I like the P/E approach. How much income will the land generate now? How much will it generate in the future? If you build housing on it, how much would it rent for? How much would that housing sell for (which ultimately is tied to rent). When will that income be available? You have to take into account holding costs.
The price of the land will take into account everybody's opinon on the income generating ability of the land. To get a better than average return, you will need to get a better than average estimate on the future earnings potential. I do not know how to do that, but I can guarantee that polling a bunch of strangers their opinion will not give you the edge over common opinion.
Or you could just buy it and hope that you will find someone who values it even more. But if you are going to do that, you might as well go to Vegas and play craps. You would probably even get better odds.