I don't think that duplicates should be closed if the questions are different enough where the resulting answers and discussion will be different.
The reason for closing is marked as "Exact duplicate". Not similar or very close, but exact. All too often questions are closed just because the title will be very similar, when the questions themselves are in no way exact.
An example may be that I recently asked about the gender pay gap, asking not if it exists but asking asking about the evidence that leads people to believe it does exist.
I was linked to this question which my question was considered an exact duplicate of.
Except that the questions are in no way an exact duplicate, with the first questions asking if something is true, and my questions wanting to examine the evidence and see if the conclusion that x is true is valid. Asking why is often different from asking if.
Jeff Atwood already advocated duplicates being left open when the questions or resulting discussion are different enough, which is something that makes a lot of sense.
Why on earth should the policy be different on the Skeptics site?