Trying to Make Ancestors Out of NADs

Ok, so color me dense, but I’m trying to figure out exactly HOW one would go about making an ancestor out of one of Ancestry’s New Ancestor Discoveries (NADs).

Bear with me while I work through this, and maybe you will have some ideas, because frankly, I can’t figure it out. I’ve had absolutely no luck with this.

If there was a chromosome browser, and given that I’m mapping my DNA segments to ancestral families, I would be able to see where these folks fall – and identify a family group by where they match. But since Ancestry has no chromosome browser, I’m in the dark about how to turn an NAD into an ancestor.

Like probably everyone else, my NADs have varied over time. Some have come and gone, and come, and gone. I have been pretty vocal about the relative uselessness of NADs, but with the recent new more refined NAD algorithm, I thought, perhaps, just perhaps, I might find something resembling a hint that I can use.

Keep in mind that yes, I am a 35 year plus genealogist, so my tree is already fairly robust. That’s one way I know for sure many of the NADs couldn’t possibly BE ancestors, because all of the slots in that timeframe are already full and proven.  At one point, someone asked me how I knew, so I wrote about how I had proven each generation in my tree, by paper. Many have been subsequently proven genetically as well utilizing triangulation.

After Ancestry’s recent revision, I’m graced with four NADs. Three are entirely new, and one remained from before the update.

New NADs

So let’s take a look at these NADs, but first, let’s look at Ancestry’s revised NAD creation criteria.

Ancestry’s New NAD Criteria

Ancestry’s new criteria for NADs is:

Previously, you needed to match at least 2 members of a known DNA Circle to be given a New Ancestor Discovery. Now, users must match at least 3 members of a small (15 members or less) DNA Circle to be given a New Ancestor Discovery. For larger DNA Circles (16+ members), users must match 20% of that Circle to be given a New Ancestor Discovery. For example, if there is a DNA Circle of 10 people, you will need to match at least 3 people to get a New Ancestor Discovery. And if there is a DNA Circle of 30 people you will now need to match 6 people instead of 2.

Now, let’s look at each NAD and see what we can determine.

NAD 1 – Robert Shiflet

Unfortunately, one of my NADs is still Robert Shiflet. The reason I have so many matches in common with him is because his wife is the sister of my ancestor, and several descendants have tested.  I wrote about this here.

Shiflet NAD chart

In the case of the Robert Shiflet Circle, I match 4 of 6, so clearly he is NAD material, even though he is absolutely positively NOT my ancestor.

NADs 2 and 3 – William Sullivan and Hariet Nickels

Let’s move on to William Sullivan and Hariet Nickels, which, according to a compilation of 355 Ancestry trees, were married to each other. (I’m sorry, but that ‘compilation of 355 trees’ makes me shudder.)

NAD Sullivan

This couple is from South Carolina and Georgia, locations where I don’t have any ancestors, but their offspring made their way to Tennessee, where I do have ancestors, but no dead ends in that timeframe.

The William Sullivan DNA Circle includes 14 people other than me, and I match 5 of those individuals.

NAD Sullivan Circle

There are three Ancestry tools to utilize for each person in the Circle:

  • Pedigree and Surnames (matching trees)
  • Shared Surnames
  • Shared Matches

Each of these tools are available by clicking on the link to the matching individual in the Circle.

NAD tools

I checked each of these three tools for all of the matches, and in one case, I found a common family surname. By looking at that link, I know that we do indeed share a common ancestor in the Dodson line.

A second person seems to also be related to the Dodson family through one of the wives lines, Durham.

A third person descends from the same Dodson line as the first person. He obviously does not have his Dodson line far enough back in time, but having worked with this family for decades, rest assured, it’s the same line.  Thomas Dodson born in 1681 in my line is the grandfather of “Second Fork” Thomas in my matches line and the common ancestor of both lines.

NAD Dodson

I utilized all three tools and could find no discernable link to the other two individuals that I match in the tree.

You can also look at the trees for the people in the Circle whose DNA you don’t match, but who match someone else in the Circle. This didn’t produce anything relevant either.

My strongest match in the NAD Circle is to the individual who also descends from the Dodson line. I checked shared matches with him first, hoping that he and I would both match someone with a leaf tree link in my match list, but unfortunately, there were no matches to anyone with a leaf tree link to me, which would have, of course, told me immediately at least the identity of one common ancestor.  Three of 5 matches have no tree and a fourth has just a minimal tree, so there is no help here at all.

NAD no shared match

Unfortunately, the best I can do with these two married NADs is to say that the only commonality I can find with some of the group is a link to the same Dodson/Durham family.

NAD 4 – Henry Garrett

Nad 4 is to Henry Garrett who was married to Nancy Farris, according to Ancestry and 179 compiled trees.

NAD Garrett

My Faires line, also sometimes spelled Farris, was from Washington County, VA, as was Henry Garrett’s wife, Nancy Farris, according to Ancestry.

So, my first thought is that we connect through the Faires/Farris family line, and that may be true. But I’m glad I didn’t stop there.

NAD Garrett Circle

In the Henry Garrett Circle, there are a total of 8 individuals plus me. Of those, two of the groups of family members connect to me through the Andrew McKee Circle where we are all members.  The third individual that I match had the Faires/Farris connection  also matches my McKee cousins.

I was confused, until I looked at the common surnames with the third person, and look what I found:


Yep, a McKee ancestor who also lived in the same location. I don’t know how Mary McKee connects, but it’s likely that she does, given his matches to me and all of my McKee cousins.  It just so happens that some of my McKee cousins also descend from Henry Garrett.

