I know it is self-selecting response, but I would have to say that the percentages of people being hit are much higher than I would have expected given the sort of figures you read about in relation to cycling injuries. According to these results I would have a 40% chance of coming into contact with a car during the course of 5 years cycling. That makes it a pretty high risk activity to my way of thinking.
Self-selection makes a huge difference! If you'd like some more comforting hard numbers, AUK just published their figures:
about 1 serious injury per 51,500km ridden (of which motor vehicles were only involved in a tiny percentage - so falling off on gravel/ice or riding into trees were more of a threat!)
Obviously this is mainly on quiet roads, but also includes some very tired/sleep-deprived riders.
(not sure what defines "serious", but I don't think there were any deaths that season).