Drivers had learned to brake for the speed camera and floor it right after, so Nottingham stopped measuring the moment and started measuring the mile

the problem

drivers braking only at the camera's exact spot, then speeding again

background

what everyone would do

The available lever was adding more fixed-point cameras along a route, increasing enforcement coverage roughly in proportion to how many points had a lens pointed at the road, while still only measuring instantaneous speed at each discrete camera location.

what they saw

Speed Check Services and the councils that bought its equipment saw that the actual problem wasn't insufficient camera coverage, it was that any measurement taken at a single fixed instant could be gamed by braking exactly at that spot and accelerating right after, meaning more cameras would just multiply the number of instants drivers could learn to brake for, not close the gap between them. Rather than adding more single-point checks, the fix was redefining what the metric even measured, replacing an instantaneous speed reading with an average speed calculated across the entire stretch between two paired cameras, removing any single instant left to brake for at all.

the move

SPECS paired automatic number-plate-recognition cameras at two points a known distance apart and calculated each vehicle's average speed over the whole stretch between them from its number-plate timestamps at each end, rather than measuring instantaneous speed at either camera. Driving fast anywhere in the interval, even for a moment, raises the average and shows up in the reading — there is no single instant left to brake for. The first permanent UK deployment went live on Nottingham's A6514 ring road in August 2000.

why it works

Calculating average speed from number-plate timestamps at two points a known distance apart meant driving fast anywhere in the interval, even for a moment, raised the average and showed up in the reading, since there was no longer a discrete instant a driver could identify and slow down for specifically. Because the metric itself covered the whole stretch rather than sampling a point, compliance now required maintaining safe speed continuously across the entire interval, not just performing compliance at a known, predictable location, which is exactly what eliminated the brake-then-accelerate pattern that had let drivers stay technically compliant at fixed cameras while speeding everywhere between them. This redefinition of the metric, not additional enforcement hardware, is why Nottinghamshire's average-speed camera network produced a 65% reduction in casualties and serious injuries across eleven fitted roads, nearly double the roughly 36% national average reduction for the same technology, evidence the average-speed approach genuinely changed driving behavior across the whole road rather than merely at isolated points.

the payoff

Nottinghamshire went on to install the largest network of average-speed-camera coverage of any English county — more than 40 miles of road — and a 2016 RAC Foundation-commissioned study by Road Safety Analysis found casualties and serious injuries fell by an average of 65% across eleven Nottinghamshire roads fitted with the system, against a national average reduction of roughly 36% for fatal and serious collisions after average-speed cameras go in.

where it breaks

The mechanism depends on the two measurement points genuinely being far enough apart that maintaining compliance across the whole interval is meaningfully different from momentary compliance at either endpoint, two cameras placed too close together would barely differ from a single fixed-point check and offer little of the averaging benefit. It also depends on there being no viable alternate route between the two measurement points that a driver could use to bypass the monitored stretch entirely, a road network with easy detours around the paired cameras would let drivers simply route around the average-speed zone rather than actually comply within it. And this specific fix works because the underlying compliance failure was driven by knowing exactly where and when the check occurred, it wouldn't help against a violation with no predictable inspection point to game in the first place, since redefining a point-check into an interval-average only closes the exact loophole created by a subject learning the check's precise location and timing.

what came after

SPECS became the reference technology for average-speed enforcement in Britain, and county after county — Cambridgeshire, Surrey, Norfolk — adopted the same paired-ANPR approach on roads where single-point cameras had failed to stop the brake-and-accelerate pattern; average speed cameras from later manufacturers (Vysionics' VECTOR, Jenoptik's successors) use the identical two-point measurement principle.

references

1. [1]Average Speed Cameras Explained and How They WorkSpeedCamerasUK, 2026speedcamerasuk.com
2. [2]The Effectiveness of Average Speed Cameras in Great BritainRAC Foundation / Road Safety Analysis, 2016racfoundation.org

keep it

same kind of clever

• 1911 · Aalsmeer Flower Auction (Royal FloraHolland)Aalsmeer sells its dying flowers by racing a price downward instead of bidding one upward
• 1911 · Roald Amundsen's South Pole expeditionAmundsen marked his supply depots sideways to the route, not along it, so a lost party could still walk into one
• 2017 · Behavioural Economics Team of the Australian Government (BETA) / Australian Taxation OfficeAustralia's tax office cut inflated deduction claims by writing to the accountant, not the taxpayer