Hah! I _knew_ I had Welsh in me. God I love these WIDE and TALL triangles. (triangulating connections) https://en.wikipedia.org/wiki/Isabel_Marshal Her mom (Isabel de Clare) is described as “Cambro-Norman-Irish” and “one of the wealthiest heiresses in Wales and Ireland”. Triangulating FAR OFF cousins is bringing some great results. Only reason I _know_ they’ll be cousins is a TINY piece of DNA shows up but with a different “mtDNA haplogroup” marker than I have. By looking up that haplogroup and picking ANY RANDOM PERSON with, say H1e1a, IF they worked as hard on their family tree as I did and hooked into geni’s “World Tree”, not only WE WILL be far off cousins, but BECAUSE we share such a TINY TINY fragment of DNA and are BARELY cousins at all, the DISTANCE BACKWARDS IN TIME where we’re connected and branched off from each other, will be HUGE. Picking a random H1e1a found us as 28th cousins, which leads to last common momma – in this case, Isabel Marshal, born 1200 AD.

====

 I think it can but only if it’s combined with lots and lots of family trees, which I think they’ve been doing.I think ethnicity estimates can be way off base (using percentages of probabilities) but if you share a small fragment with someone who is FAR away from you genetically , that makes a big triangle.  It’s got a long base.  So your connection as cousins will be many generations back.But, if you BOTH did extensive family trees, they’ll connect.  I’m absolutely floored at how well this is working.

======

 We might not have any shared segments but still be cousins.  If we have different parts of the original DNA, they won’t line up today but as we get closer together up the trees, more and more “parts of the original” would show up.

====

 really?  I know so little about the part of my ancestry.  Hard to trace. I could only get back to 1880s or so.I changed the resolution down to 1 and we match
Largest segment = 6.3 cM
Total Half-Match segments (HIR) = 22.2 cM (0.620 Pct)
5 shared segments found for this comparison.

chr 3, 5, 6, 18, 19

——

 What I’m doing is “spot checking” cousins, jumping around to find commonalities.  So if it shows we are 15th cousins and you’re H1b1-T16362C, I start going to other H1b1-T16362C that are on geni and look for family tree connections in common.

It’s been hooking together really nicely.  this is all following the Rogers / Luster line that was mostly in Connecticut, NY and NJ, but it’s got hooks everywhere from spreading so far.

====