Since all three of my matches in the Henry Garrett Circle also have McKees in their trees, two of those proven to my line, and the third from the same location – I’m guessing here that my Henry Garrett NAD is really a McKee connection, perhaps with some Faires/Farris thrown in for good measure. 

NAD Summary

So, in summary, none of the NADs are my actual ancestors, but are connected in some other way.

Name of NAD Common Line
NAD1 Robert Shiflet His wife is my ancestor’s sister.
NAD2 William Sullivan Married to NAD 3, three of 5 matches have common Dodson line.
NAD3 Hariet Nickels Married to NAD 2, three of 5 matches have common Dodson line.
NAD4 Henry Garrett 3 matches of which 2 are two family groups of individuals who are in my Andrew McKee Circle. The third match also had a McKee ancestor in the same location. Henry Garrett also married a Farris who may be related to the Faires family from the same location and who are my ancestors as well.

The Question

So I’m still back to the same question I started with. How would I actually work any of these back to prove they are an actual ancestor?  So far, none of the NADs are ancestors, and these all seem to be connected via a spur of some sort, or “spuradically.”  I know, bad pun.

Let’s look at my actual Circles of proven ancestors to see which ones of those would qualify to be NADs, if I didn’t have them listed in my tree as ancestors.

Circles – Proven Ancestors

I created a Circle Chart to see which of my confirmed ancestor Circles qualify to become NADs.

Of my 21 DNA Circles, only one has 16 or more members if you count family groups as 1 and not the family group members individually. Two have more than 16 if you count individuals in family groups separately.  Family groups consist of people that are closely related, such as siblings. In the chart below, I have counted groups as “1.”

Generations Ago means counting me as generation 1, how far back in time does this ancestor occur in my tree.

My Matches – Total Circle shows the number of matches I have to circle members, and the size of the circle, counting family groups as only 1.  In the first example of Jane Dobkins, there are two total in the group, and I match 1 which is a family group, not an individual.

NAD Qualifications shows whether this Circle should qualify to be a NAD if I didn’t have this ancestor is my tree.

NAD Created shows whether a NAD was actually created for this Circle when I replaced my current tree with a very small tree that only included my parents and grandparents.

Circle Name Generations Ago My Matches -Total Circle NAD Qualifications NAD Created
Jane “Jenny’ Dobkins 6 1 group of 2 matches total No No
Daniel Miller 6 3 of 5 total Yes No
Elizabeth Ulrich 6 2 of 5 total No No
Jacob Lentz 5 2 of 5 total No No
Fredericka Moselman 5 2 of 5 total No No
Fairwick Claxton 5 2 groups of 3 total No No
Agnes Muncy 5 1 group of 2 total No No
Andrew McKee 6 3 of 4 total, of those 3, 2 are groups Yes No
William Harrell 5 1 group of 2 total No No
Mary McDowell 5 1 group of 2 total No No
David Miller 5 1 group of 2 total No No
Rachel Levina Hill 4 3 of 3 total, one of which is a family group Yes No
Jotham Brown 6 1 group of 6 total No No
John Hill 6 1 of 2 total No No
John R. Estes 5 2 groups of 3 total No No
Nancy Ann Moore 5 2 groups of 2 total No No
Henry Bolton 5 3 of 11 total Yes No
Nancy Mann 5 6 of 17 Yes Yes
Joseph Preston Bolton 4 3 of 7 Yes No
Joel Vannoy 4 4 of 4 Yes No
Phebe Crumley 4 4 of 4 Yes No

These Circles are all confirmed to be my ancestors. It’s unclear how Ancestry would “count” individuals in family groups relative to creating NADs.  In the chart above, I counted a family group as “1” because that’s how it’s shown, but I suspect that even through Ancestry groups the family group together, they are counting the group members separately.  The reason I think this is that some circles only have two members total, plus me.  I don’t match both other individuals, but in every case, I do match the family group, which consists of at least three people.

NAD Jenny Dobkins

On the main Ancestry DNA page, this Circle is shown with 5 members, which counts the family group members individually.

NAD Jenny Dobkins main page

I decided to do an experiment and I linked my DNA results to a much smaller tree consisting of me, my parents and grandparents, to see how many of my Circles would actually become NADs.  This is where a lot of newbies begin, so let’s see what the newbie experience would be, relative to NADs and which NADs really could be turned into ancestors with enough research. 

Reverting to a Newbie

By connecting a very abbreviated tree, I have put myself in the same position as a new person who just knows their grandparents names – or that of an adoptee, except adoptees don’t even have that much information. They are truly flying blind.

Let’s see what the newbie experience is like. After giving Ancestry enough time to cycle through the process, about three days, just to be sure, my Circles disappeared, of course, which I fully expected and is appropriate because there is no one in my tree beyond two generations.  Because there is no common ancestor in a tree, Circles can’t form, but NADs can form, and should, from some of those Circles.

So what happened?

The same 4 NADs remained, which is exactly what should have happened of course. I expected that too.

However, what I very clearly didn’t expect was for only one new NAD to appear, out of my 21 total Circles and 8 Circles that clearly met the NAD qualifications.  Only one Circle became a NAD – Nancy Mann.


I fully expected at least A FEW of my previous Circles to become NADs. Eight Circles appeared to be qualified based on Ancestry’s stated NAD criteria, but only one actually turned into a NAD.  Even the 100% group, Joel Vannoy and Phebe Crumley, where all 4 people in the Circle matched each other for some reason didn’t become NADs.

