and GoSafe) use the FTP method. The difference in results is because the FTP method was not applied accurately in the GoSafe report (they only assumed when SSP (Site-Selection Period) might have been) whereas Professor Maher identified when the SSP actually was.

By applying the FTP method accurately, Professor Maher has produced the most accurate speed camera reports (London and Wales) by anyone within authority.

There are now 4 reports that use the FTP method to it's full accuracy (most recent first):

Wales speed cameras, M Maher
London speed cameras, M Maher
Thames Valley Fixed speed cameras, D Finney
Thames Valley Mobile speed cameras, D Finney

[Dave Finney]

cmaguire said:

What I would want to know is how the mobile increase actually rates against the actual presence of a van rather than the possible presence.

That is a very important question. When similar increases in fatal and serious collisions occurred at mobile speed camera sites in Thames Valley, the 2nd recommendation was:

http://speedcamerareport.co.uk/08_mobile_report.ht...
"8.3.2 An independent investigation is recommended to determine why there has been an increase in the number and severity of collisions and casualties following the deployment of mobile speed cameras and to establish whether there were more collisions when the mobile speed cameras were on site (a direct effect) or when the sites were unattended (an indirect effect). The author requested data on the number of collisions recorded by mobile speed camera operators but was advised by the partnership that such incidents were not logged. Further research could compare dates and times of collisions with dates and times of mobile speed camera operations. Also, Police collision investigation reports could be analysed to determine how the factors that contributed to collisions changed at these sites following the deployment of mobile speed cameras."

[Dave Finney]

AyBee said:

And what was the increase/decrease in traffic at the same points?

Good question. The short answer is we don't know, none of the reports have that data.

The longer answer is that any change in traffic flow is just one of many factors that might influence collisions rates at the speed camera sites. Other factors might include changes in: vehicle safety features (ABS, ESC, SIP etc), the state of the economy, cost of fuel, vehicle type (cycling rates) etc. These factors tend to influence collisions rates across the entire area therefore area-wide trends can be used to compensate for all of these factors combined.

That is how changes in traffic flow (and other factors) were compensated for at the speed camera sites in the reports.

[Dave Finney]

Ken Figenus said:

Thanks for the info Dave. Good to have fact to back instinct.

You're welcome, Ken.

If anyone has any questions regarding the evidence presented in the above 4 speed camera reports, just ask and I'll see if I can answer them.