See, this is the problem with ideologues. Instead of dispassionately observing the data, and drawing conclusions from it, ideologues "put the cart before the horse" and advocate a given stance...then scramble to find data to support it.
Do I believe that bicycles, when operated on roadways, fare best when they adhere to the "rules of the road?" Yes.
Do I believe that novice cyclists (a) over-estimate the risks of "playing in traffic" and (b) under-estimate the risks of sidewalk riding? Yes.
Do I then avoid sidewalk riding completely, and use whatever road is available? Uh, no.
As a commuting cyclist, I use whatever route I determine to be "optimal." That might be the shortest, flattest, most aesteically appealing route...or the route with the least traffic exposure. Also, frequently I can cut distance off a trip by utilizing anything from MUP, to singletrack, to sidewalks, that go direct where cars are routed in a circuitous manner.
A large part of a bicycle's utility is that (with the possible exception of a dual-sport motorcycle) it can be operated over a greater percentage of the Earth's surface than any other mode of transportation. I would never willingly sacrifice so much of that utility to operate "just like a car."