| London, 3 Oct 2003. For immediate release. |
![]() |

| The actual number of fatal accidents for 1978 to 1993. | |
| The number of fatal accidents that would have occured between 1994 and 2002 had the 1978–1993 downward trend continued. | |
| The number of deaths that have occured in excess of this trend. |
"The figures are even worse than we had expected. Regrettably, they are quite clear and undeniable. I wouldn't want to be running one of the so-called Safety Camera Partnerships when this gets out."The partnerships operate both the familiar roadside cameras and the increasingly unpopular "Talivans". The partnerships, made up of police, magistrates, and local councils, have always sought to justify their actions by claiming to save lives. The usual basis for this claim is that in locations where cameras are installed, the number of accidents reduces in the period immediately following their installation. What the partnerships invariably fail to acknowledge is that they place cameras at locations where there have been unusually high numbers of accidents in the three year period preceding the camera installation, and where accident numbers would probably have reduced by a simple process of chance. This effect is extremely well understood and is known to statisticians as "regression to the mean" - the tendency for unusually high numbers of crashes to occur from time to time but not to be repeated year in and year out.
"This makes me so sad and angry. We have been warning for years that speed cameras make the roads more dangerous. How many more will have to die before we return to traditional and proven road safety values based on skills, individual responsibility and consideration? This speed camera madness must end right now"