Of the 5 NADs granted by Ancestry, we know that the original 4 are incorrect, and we know that the one NAD created from Circles that I had with my robust tree is accurate.  This is what a newbie would see.

How would a newbie ever go about telling the difference, except by beginning to work the genealogy backwards in time from their grandparents. And in this case, they will only be able to “hit” one of 5 NADs, because only one is an actual ancestor – 4 are false positives, red herrings or maybe hints, but only hints if you have a robust ancestry to figure out WHERE that hint resides – an advantage a newbie wouldn’t have.  And frankly, none of those hints were one bit helpful.

Given this situation, where 4 of 5 NADs are wrong, are NADs useful at all or are they exciting distractions leading people down dead-end paths?  I feel particularly bad for adoptees who have no information to utilize to try to build backwards to connect with their NADs.


In the case of an adoptee, they can’t build backwards from any known family, so they would have to contact a group like and utilize special methodologies developed by the adoption groups that match groups of people with common ancestors in their trees.

During one of our conference calls, one of the Ancestry folks talked about how excited adoptees are to see a list of NADs. For many, that would be their first clue as to their family history or genealogy, and their first connect to their family, ever.  I’m sure it would seem like a gift from above.  Of course, adoptees wouldn’t have any Circles, because they are hunting for their ancestors and they don’t yet have trees.

I couldn’t help but wonder when the Ancestry representative made that comment how many of those NADs are accurate – and if that adoptee is embracing people as ancestors who are somehow connected to them, but not their actual ancestors.

Not being an adoptee, I know how hard it is to saw branches off of your family tree when you’ve proven your own work to be incorrect, or the work of another in whom you had confidence (or if you’re a newbie, that tree you copied) and I’d hate to be the one to have to take that NAD (or 4/5ths of their NADs) away from an adoptee, because it’s not really an ancestor.

The sad part is that while I have enough information to determine that 4 of 5 NADs are incorrect – the newbie or adoptee doesn’t.  They just have to go on faith.

Common Segments

It’s common knowledge that Ancestry does not give us a chromosome browser. I routinely use segments to prove a common ancestor, or at least an ancestral line.

In one case, we had oral history that Marcus Younger’s wife was a Hart.  Sure enough part of the Younger group matched individuals from the Hart family dead center in the middle of a Hart triangulated segment.

Here’s an example of what this kind of triangulation looks like.

NADs triangulation example

These particular segments are triangulated to the Hart family and triangulated to the Younger family as well, meaning that all of these people match each other on this segment, as well as me, so this is as much confirmation of Marcus Younger’s wife being a Hart as we will ever receive, short of a Bible turning up on E-Bay. The county records where this family lived no longer exist, so we were left with family rumors and later, DNA.

I keep waiting for a Hart NAD to appear. That’s one I could really embrace.  However, it’s quite far back in time, 8 generations.  Would a Circle or a NAD even be formed?

NADs are formed when you match multiple people in Circles who have a confirmed common ancestor. A Circle has to exist before NADs can be formed.  How are Circles formed?

NAD and Circle Formation

First of all, you have to have enough people matching each other to create a Circle or a NAD. That means it’s unlikely that you’re going to have Circles in the closest few generations – because there just aren’t enough descendants of your grandparents, or maybe even your great-grandparents to create a Circle, which is required before the creation of a NAD happens.  My closest Circles are my great-great-grandparents, the 4rd generation counting me as generation 1.  I do have leaf matches to the 9th generation, but only Circles to the 6th generation.

Second, leaf matching and Circles don’t go beyond 9 generations, so if the common ancestor is beyond that in your tree, you won’t get a matching leaf, a Circle won’t be created, and neither will a NAD. That’s really unfortunate, because I think a lot of us really do carry family DNA that is recognizable from that long ago.  We see it routinely elsewhere.

Third, Ancestry creates what they call confidence scores and Circles are created based on confidence scores. They don’t tell us exactly how these confidence scores are created, but in their white paper, they do tell us that more distant matches have lower confidence scores which is also confirmed by looking at the last page of my “leaf” match list. It appears that Ancestry does not display matches below the moderate confidence level.

NAD confidence level

Based on my Circles shown in the Circle Chart, the new person is only going to receive Circles or NADs for generations 4, 5 and 6.

I have matches through generation 9, and in some cases, 10-12 “leaf” matches in generations 7-9, but no Circle has been formed, which causes me to wonder if anyone has Circles between generations 7 and 9?

Being Alone and Right Means No Circle

This past week, I discovered that my ancestor whose name has been believed for years to be Fredericka Moselman was Fredericka Ruhle.  Actually, her baptized name was Hanna Fridrika Ruhle.  I now have her baptismal record, and her marriage record to Jacob Lentz, both confirming her surname.  I corrected her surname on Ancestry to Ruhle, and boom, I’m gone from the Circle.  And Fredericka has not shown up as a NAD.

So, now I’m left with a quandary.  The only way to see who else is in the Fredericka Moselman circle is to change her name back to the erroneous surname.  Or, in this case, to look at her husband’s Circle which is identical to hers.  However, if I correct his name too, I’ll be thrown out of that Circle as well.  If a former Circle doesn’t appear as a NAD, I have no way of viewing Circles that aren’t connected to me.


Back to the Question

I think we’ve come full circle (pardon the pun), and I still have my original question. How does one go from seeing a NAD to proving that NAD is an ancestor?  We can’t do it with DNA at Ancestry because we don’t have a chromosome browser.

If you have identified a NAD as a direct ancestor, or even used a NAD that was not an ancestor to find your way to a new ancestor, please tell me how.

