Out-of-project close matches

[Itzhak Epstein]

Tom Epstein, Dick Epstein, and Harry Epstein are members of my surname project. Their presumed haplogroup is G. Tom and Dick match on all 12 markers, including 389-2 = 32. Harry almost matches them, but his 389-2 = 31. Nobody else in our project comes close to matching them. They opted to compare their results to the entire database.

Under FTDNA’s “Y-DNA matches” tab, Tom and Dick see 73 12-markers exact matchers, including each other, and only Harry as “Genetic Distance - 1". Harry sees 20 12-markers exact matchers, and only Tom and Dick as “Genetic Distance - 1". I, as the project administrator, can see both relevant flavors of 389-2. How can I show all three of them what their imperfectly matched surnamesakes see?

Short of getting everybody to join Ysearch, the only way to provide these three members with the necessary information to work out their relationship would be to copy the relevant part of the others’ web pages and email it to them.

[efgen]

For 12-marker matches, FTDNA only shows exact matches except when the close matches are in the same project. That's exactly what you are seeing. There is no way for Harry to see Tom & Dick's exact matches through his FTDNA account, and vice versa.

However, keep in mind that 12-marker close or exact matches with people of different surnames are practically meaningless. The best that a 12-marker match can tell them is that they shared a common ancestor thousands of years ago.

So their best option is to upgrade to more markers, preferrably to at least 37, and then see if they have any close or exact matches with more markers. FTDNA does show out-of-project close matches for the 25-, 37- and 67-marker tests.

[lgmayka]

IMHO, there is a very good reason for this which you should not try to thwart. 12-marker matches are very common, and not very meaningful. Many customers, especially with "popular" haplotypes, turn them off entirely. Even those who don't turn them off completely do not want to be bothered with 1-off near-matches (again, at a mere 12 markers). FTDNA includes 12-marker 1-off near-matches only for members of the same project, who presumably have a specific reason (same surname or same ethnicity) to believe they might be related.

Frankly, the real "solution" here is for your project members to upgrade to 25 markers or more. They will then get a much better picture of who they may be related to in a genealogical time frame.

[Itzhak Epstein]

My co-administrator already urged the three to upgrade.

Since Tom, Dick, and Harry’s ancestors very probably did not adopt surnames before the 19th century, geography and genetics are even more significant indicators of kinship. Eastern Europe, as the respondents very well know, is not the British isles. Among the 93 combined 12-marker matches, there almost as many surnames.

According to FTDNA’s estimates, there is a 56% probability that Tom and Dick share a common ancestor within the last 8 generations. There are 48% probabilities that either of them shares a common ancestor with Harry within the last 16 generations. Ignoring Dick, there are several principal possibilities:

1. Tom and Harry are unrelated.
2. Tom’s ancestor mutated from Harry’s.
3. Harry’s ancestor mutated from Tom’s.

Since Tom and Dick have 73 matches and Harry only 20, I would gamble that Harry is more likely to be the mutant. There is also a possibility that Tom’s and Dick’s ancestors adopted the same surname independently of each other.

Of the three, Dick is the most active and interested genealogist and I want to provide him with as much information as possible.

[lgmayka]

Of the three, Dick is the most active and interested genealogist and I want to provide him with as much information as possible.

Perhaps, then, you have sufficient reason to deputize Dick as a 'virtual deputy co-administrator'. In other words, give him the project's kit number and password so that he can see what you see.

[Itzhak Epstein]

Perhaps, then, you have sufficient reason to deputize Dick as a 'virtual deputy co-administrator'. In other words, give him the project's kit number and password so that he can see what you see.

Are there such entities as read-only co-administrators?

[lgmayka]

Are there such entities as read-only co-administrators?

Not that I know of. You would have to trust him enough not to change anything.