Originally Posted by
tjspiel
From the studies he's presented, you can't really come up with any conclusions about the safety of riding on a sidewalk vs on the streets. I agree with that.
I think it really just boils down to this: If you understand the risks of riding on the sidewalk and take steps to mitigate them, then it can be more safe than riding on the street. To me this means basically getting off the bike at every driveway and street crossing.
There are other occasions when riding on the sidewalk is safer than being on the street. There's a sidewalk near my house that runs along the border of a golf course uninterrupted for about 1/2 mile. I would argue that taking that sidewalk instead of the street running parallel to it is safer, though neither are particularly dangerous. It's not safer for the pedestrians I may encounter and it's always possible that I may get hit by a stray golfball.
There are streets that are particularly dangerous for various reasons for cyclists and there may or may not be sidewalks nearby that are safer. The problem with all the studies is that they don't take into consideration local conditions. I can ride all the way to my job on streets where the speed limit doesn't exceed 30 mph and where the drivers are used to sharing the roads with cyclists and pedestrians. It's not like that everywhere.
It's always possible to find a particular sidewalk that is safer or a particular road that is more dangerous - those are really meaningless exceptions because because cherry picking examples is not representative of the whole. Mile for mile ridden, I think it's been pretty clearly shown in a number of studies, that a given cyclist is safer on a roadway than on a sidewalk and safer by a wide margin.
Finally, one can talk all the precautions one wants to but it's impossible to foresee the unforeseen but it is accounted for in the aggregate of the data. Let's just say that you could have a cyclist who could prepare and anticipate every potential hazard and never have an issue. That cyclist would be no more representative than the uber safe stretch of sidewalk along a golf course that is never used by anyone except (say) you. Either way, those are outliers.
Now, if the OP is referring to a particular stretch of road or sidewalk, then the generalization wouldn't apply (obviously) and this is meaningless. However, that wasn't the way this was presented and it still remains that all the data and studies point to the fact that sidewalks are multiple times more dangerous to a cyclist than are roadways.
J.