And I hope, I really hope, it wasn’t just by copying someone else’s tree – because if it is – you’ve very likely just copied the cumulative errors of many – especially if they copied someone else’s tree, who copied someone else’s tree, etc. Tree copying is the equivalent of a genealogical social disease.

Did you simply use the NAD as a hint and pursue traditional genealogy to prove the connection? Was the ancestor the person actually listed as the NAD, or a different person?  Do you have proof in the form of documentation?  And by proof, I mean proof that the documentation is actually for this particular person.

I only mention this because I’ve seen so many conflated trees where someone took any documentation by the same name and added it to their tree as proof.  Let me give you an example.  A man who lived in North Carolina and from the census, was born in Virginia, was not naturalized in the state of New York, a location where he never lived.  Obviously a man born in Virginia had no need to be naturalized.  Same name does not mean same person.  Just saying.

If you turned a NAD into an ancestor, did you track from the NAD forward in time to you, or from one of your lines backward in time to the NAD? If so, how did you know which line to track backwards?  Did your match or matches from the NAD circle download their DNA to either Family Tree DNA or Gedmatch where you could utilize chromosome matching?

If you’ve had success turning NADs into ancestors, please let me know and explain how in the comments.



I receive a small contribution when you click on some of the links to vendors in my articles. This does NOT increase the price you pay but helps me to keep the lights on and this informational blog free for everyone. Please click on the links in the articles or to the vendors below if you are purchasing products or DNA testing.

Thank you so much.

DNA Purchases and Free Transfers

Genealogy Services

Genealogy Research

42 thoughts on “Trying to Make Ancestors Out of NADs

  1. I had 26 NAD circles, half of which were related to each other. Another NAD was the same surname as my 4xs grandmother of which I didn’t know who her parents were. I started a separate tree for each of these two lines and worked the line by documentation. No tree copying, although I checked their public tree for records not found on line, such as Bible or book records, until I came to someone on my tree. One group was the lineage of the wife to the son of a brother to my 5xs grandfather. The 2nd one ended up as the parents of my 4xs grandmother. Now that was a fabulous find! After I found the connection and worked the tree, I found out I couldn’t merge the tree into my tree. The first one, I just added the wife and her parents names on my tree. The 2nd one I had to reenter the information onto my tree. So, working the 2nd tree was worth my time and energy. The 1st one, not so much. (I am new to Ancestry DNA, having previously tested at ftdna and 23andme and on gedmatch.)

  2. I have written in my blog, about my New Ancestor Discoveries, both before and after the recent changes. The article from before the change is called “New Ancestor Discoveries,” and the one after is called “New Ancestors from Ancestry?” I’m certainly not as experienced as you are with DNA, so forgive any of my mistakes. A couple of NAD’s have given me a clue to a relationship I already suspected, but I’ve not had any luck tracing back to a NAD cold.

  3. I have had one NAD that ended up being my 4th great grandmother’s great nephew. It was a very easy relationship to trace because his surname was the same as hers, plus his given name was common in the family. It took less than an hour of poking around to identify him and our relationship. This was pretty useless, since I already had his father in my tree.

    I do have some new nads I haven’t explored. Doing so is quite low on my priority list, since the ancestry tools are so hit or miss. More and more, ancestry seems to be moving toward razzle-dazzle, and ignoring serious efforts toward both traditional and genetic genealogy.

    Thank you for this very interesting investigation. I was thinking I was the only one having little luck with nads!

  4. Roberta, My experience is pretty much in line with yours. NADS coming and going—all the ones I can identify are second spouses or sibling to someone in my tree. I can see that they may be helpful but to date I have found them a wild goose chase and just ignore them—until a itme when I have nothing better to do.

    The CIRCLES on the other had are pure gold…however as I mentioned elsehere I have lost a couple of further back circles in favor of their offspring—that is a 3rd great grandparent circle went away to be replace by his son a second great grandparent—I hope his father will return—there was something like 18 in that circle—so I love them but I don’t like their ephemeral nature so much…

  5. Heck Roberta Ancestry staff has enlightened me today that I can be in and have a DNA circle for my 2nd great grand uncle and his father as well. I thought those DNA circles were for ancestors only. You have any DNA circles for your aunts, uncles and cousins? Ancestry says I can.

      • I am in a circle of seven members, four of whom I match, where the “ancestor”, according to the circle, is the “husband of 2nd great-aunt”. The 2nd great-aunt in question is my great grand aunt (is that the same as a 2nd great-aunt?), the sister of my great grandfather. I actually do have a blood relationship with this man, but it is three generations further back. I do not have a circle for my 5th great grandfather, who is the second great grandfather of this man. Possibly not everyone has their trees that far back. I was surprised when he showed up, not expecting circles with other than direct ancestors. Oh, well!

    • I think the error Ancestry makes is in calling the NAD an “Ancestor” discovery. I have found many of the NADs proposed for my husband and his cousins to be on lateral lines. (Note: I do not have any NAD myself.) I don’t think that is a bad thing though as any prudent genealogist will work more than just the direct line. One should follow the siblings, aunts, uncles, etc because some people create more records than others and their records may apply to your ancestor or provide you with indirect evidence. All DNA matches need to be documented with traditional genealogical research as much as is possible. There may be instances that paper documentation does not exist.

      It is my recommendation that every DNA match in your Circles be added to your tree. This often in a sense is adding the lateral lines so that when you are doing the traditional genealogical research you have other names to research. You may also want to add those that match someone in the circle if you know how they are connected.

