This article

http://www.thetimes.co.uk/tto/public/cyclesafety/article3986796.ece

argues that, on a per mile traveled basis, cyclists in the UK are about as likely to injure pedestrians as are drivers.

This statistic is, I think, very misleading about the relative danger to pedestrians from cyclists and drivers.

Of all the miles ridden by cyclists, I will wager that most are on streets where there are provisions for pedestrians, and where pedestrians are often present.  In contrast, a significant fraction of the miles traveled by drivers are on highways and other roads purpose-built for auto traffic, where either no provision is made for pedestrians, or where pedestrians are very rarely present.  A comparison on the basis of 'miles traveled in the presence of pedestrians', almost certainly would show drivers to be a much greater threat.

The article would have been better if it had merely noted the fact that a significant number of pedestrians are injured by cyclists.  After all, even if pedestrians are far less likely to be injured by a nearby cyclist than by a nearby driver (which I think is pretty obviously true), cyclists DO have the potential to cause injury, and we should remember that.  Unfortunately, by leading with a provocative yet totally BS statistic, this article will merely cause drivers to feel smug, cyclists to feel put upon, and it will influence nobody to be more careful around pedestrians.