The venture capital industry likes to talk about the importance of investing in ideas that are and turn out to be both non-consensus and successful. The idea here is that if an idea or opportunity is already consensus, then there's too much money flooding into that space and it becomes too difficult to make money. This is particularly true in venture capital where a select few companies usually end up generating most of the returns. This is a high risk business. Supposedly, even the best VCs end up having to write off a big portion of their deals.
But I don't think that this logic need only apply to venture capital. In real estate development, you are often faced with similar situations. For example, if an area is already consensus -- that is, it is already considered to be highly desirable -- then capital is going to naturally flow into it and land prices will be relatively high. These high land prices might be justified by the revenue side of your pro forma, or they might not be. I know many developers who avoid "core" locations simply because the land is too much and the margins are too little.
On the other hand, if an area is non-consensus -- that is, you're not sure people will want to rent or buy new space in the area -- then the land prices should reflect this. But here's the thing. What you're doing is trading, among other things, a lower land price for greater market risk. Because the non-consensus bet could turn out to be either successful or unsuccessful. People will either want to occupy space here or they won't. And remember, by definition, it being non-consensus means that most people believe they won't -- or at least not at the prices you might need in order to make the math work.