      I also recommend that the NADs be put into your tree even though they are not an ancestor or connected to you yet. At the beginning, they will be a loose “twig” that may some day connect to your family in some way — not an ancestor, but possibly a lateral line relative.

      Too many people focus only on the direct line which is a mistake when doing traditional or genetic genealogy. With DNA evidence now I believe our “Family Trees” will turn into “Family Webs.”

      I believe the NADs and Circles can be very helpful. I do think Ancestry should provide more information about how they expect people to use these tools.

  6. Prior to ancestry’s change, I had a NAD with about 8 people. I eventually determined that the wife of the NAD person was actually a distant cousin. About half of the group had the wife’s last name incorrect. Since ancestry’s change, I am now a member of the DNA circles for each of the wife’s parents and no longer a member of the nad for their daughter’s husband. So I’m now in a dna circle for a 1st cousin 4x removed and in a dna circle for the wife of a 1st cousin 4x removed. There are 40 members of the dna circle and I match 8 of them. Looking at the details for the 8, only 2 identify the common ancestor. Seems like this should be two NADs instead of two dna circles.

  7. Several months ago I had Pheobe Crumley listed as an NAD. One of my dna circles is William Vannoy who is my 4th great grandfather. I started with my 5th great grandfather Francis Vannoy married to Millicent Henderson and followed their children down. I found Elijah brother of William and then his son Joel. Once I had found Joel I found Phoebe Crumley born in 1818 in Virginia. Although I am not related to her except through marriage she still showed up as an NAD. That was back when I had about 15 of them. I now have only 5 and she is not one of them. I found her when I was filling in my tree.

    I also have an NAD and one of the matches is a dna and tree cousin of mine. Peter Bloyed is in her tree. I sent her a message on facebook since we are friends on there figuring I will hear from her sooner there than if I messaged her on ancestry. I am guessing it is through marriage.

    I was just trying to look at another NAD and as I am looking I could not get the info I was looking for because it said I must be looking when they rolled in an upgrade. Not sure what it is but unable to find anything I want on ancestry right now.

  8. I have had similar experiences with NADs; they usually do not fit into any gap in the time-frame identified. If I do not pick up on a common surname just from viewing the other people in my NAD DNA matches’ trees, I usually start examining each of their shared matches to me and hope to find a few that have notes on them where I had previously identified a probable or known cause for our DNA match. This can sometimes point me to the ancestral line of interest, even though I still don’t know the exact source of the match without further examination. I have in the past chosen a few of these folks from NADs and started a tree for them, knowing that the common ancestor cannot be the one named, but thinking it must be further back up their tree (possibly further up mine also). However, this method has so far led to me doing a lot of genealogical searching on other folks’ family trees where I finally just gave up on finding a common ancestor that way because it was not looking promising. I suspect that one or both of us is missing some collateral people that would explain the cause of the match, although naturally, we should have upstream ancestors in common somewhere if Ancestry is even close in their DNA matching. In many cases, we lose track of a female daughter in our tree who either died or married where a marriage record cannot be found. Many of these court records (especially in the south) are burned or missing/damaged and will never be found. I have so far had no luck with the new NADs and finding some of these dangling collateral lines. Although a recent one involving what might be a common surname of Winchester looks promising. I have one line that ended at Anne/Anna Winchester born in KY, who I could never find with parents and positively identify. I was seeing a Winchester trend recently and hope it leads to something in the future to get past that unknown wall. Under the old NADs, I WAS able to spot connections on a very few NAD tree matches due to the fact that I recognized lines that were collateral to my tree but where I had worked with the MRCA and their descendants enough to recognize where their tree stopped short of identifying the true source of the match. Those NADs where I could determine a definitive match are now gone (possibly the cousin relationship was too far away from me).

    I finally gave up on one of my ancestors who has always been in my tree with just the given name, Lucinda, born in 1808 in NC or TN (probably that disputed area where boundaries shifted). I sort of decided to “go with the majority” for matching purposes. I knew she married a Gilliam and had quite a few Gilliam children on the 1850 census, and her Gilliam husband died before the 1850 census when he would have been identified with his wife and children. She later married a Severs, presumed to be the hired man from IL listed in her 1850 household (likely the older of two probable brothers). The older “brother” presumably also had children and was a widower, based on children with no wife listed and the young age of the “brother.” He was dead before 1860 and the marriage is only known because Lucinda had three Severs children on the 1860 census. She died sometime between 1860-1870, leaving her children orphaned. My ancestor was one of these three Severs children. Many folks presume Lucinda to have the maiden name Owens since a TN marriage record with a female Lucinda Owens and a male Gilliam exists in the right time-frame in TN (the right state where the first child and some subsequent children were born also). However, the names are all fairly common and I remain unconvinced due to the inability to find a Gilliam census in 1840 in AR where the number of children lined up anywhere near close enough to what we know. I admit the lack of an overwhelming amount of Owens in my DNA matches’ trees bothers me, although it does not disprove the “Owens” theory. Perhaps the “Owens” descendants have just not tested in high enough numbers. Because of the overwhelming number of people who have accepted that Lucinda is an Owens, I could not get leaf hints due to lacking the surname in my tree. Just recently, I did go in and add “owens” in quotation marks and all lower case to indicate that I was uncertain that was correct. I did pickup additional leaf hints, but no circles at that time. One way to get around your Ruhle issue might be to list her as Fredericka “moselman” Ruhle. I highly suspect this would pull up matches under both surnames. I have done this with a few of my lines with “morphing” surnames. Some seemed to change every few years or at least every generation or so. I often make notes concerning the surname in birth information in Ancestry, or else write a “story” and attach it if I feel it needs a long explanation. It would be nice to know how Ancestry handles hint/NADs/Circle matching when we use quotations, several surnames, variants, etc. They must have “rules” that are used concerning surname matching and hints/circles.

