On FamilyTreeDNA, they have their
TiP Report that gives the probability that two matching members shared a common ancestor within the last number of generations. E.g., for a yDNA-37 match with 4 variations it suggests 90.6% within the last 12 generations. Now for this yDNA test on my side of the match, we know the paternal line with a very high level of confidence back to the 1600s and ten generations due to yDNA triangulation of the paper trails.
Therefore, I'm wondering, could it be this is a false positive? It seems to me that you could have two separate lines that go back many hundreds of years that just happened to mutate and by coincidence end up in the same general area of yDNA and thus appear as a match. I know a NPE is also a possibility, but perhaps this type of coincidence is too? If so, what is the probability of a yDNA match where surnames do not come close to each other being a false positive?