Announcement

Collapse
No announcement yet.

Chances of 3rd party matching 2 confirmed relatives, but NOT through same lines?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • John McCoy
    replied
    It is always possible that the segments that match come from farther back than the estimated generations to most recent common ancestor would indicate. Most of my predicted "third to fifth cousins" are apparently several generations farther back than that and so remain a complete mystery.

    Leave a comment:


  • John McCoy
    replied
    It is always possible that the segments that match come from farther back than the estimated generations to most recent common ancestor would indicate.

    Leave a comment:


  • Tmason
    replied
    Thanks, yes - you understood correctly.

    C matches both of them, but does not recognize any of their shared ancestors.

    Another person - let's call them D - matches both of them, does not recognize any of their shared ancestors, but does recognize an unshared ancestral surname of just one of them.

    Leave a comment:


  • loobster
    replied
    If I understand correctly, situation is:
    A matches B. C matches both A and B.

    It is totally possible that C is not part of the shared family of A and B.

    Even if Chromosome Browser shows C matching on the same chromosome segment - still, could be C matches A on one version of that chromosome, matches B on the other version of the chromosome - so it may not actually be the match it looks like.

    Leave a comment:


  • Chances of 3rd party matching 2 confirmed relatives, but NOT through same lines?

    Well, seven long years after I first got Family Finder tested - and I admit I pretty much lost interest after the first two - I finally got a match (out of more than 3500) where we could reconcile the paper trails!

    Although I hadn't even known she existed, I realised as soon as we compared that she is descended from my great grandparents - my mother's father's parents - and it turned out we are second cousins once removed - actually within FTDNA's prediction of 2nd to 4th cousins, so all very good and exciting.

    After we had helped to fill in some of the blanks in each other's trees, I naturally started to wonder what else we could do with this new information of a confirmed relationship, and realised I could use "in common with", to see who matched us both.

    There were only two people in all of my thousands of matches who match us both - so I thought the odds were good that we could work out how these two were also related to us.

    It could only be through the ancestors we already had in common, right?

    But to my great disappointment, although I was able to supply nine ancestral surnames that I and my "new" cousin have in common, the two third parties denied sharing any of them. I don't mean, of course that they were lying - but they didn't recognize any of the names at all, or even have a similar name, that might have got corrupted.

    Oddly, though, one of the third parties did have a surname match with one of my cousin's ancestral surnames, but not with mine - i.e. they had a surname match with one of my cousin's ancestors, who is related to me only by marriage - definitely no contributor to my DNA.

    So I'm just wondering how likely it is that someone who matches two kits which each match each other does so only by chance - i.e. no shared ancestry with both, but matching one through one line, and the other through an entirely different one?

    I suppose I should say at this point that I and my newly found cousin are both from the UK, and our common ancestors are all from the UK, so it would have been a smaller gene pool than if we were talking about the whole world, or even the whole of the United States. The third party matches are in the U.S, though.

    I've no idea whether, statistically, the "in common with" matches might simply be noise.
Working...
X