Originally Posted by John Forester

There have been studies of the crash rates of various cycling groups, made in the 1970s, some done by the National Safety Council for school and university-associated groups, some done by club cyclists for other club cyclists. The data show that the crash rates for club cyclists were about 25% to 20% of the crash rates for the other groups, depending on the type of crash (for example, all crashes, or just car-bike collisions).

There were distinct differences in behavior between the two general groups.The general bicycle commuting populations of several N. Calif. Bay Area cities with large (compared to the US in general) bicycle mode proportions, and hence most likely to be better than the US average, scored only a flunking 55% when measured against conformity with the rules of the road for drivers of vehicles. In contrast, the better club cyclists scored around 98% when measured by the same scale.

The Effective Cycling Program taught its students, and tested their accomplishment, in the traffic skills employed by the better club cyclists that were in accordance with the rules of the road for drivers of vehicles. In the absence of other evidence, which has proved to be very difficult to obtain, it is reasonable to assume that the skills learned in Effective Cycling are most likely able to reduce the crash rate of its graduates to about 20% to 25% of that of the general bicycle-riding population.

That is the only claim that I have ever made on this subject.