Interpreting "Shared Matches" on Ancestry - a difficulty

[bugbear]

This is obvious, right? The people in a Shared Match share DNA, they (pace Endogamy) have a shared ancestor. Simples.

This interpretation of matches has worked well for me, and I have managed to build trees (where trees were not available) for many people in shared matches, using other members of the Share Match group as clues.

Until now...

I'd formalised my thoughts into a quick 'n' dirty (Perl on Linux) implementation of the Leeds method, although I had to enter the Shared Match data onto a spreadsheet by hand.

I'll call my base person 'S'. I had a small Shared Match group, which included 'X' and 'Y'.

It was quite confusing, because 'X' also shared matches with 'O's maternal line, and 'Y' shared matches with 'S's paternal line. 

After a fair bit of head scratching and research, I found the answer (and built fully cited trees for both 'X' and 'Y').

To cut a long story short, X,Y,S do NOT SHARE A COMMON ancestor.

S and X share a common ancestor.
S and Y share a different common ancestor.
X and Y are quite closely related, and share yet another (quite recent) common ancestor.

So beware...

   BugBear

[phenolphthalein]

 
NO ONE got it !!! Blood types. Where is the rhesus factor?
pH

[JohnDoe2020]

X and Y share S

[bugbear]

X and Y share S

well, no. There's nothing in common between all of X,Y,S.

They merely form a set of 3 pairs, and each pair has a connection.

I admit it's complicated, which is why

• It took me a while to realise the true answer
  
• I posted it in the hope of helping others
  

  BugBear