[cb18]

most of those deaths come from inexperienced riders

Similar things can be said for driving, in that motor vehicle accidents are concentrated in certain segments of the population, (and people having unlucky interactions with those segments.)

Though this niggle of the statistics is more of an issue with bicycling statistics because the bicyclists who know more or less what they are doing get lumped in with the accident rates of people like the "DUI bicyclist" riding home after dark with no lights into oncoming traffic, or terrible tragedies of people driving carelessly through neighborhoods and running over little kids on their bikes.

Then consider, even if we had statistics showing moderately higher injury/death rates for comparable travel styles with a bike opposed to a car, the overall mortality rates in the bike vs car cohorts would no doubt be reversed because of the physical activity vs sitting factor, and likely lower stress levels of the cyclists. Physical activity tends to reduce stress.