Announcement

Collapse
No announcement yet.

First results in...

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

  • #16
    Originally posted by Seth_Rogoff View Post
    I think my brain is going to explode!!
    Try this first:

    https://sites.google.com/site/wheato...etic-genealogy

    Comment


    • #17
      Originally posted by jbarry6899 View Post
      Thanks will have a look at the link.

      Comment


      • #18
        ah, this thread seems to have a lot of the answers I was looking for when I posted mine... good, I'll just sift through this

        ;-)

        Comment


        • #19
          Originally posted by dna View Post
          Let me exaggerate a little. The one with LB of 30 would only have 20% chance that yours and his DNA are matching by a pure random chance. The one with LB of 15 would have over 95% chance that yours and hers DNA are matching by a purely random chance.
          Do you mean to write elaborate or exaggerate? Could you please supply a source for your claim that a 15 cM longest block shared has over a 95% chance of being a false match? DNA blocks over 7 cM are more than likely IBD segments, it is when you get to around 5 cM that segments more than likely are IBS.

          Comment


          • #20
            Originally posted by travers View Post
            [----]
            As I wrote, I was exaggerating. That made me pull the numbers from the sky. Hindsight is always 20/20, I should have also written: not actual numbers, for illustration purposes only.

            My apologies to all the readers!

            Two points to be made:
            * quality of matches keeps decreasing when the Longest Block gets smaller/shorter;
            * many people experience a drastic drop in quality of matches when the Longest Block drops to 15-20 cM.

            Unfortunately it is all (advanced) statistics with proprietary algorithms

            Warning: actual cM numbers used, however the poster is not a trained professional, but an amateur; use at your own risk; actual results may vary... I do not know how to nicely explain that it is a continuum with the longest block above around 90 cM giving extremely good predictions on number of births separating two individuals, around 4 cM entirely useless, and for the values in between giving in between quality of predictions. And also that the relationship curve that ties these two is evidently not a straight line.

            Let me repeat, for anybody who has thousands of matches, the best strategy is to investigate them starting with the ones that have the longest block above 20cM. Or in other words investigate them in the order Family Finder is displaying them to you (by the Longest Block), however be extra careful in taking for granted the relationship between you and your match when the Longest Block is less than 20.

            Comment


            • #21
              dna

              thanks so much for clarifying, Your answers are very helpful! Yes I thought there was a reason why FTDNA defaulted to the longest block first.

              So in the default FTDNA list order my top 5 are

              (known 3rd cousin)
              2nd Cousin - 3rd Cousin
              Suggested Relationship: 2nd Cousin
              Shared cM: 267.51
              Longest Block: 49.76

              4th Cousin - Remote Cousin
              Suggested Relationship: -
              Shared cM: 72.55
              Longest Block: 34.75

              2nd Cousin - 4th Cousin
              Suggested Relationship: 3rd Cousin
              Shared cM: 87.39
              Longest Block: 27.5

              3rd Cousin - 5th Cousin
              Suggested Relationship: 4th Cousin
              Shared cM: 135.61
              Longest Block: 27.21

              4th Cousin - Remote Cousin
              Suggested Relationship: -
              Shared cM: 87.33
              Longest Block: 26.56


              So the top one IS my 3rd cousin, that's a known match, but going by yours (and another couple of posts I have seen on the forum) I should now start on the next four LB's even though the Shared cM is <100 on 3 of them? The relationship range on them is quite distant, if they are correct, I think the only one I may have any chance of tracing would be the other suggested 3rd cousin? As all my ancestors originally came from Eastern Europe, getting definite paper trails for every one is quite a task!

              Thanks

              Comment


              • #22
                If it were easy, you would not be here

                The best I can recommend, it is to reach out to these individuals. May be they have accumulated some evidence that would be a good starting point. Or you find something in common (let's say both of you have a copy of the same photograph).

                I am aware, that the paper trail in areas beyond current borders of Poland and Lithuania (Central Europe) does exist on occasions, but access - if at all possible - is a nightmare.

                P.S. Family Finder estimates the number of births separating you and your match. Just for our convenience, it displays that estimate as uncles, cousins etc. If you count the number of births in the tree, that might be an easier visualization of how many relatives are missing between you and your match depending where in the tree they potentially could be.

                Comment

                Working...
                X