This comes up every now and then. I'm really not sure there is as much NIMBYism in SF relative to other cities as many people believe. By the way, by "not really sure", I mean not really sure. I just view it as more of an open question than many people here on HN do.
Here's where I'm coming from - SF is small, old, and largely built. It's in the middle of a very large metropolitan area, but it includes many of the old parts that have had high population densities for a long time. Contrast that with the population densities of most American cities, and the general land area. With the exception of New York, SF much smaller and more densely populated than almost all of the regions used in these comparisons.
Now, when someone builds a large housing development far from the urban center of, say, Houston, that counts as development for Houston. When this happens in Walnut Creek, it goes down as not developing in SF.
So the way I'd answer this question - and I haven't yet - is to look at subregions in major met regions in the US that have had a population density of above 17,000/sq mi for more than 50 years, and try to measure how much development has occurred within those areas in the last 20 years, or what kind of opposition to new development happens in those regions now.
My refutable hypothesis is that SF will no longer show a dramatically different pattern of NIMBYism from those sub-regions outside San Francisco. That doesn't mean it won't show NIMBYism, or that it won't show more NIMBYism than other old, relatively dense regions - I just think that it may reveal that "population density over 17000/sq mi for more than several decades" correlates with general opposition to new construction regardless of whether it happens in SF.
Like I said, I haven't checked, it's just an idea. This would actually take some work to investigate.