"57% of cyclists jump red lights" screamed the headline on the press release from the IAM.
Perhaps you think that's a little high. Peter Walker of the Guardian's Bike Blog  certainly did. When he asked for a more detailed breakdown of these figures he found that only around 2% of the respondents to their survey admitted to jumping red lights 'regularly'. Over 85% said they never jumped red lights or had done so only once or twice, or 'rarely'.
Of course, as anyone who ventures onto Britain's sometimes fractious road network knows, law breaking happens on all sides. 82% of motorists admit to breaking the speed limit , nearly half talk on their mobile phones  and, in my experience, it is a good day when more than a handful of drivers manage to overtake following Rule 163 of the Highway Code .
Red light jumping is often attributed specifically to cyclists, but in terms of a road safety risk it's much more of a serious problem when drivers do the jumping: of all the pedestrians injured in an incident involving red light jumping in London, 4% were due to cyclists; 96% were due to drivers of motor vehicles .
But it is the reaction from the cycling media, like the Guardian and BikeBiz that is interesting. This press release is tabloid fodder and is knowingly, and in my eyes quite disgracefully, spinning the results to fulfil stereotypes. Faced with such behaviour Peter Walker and Carlton Reid decided to ignore the embargo.
The IAM should know better - it is usually a responsible organisation, which seeks to represent all road users. They even run their own, reasonably good, cycle training and take sensible, conciliatory positions on many road safety issues. Feeding the anti-cycling media doesn't help their reputation in the eyes of cyclists.