  9. Two of my NADs are a man and his wife. He is the great grandson of my 5th great grandfather. It does say at the top of the NADs that these are “potential new ancestors or relatives” and he is my 2nd cousin four times removed. I do have six more that I can’t figure out.

  10. I’ve always viewed both the circles and the NADs as silly gimmicks, thanks for doing the proof on that. I guess you don’t have to give people useful things like chromosome browsers if you can spit out small shiny objects from a genealogical “magic box” that requires no work or intuition to discover.

  11. Well this is interesting Roberta that your understanding of DNA circles the same as mine. So there is something else for you to investigate. Maybe you can get a better answer to this than I have today. I have a screen grab of the 2nd great grand uncle DNA circle if you want it. I have been on the phone with two different AncestryDNA support techs today and received a definitive it is a glich in the system from the first one, and the second danced so much I asked to be transferred immediately to the support survey so I could blast them. When I asked that tech if it would be worth her time to put all her ancestors siblings in her ancestor only tree to see these DNA circles I got crickets.

    To add insult to injury now Crista Cowan has weighed in and says yes, you can get a DNA circle for a 2nd great grand uncle. When I asked where that was written in their helps or which video covered that topic, I was told that her written reply to me was were it was written. I won’t comment here what I said in my genealogy office.

    Keep in mind here I am NOT talking about NADs, but the DNA circles. Something just doesn’t seem right in this, again!

    The last problem I queried them about was how I could have a NAD and neither of my parents had those NADs on their pages. I had 17 NADs on my page and my parents only had 8 of them. Hard to have a NAD and your parents do not. I think this whole thing is a mess. Like you Roberta have asked to have a chromo browser and have had it fall on deaf ears publicly and privately. IMHO as far as I am concerned these DNA circles and NADs are no better than an un-sourced genealogy. After all these years I am having my doubts about certain aspects of their DNA product, tech support, and customer commitment, and I have been with them as a full subscriber since Dec 2000. I am not a happy camper.

    • I would suggest referring them to their own paper on DNA Circles where it says “DNA Circles attempt to quantify the strength of each members’ connection to a given ancestor as well as to other descendants of that ancestor.” And this, on page 3: ” A DNA Circle is a set of likely descendants of a given ancestor and is generated by combining pedigree and IBD information across the entire AncestryDNA database. A DNA Circle is always in reference to a particular ancestor. Each member of a DNA Circle has the ancestor in his or her online pedigree and shares DNA identical-by descent with at least one other person in the circle.” I’m surprised about Christa, of all people.

    • I was just thinking that if your parents actually had some distant ancestor(s) in common (knowingly or unknowingly) and you received a portion of your distant ancestors’ DNA from one parent, and a portion from the other, you could theoretically (I think) match people in a DNA circle whom your parents did not match with a long enough segment. That is the only scenario I can imagine at the moment.

  12. You have done a lot of work and still have questions. I find the NADs almost useless. I too recently did some of what you have demonstrated here. Most of my NADs seem to be related in some way to spouses of family members or to be connected to people in my Circles, otherwise, they seem to have no obvious connection to my ancestors apart from marriage. This makes me question if Ancestry is actually using DNA or if, instead, they are just finding Tree matches. It’s frustrating and annoying. After months of what seem to be useless NADs, I finally got one that may have an actual genetic relationship to me or rather to the spouse of a cousin’s ancstor and that spouse has a surname that is in my tree. I therefore have at least one spouse of a cousin who may also have a genetic relationship to me of some kind. So, I used Ancestry’s “Search” to see what I could find. Not much at all. The NAD only appears in 2 or 3 matches and those are to FTs and not to any record, not even the census! I do think there is a connection. The surname and home of the person are matches to a surname in my tree and to an ancestral home of one of my ancestors. I am still in the dark but at least this single NAD has some potential. Next, I must try some inventive spelling and maybe one or two other tricks.

  13. Roberta,
    Your comment about the nad’s not fitting the time frame that you have slots for is very apt. We have some LONG generations in my family. I’m the “baby” of a large family and so were both my parents, so think and average of 40 year generations. My grandfathers were born in 1866 and 1868. Both grandmothers were born in 1872. All four had died years before I was born. All of my great grandparents were born before the Civil War. NAD’s tend to be born between 1800 and 1900. Most of the latest group of NAD’s were born after my great grandparents and some after my grandparents were born. And these later ones have surnames I have never heard! Obviously they are not my ancestors. To find ancestors for me they will have to actually look at my tree and go back before the Revolutionary War. But that isn’t how they choose them. Apparently the problem is that they are only based on having a few matches to descendants and not on any empty slots in your tree begging for ancestors to fill that slot.

    I also have the gg uncles and aunts as nad’s. Of course that is because their descendants are some of my closest cousins. And we have been recruiting people to test! Lots of matches there!

    My biggest complaint about nad’s is that they totally leave out many of my family lines that I have dozens of matches for. All of my dad’s paternal line is missing and all of my mother’s ancestry except a Ross nad. That one is not an ancestor but it helps me find some of my Ross matches. It is the only useful nad, I have.

