Thursday, June 9, 2016

AncestryDNA New Ancestor Discoveries

Many view’s “New Ancestor Discoveries” as so much bunk. I’m one of them, or was one. Ancestry has presented me with 22 Discoveries that aren’t in my tree. Trouble is, my tree is full for 6+ generations. Granted, there could be some non-paternal events hiding in there somewhere. But 22 out of 31? Not a single one of the 22 shares a surname with anyone in my pedigree! Like I say; I have my doubts about New Ancestor Discoveries.

One of the New Ancestor Discoveries that has shown up since last I checked is Joseph F. Smith, president of The Church of Jesus Christ of Latter-day Saints from 1901 until his death in 1918.

Four of the Ancestry Insider's 22 New Ancestor Discoveries

While I don’t think I am his descendant, I do happen to know we’re related. I’ve always known I’m a 2nd cousin of the mother of Joseph Smith, prophet and founder of the Church. That makes Joseph F. Smith my 4th cousin (5-times removed).

Relationship chart for The Ancestry Insider and Joseph F. Smith
This chart was produced by BYU’s Relative Finder.

I started to entertain the notion that Ancestry’s New Ancestor Discoveries was confusing cousins with ancestors. Then it hit me: read the fine print. Sure enough, the fine print reads, “These are potential new ancestors or relatives who are not already in your family tree.” (Emphasis added.)

Ancestry has found 18 Joseph F. Smith descendants (or tightly related family groups of descendants) who share enough DNA that Ancestry has called them a DNA Circle. As a cousin, I share significant amounts of DNA with 3 of the 18. But that doesn’t make me a descendant. I have a feeling the same thing applies to my other 21 “Ancestors.”

The Ancestry Insider and the Joseph F. Smith DNA Circle

Ancestry knows they have a problem.

“As DNA Circles get larger and more DNA matches are delivered, more people are connecting into the DNA Circles, which results in more New Ancestor Discoveries, but with a decrease in accuracy,” Ancestry said in a statement. “We are updating the criteria to make it more conservative and increase the accuracy of New Ancestor Discoveries.” The changes are said to “significantly decrease” the number of Discoveries while increasing the accuracy. Before, you had to match just two members of a DNA circle to be considered a “Descendant.” Now, you must match at least three. And for circles with more than 15 members, you must match 20% of them.

For more information, see “AncestryDNA: Improving Accuracy of New Ancestor Discoveries” on the Ancestry Blog.


  1. They may not even be a blood relative, but the spouse of a blood relative. Just like your circles show the parents of the relative you are related through, the DNA in question does not come from both of the parents.

    I have been able to place many of them in my tree. But it was confusing until I realized that most were relatives or spouses of relatives.

  2. If only AncestryDNA would provide chromosome mapping, this would become so much clearer. I actually think this could be useful working through backdoors of brickwalls. But only if you could know which line the match is on which requires chromosome mapping.

  3. I'm not nearly that close being 19 steps from Lucy B. Mack and 20 steps from Joseph Smith, Jr. I'm experiencing some of the same disappointments in all those matches, but occasionally find a real gem like a first cousin once removed. She put a family of only 9 members on Ancestry but I recognized her father's name immediately.

  4. This feature is totally new to me. I have never seen "New Ancestor Discoveries" ANYWHERE on my profile.

    Maybe it would be a good idea to tell us WHERE we can find this information? Otherwise this post is more confusing than helpful.


  5. I too have been frustrated looking at these "discoveries". More work needs to be done to help identify the relationships.

    I wish we could "ignore" these discoveries after we have researched them, or at least hide them for later consideration. Adding comments to them would also be helpful. As it is they just drive me crazy.

  6. Mine come and go..had 41, then they disappeared. then I had 13, now I have 5 or 6. And some of the ones that disappeared I had already glanced at and knew how they fit in my tree, so they were correct...disappointing..

  7. It would be infinitely more clear if they just called it "Relative Discoveries"

  8. I have also had a large number come and go. maybe 30 gone.

  9. Thanks so much for this post. I have been trying to decipher "potential ancestors" for my father, whose DNA is matching him with 14/14 in one DNA circle, 17/21 in another DNA circle, and 8/8 in yet another circle. From glancing through folks in those circles, it is evident they all hail from the same area/county in Alabama circa mid-1800's. Surnames Robbins, Williams, Cantrell. suggests that he is descended from a certain Williams/Cantrell union.

    My dad does have an unknown great-grandmother whose name was Nancy (widow of Howard) and married his great-grandfather in Mississippi in 1885. Family history is that she brought to the marriage a daughter about 10 years old named Jennie Henretta. Nancy died in 1888 after giving birth to my father's grandmother. Her daughter by Mr. Howard was sent to live with relatives who lived in Arkansas. But we know from multiple census records that both the daughter (Jennie Henretta) and Nancy were born in Alabama.

    For 50+ years, our family has been looking for a Nancy Howard and daughter Jennie/Henretta Howard in the 1880 census in Alabama with zero luck.

    When I saw my father's VERY strong connections to these DNA circles, I narrowed my search to women named Nancy Williams, Nancy Robbins, or Nancy Cantrell. I looked for marriage records to a Mr. Howard. I found a Nancy Robbins married to a James Howard (signed by Nancy's father William Robbins) in the very county where all my father's DNA matches occurred. So I then looked for the couple in the 1880 census. No luck. My next step was to see if I could find Nancy Robbins in the 1870 census, same county. I did... with William Robbins listed as her father.

    Thinking that this may be the right family and perhaps Nancy, Mr. Howard, and Jennie Henretta lived nearby in the 1880 census, I followed William Robbins forward to 1880. And there it was... I think the smoking gun. In 1880, William Robbins is listed with his wife, known children from the 1870 census, and..... Nancy Howard, age 26. There is also a 6 year old "daughter" in the household named Henretta J Robbins. It seems likely that Henretta J Robbins is actually Jennie Henretta Howard, as William Robbins and his wife are 60 and 54 years old. My gut says that Mr. Howard had died early in the marriage and that Nancy and daughter Henretta Jennie moved back in with her parents, where they were recorded as "Robbins" on the 1880 census.

    It's taken some digging, and I'm not positive I'm right, but the evidence suggests I am. My concern was that the Robbins family I've found isn't in the pedigree of folks in the DNA circles my dad matches with. Same last name, same county, but no shared common ancestor (yet).

    But thank you so much for your post. I think the likeliest explanation is that the connection is a generation or two earlier.

  10. This comment has been removed by the author.