I like the look of the hill's placement on the horizon, even if the alignment is a fluke. I would say it isn't one, though. I'm not saying it meant anything massively significant, but it both the hill and the pretty sunset could have been the reason that the location was chosen.

Every site has to be assessed on its own merits. If, once you've considered every site, there is a pattern then you have a general theory. If there isn't a pattern to particular aspects this does not make the observations at each site invalid - it just means there's no pattern in the data you've collected. You may get local patterns amongst monument types that are global.

I was at a coastal Neolithic citadel in Spain the other day. As we approached I could see a whirlpool about 50m offshore. Was the citadel built here because it overlooked this wondeful phenomenon? Was it built there because it was next to the sea? I dunno, because I don't know what the coast here looked like when it was built. However, I do know that overlooking this whirlpool there is a great simulcra. I assume that this has been there for a long time. There were no such simulcras at the other similar settlements we visited. So, is the presence of the simulcra at one site a fluke because there aren't any at other sites?