  14. Hi Roberta,
    I can tell you from an adoptee’s view point that it’s not easy. I have pages upon pages of people I can’t connect. So primarily I do the same thing with ancestry that I have to do with ftdna & 23andme. I research all my close cousins hoping to fill in some blanks. I don’t waste my time with moderate matches at all or anyone that has no tree. If I can get a gedmatch number then at least we can move forward. In the meantime I have multiple trees to research back & forward for these matches. In the notes I specify maternal or paternal, the number & length of segments, matching surnames and locations, and if I find someone in their tree that matches one of my known surnames. It’s daunting though because as you know the NA research is tough enough. On the flip side, most NA are so admixed that the “IW” side is easier to find.


  15. Thanks for another well researched, thoughtful post. I continue to be one grumpy customer who doesn’t give a rat’s fanny about NADs or Circles. Total waste of resources.

    On the other hand, if Ancestry would even put something as simple as CH2 89209101 – 121337298 (19.135 cM) next to the “Amount of Shared DNA” data hidden under the little grey logo, I could enter it into my own spreadsheet. I could include all the Clough’s or Potter’s or Frasier’s or… and at least see who overlapped. I know it wouldn’t be perfect but I would at least have a chance to determine which line my brick-wall grandma is most likely descended from. That research would be a hell of a lot more productive that researching NADs IMHO.

  16. I am a newbie at the DNA tests confirming the paper genealogy. I have 3 different endogamous groups in different areas of the world, none of whom intersected until my parents met. (Within each group, it’s a mess, particularly when there are only a dozen or so surnames in a village of a couple of hundred people!!).
    Are the NADs in sort of like a FAN club, until one goes into the actual details?? Possible nice clues, but no definite brick wall solutions??

    • Yes, I believe that the NADs are like the FAN that Mills refers to in her talks and journal articles. I believe that it is important to add them as a loose branch in your DNA tree. My recommendation is that each person have a “DNA tree” connected to your AncestryDNA results. This tree will have your own relatives in the tree as well as the loose “lines” that are a result of adding your DNA matches from NADs to people in your DNA Circles. Most people are so focused on the direct line that they forget they are related to the lateral lines.

      It is a lot of work to research other people’s direct lines, but not everyone has the interest or time to do this. In order to get the most out of your results you need to do this.

      I agree it would be very helpful if Ancestry would give us the segment data too so we can see the actual genetic segment where we match and we could do our own triangulation work. However, they have chosen not to do so at this time and we should use the information they are providing (i.e. DNA Circles and NADs). It is also essential that you put the “Shared Ancestor” matches into your trees.

  17. Meanwhile I have not had any NADs or Circles at all (and only one leaf match). From the sound of it, I guess I’m not missing much, but it was disappointing to hear they were making it harder for them to form rather than easier.

  18. Hi Roberta, Thank you for this post, great job 🙂 I have spent countless hours researching NADS. What I have found (abt 75%) they are the in-laws of my ancestors. I no longer research them.

  19. Thank you for persisting in showing the value and necessity of matching segment details. I finally have what appears to be a good NAD but it looks like putting the NAD to good use will be exceedingly difficult unless I succeed in recruiting some of the members of the Circle to GEDmatch.

  20. Good work Roberta. My small handful of NADs didn´t hang about long enough for me to form an opinion as to whether they were good or bad and I have yet to meet a circle.

  21. I have 41 DNA Circles, 38 of which are direct ancestors, and I have zero NADs. I did have a couple of NADs initially, then when Ancestry made changes, the number went up to 19 NADs, then they made changes again and it went to zero, which is fine with me. Interestingly, my brother has 10 NADs and I can’t figure out any of them.

    You asked if anyone had turned a NAD into an ancestor. Well, not exactly — yet. I’m still hopeful. One of the first NADs I had was for a Robert M. Douglass, b. 1815 in Tennessee. I match 7 out of 16 people in the Robert Douglass DNA Circle. I scrutinized all of the trees of the circle members to gain any clues. They all showed the Robert Douglass was the son of Issom Douglass, b. 1787 in TN, and Nancy Martin, b. 1790 in VA. I then found a marriage record for Issom and Nancy in 1812 in Blount County, Tennessee. I got really excited because I thought this clue was related to a brick wall I have. My maternal third great grandfather was James Martin. He married Jane Glass in 1819 in Blount County, TN, and lived his entire adult life there. Census records show him born about 1792 in Virginia, but I have no idea where in Virginia, and have no idea who his parents were. I have found nothing in any other trees that would help figure out the puzzle. All of the descendants of Nancy Martin have hit a brick with her as well. So, when I saw that James Martin and Nancy Martin were born two years apart, both in Virginia, and were both married in Blount County, TN, I started wondering if they could have been siblings.

    I messaged members of the Robert Douglass DNA Circle to see if they would upload their DNA results to GEDmatch. Only one person, Merrill, responded but she very graciously agreed to upload her results to GEDmatch for herself and her elderly maternal aunt, who are both in the Robert Douglass DNA Circle. The results showed Merrill and her aunt matched me, my brother, and my maternal first cousin, on chromosome 9 in a triangulation match (about 10 cMs). Merrill’s Aunt Emily not only matched us on chromosome 9, but also matched me and my first cousin on chromosome 2 in a triangulation match (about 8.5 cMs). And Aunt Emily matches just me for 18.4 cMs on chromosome 5. Aunt Emily and I share a total of 42.5 cMs and GEDmatch estimates our common ancestor to be 4.2 generations back. If James Martin and Nancy Martin WERE siblings, their parents would be 5 generations back from me. Pretty close. Unfortunately, there were no other matches to us at GEDmatch that were helpful.

    I know that my matches to Merrill and Emily are through my maternal line since our triangulation match includes my maternal first cousin. I have identified all of my maternal 4th great grandparents except for the parents of James Martin and I see no other ancestors that could be the connection between Merrill, Emily, my brother, my cousin and me. And my Martin ancestors are the only ones who lived in Blount County, Tennessee. So, although the brick wall remains, I am hopeful that I will be able to break it down one of these days. Just for fun, I added Nancy Martin as a sibling to my third great grandfather, James Martin, and voila. The NAD I had for Nancy’s son, Robert Douglass, turned into a DNA circle. Now, I know I haven’t proven James and Nancy were siblings, but I just wanted to see what would happen to the NAD if I added Nancy and her son to my tree.

    One other thing I’ve done that’s helpful is that I took a screen shot of the blue and white circle that displays at the top of each DNA circle page and I made that emblem the profile image of each ancestor who has a DNA circle of which I’m a member. I’m a visual person so it’s fun when looking at the pedigree chart of my tree to see which ancestors show a DNA circle emblem. Roberta, you have access to my tree from the time we were looking at a possible Hickerson link so if you should view my tree in the pedigree format, you can see what I’m talking about.

    • Great idea … using the circle emblem on your circle ancestors!

      I have nine NADS. I have figured five of them out. They are all cousins with the common ancestors being too far back to create a circle. I think circles are only created when the ancestor is not more than five of six generations back.

  22. I just this week had a minor breakthrough using one of my mother’s NADs . I wish there was a way to confirm with a paper trail but I suspect that ain’t ever happen. I think my theory is pretty reasonable though. I posted it on my blog this week . Easier just to post a link to it here than rewrite it.

    Unfortunately no one in the circles have responded to my messages and none appear to be on Gedmatch so I can’t confirm with a chromosome browser either.

  23. Read your articles with interest, even tho I’ve done no DNA testing. The premise I follow is that anyone I enter in my tree who has little or no documentation elsewhere is “suspect” or considered a “hint.” Sometimes when I change the surname spelling in Ancestry, I gain or lose hints, too. I often note this in my “comments,” but all should ALWAYS be leery of tree entries with no reasonable sources!

  24. Pingback: Ancestry V1 vs V2 Test Comparison | DNAeXplained – Genetic Genealogy

  25. So, one year later I’m reading this excellent post. I’ve had a NAD that is very interesting for me. I was trying to locate a distant African ancestor, and at 2%, good luck with that, right? However, I discovered one particular surname, my maternal grandmother Morgan, where approx. 85% of my shared matches in that line also show trace African, sometimes the same Ancestry named region (Senegal), sometimes not. None of my other lines show anything close to this, mostly no higher than 10% of the matches showing African. Did I say that I’m Louisiana Acadian on my father’s line, and Louisiana English/Welsh on my mother’s, with some crossovers that you’d expect from Louisiana families that have been in the same area since the late 1700’s.

    Here’s where the NAD comes in. I get an Ancestry NAD to a completely unknown surname (Richardson) line, matching 8 people in this circle. I start looking at their trees, which seem to be well documented, and come to a Morgan female two generations up. Some trees list her as a Morgan, others with the same surname as the husband’s, which I assume is why I wasn’t getting an NAD on the Morgan. The Dawes applications is where it shows her surname as Morgan, and evidently not enough people in that line have made the edit to change her name to Morgan. The Dawes discovery was made by an excellent genealogist in this subject matter. But, Rhoda and my John are in the same state, same general time period. This Rhoda Morgan was born in 1796. My known Morgan ancestor, John, is a brick wall. Many trees try to connect him to a family (probably wishful thinking to get back to Welsh royalty hundreds of years ago), but there is no evidence that connects him to that family, or any other. Now, these people in the NAD Circle are all from a Free People of Color line in northern North Carolina, and very endogamous. They are actually a triracial group and members of the Halawi Saponi Native American tribe. But there is only one Morgan in the line.

    As fortune would have it, I start researching Free People of Color in North Carolina and looking up FaceBook groups. I post my story/theory in a group and get an IM immediately from a genealogist with Rhoda Morgan in her line, and she is a brick wall for them. She invites me into her secret Facebook group for FPOC from this area, and most of the members are on Gedmatch. So we start sharing kit #’s. Now I’m up to 12 matches that share Identical segments, mostly on the 4th chromosome, and with some on the 9th, but the 4th is where most of the identical matches are. So I believe (and please tell me if I’m looking at this all wrong), that their Rhoda and my John come from a common ancestor, probably 2-3 generations back. They are both brick walls, and known documentation seems to be exhausted, so we may never know. It’s sure an interesting journey though, and one that a NAD hint started.

  26. Thank you! It hit me as soon as I posted the question…chromosome browser. The 9th chromosome paints almost completely African at the matching segments. :). Interestingly, the 4th chromosome paints as North Atlantic and Baltic (using Eurogene K13), which would match the Welsh side of the DNA.

    And yes, my Hebert line is Acadian and shares every possible name from the 18th century Nova Scotia Adadians. And one side was Italian disguised as an Hebert. I have three Hebert lines coming into my family. Two from the Acadiana Hebert brothers, and one from Southern France, but was Italian. Very very sticky genes from that Italian line. He married another Italian in Louisiana in the late 1700’s. Their descendants married into German, Acadian, and English lines, yet, I get 22% Italian in Ancestry admixture…somewhat less in Gedmatch, but not much less.

    Thank you for your help!

Leave a Reply