(Note: I did not view the updated part of the question before writing this answer, so my original answer does not address the specific scenario added in the update. See the section below the dividing line for the specific question of the generational 'range' of various tests.)
The Problem with Ancestry DNA
You're probably already aware of this, but Ancestry recently changed the manner in which DNA test results and tree matches are shown to customers. Crista Cowan reviewed the changes in a video AncestryDNA: Cousin Matches and DNA Circles; DearMYRTLE also discussed the changes in her 24 Nov 2014 episode of Mondays with Myrt -- this link is to the timestamped version, so you can go directly to the relevant segment at 17:36. Under the video there are links to other bloggers, including Judy D. Russell's post Changes at AncestryDNA: The good, the bad and the ugly.
If you have tested there, I encourage you to download all match lists and raw test results from Ancestry ASAP while your data is still available for you to access. There is also a white paper available to people who have tested via AncestryDNA, which explains all the recent changes.
A disclaimer: I am far from an expert on DNA testing -- I think it's fair to call myself a skeptic, especially with results from companies like Ancestry which do not reveal how you match to other individuals in the database. Most of my readings in DNA have been books and articles about specific traits which discuss specific alleles, so the matches which express only a percentage range seem way too fuzzy to me.
This recent blog post "Chromosome Browser War" from Roberta Estes on her blog DNAeXplained – Genetic Genealogy explains why people who would like to do serious DNA research are not happy with the product put out by Ancestry.com. The link to Estes' post is courtesy of Hoosier Daddy's author, Michael Lacopo, who said in his post of 14 Dec 2014:
People test at AncestryDNA because it sounds fun. Because they want
the warm fuzzy experience. The company gives you fun colorful graphs
of what countries your ethnicity derives (which I have already stated
is as accurate as measuring a doorway with a cat.) Further, if you
find matches with others using AncestryDNA, the tool they use to show
you how you are related is by comparing your family trees to each
other. And if you are matched with someone with shoddy research or no
family tree, the results are meaningless. Lastly, having matching DNA
segments and knowing you have a common ancestor is great, but it is a
dangerous assumption to think that they go hand in hand. That is why
triangulation with another researcher with the same common ancestor is
necessary to see if you all share the same strand of DNA at the same
chromosomal location. This is something you cannot do at AncestryDNA.
The need for a paper trail
My question to you is -- how many generations back is your "pre-1800" void? If I recall correctly, in one of the recent videos, Angie Bush told DearMYRTLE that the results for autosomal were good to about six generations back (see link in the resource list). The big problem is that you may have a match in your paper-trail trees with a lot of people, and a DNA match, but there's no guarantee that your DNA match is because of the line in your paper trail -- you could match via some other means which you are not aware of, especially if you're talking about distant cousins, and both of you have gaps in your paper-trail trees. See How Much of Your Family Tree Do You Know? And Why Does That Matter? by Blaine Bettinger (the pointer to this article was shared by Debbie Kennett).
My personal feeling is that while the autosomal DNA information is fascinating, and interesting to explore in its own right, it's important not to put too much emphasis on it. The deck of cards shuffles so much with each generation, it's possible for distant cousins to be related to you but share no DNA just because of the way the chromosomes fell.
I think your question is an important one -- if your relatives are willing, you don't want to lose their information by waiting too long (which is true of all family history information). Here are some criteria I would use to choose test priorities and my reasoning for each one. Just like paper trail research, choose your test by what information the test will give you, and what questions you most want to answer.
Disclaimer: my personal bias against autosomal tests where I can't see the raw data is reflected in this list.
Some things to consider:
- Which company offers the best takeout of the data? A lot of people got burned because their Y-DNA data and mtDNA data got locked up when Ancestry dropped those tests. If you have older relatives pass and their data is tossed, you can't go back and ask them for a fresh sample.
- Your 93 year old grand father on your father's side (barring an unknown 'non-parental event') will probably match yours. Do you want two data points for that line instead of just one?
- mtDNA tests on all of the individuals in your group A-D would give you results from your direct line on your mother's line, but this is more useful for "deep" results. What do you want to know? (see updated section)
- For autosomal tests, I suggest starting with yourself and group A-D. Think of it as establishing a baseline against which the people in group E could be judged.
My reasoning is that you say "I have a long way to go to understand how to process the results". It makes more sense to me to learn by comparing the test results from your known relatives in A-D first, before you ask any of your group E people to take more tests. See Triangulation for Autosomal DNA and the other articles about Triangulation on Roberta Estes' blog, DNAeXplained – Genetic Genealogy. However, since I am a complete rookie at this, my reasoning could be completely wrong -- a sensible person would ask advice from someone much more familiar with the tests, like Judy G. Russell or the other bloggers listed in my resources list.
Like all family history research, one can often make better progress with specific, focused research questions than with wide global searches. As Judy G. Russell says in her post "It isn't so", "DNA testing only works with traditional paper trail genealogy, not instead of it!" So use what you learn from the tests of your own family to help you decide which test results would be most useful from the people in your group E.
For the question of how to get the most "bang for your buck" for your test results, see:
Posts by Judy G. Russell, from her blog The Legal Genealogist:
- Category Archives: DNA from Judy G. Russell, The Legal Genealogist
- CeCe Moore, Your Genetic Genealogist
- Blaine Bettinger,The Genetic Genealogist
- Roberta Estes, DNA eXplained – Genetic Genealogy; category Triangulation
- Genealogy and Genomics
- DearMYRTLE, Video Series: Demystifying My DNA Tests (intended audience: newcomers to genetic genealogy)
- Michael Lacopo's, Hoosier Daddy? -- if for no other reason, it's a great example of how to write a cliff-hanger! The link should take you to the first post, "Beginnings".
- Thomas MacEntee's blog GeneaBloggers, Genealogy Do-Over – Week 10: 6-12 March 2015 (see resources for Topic #1, Reviewing DNA Testing Options)
- Genetic Genealogy Links & Resources by Louise Coakley (thanks to Debbie Kennett for sharing the link to this post).
- "How I DNA" -- a new series by SallySearches. Part One posted 28 Sep 2015 has a chart summarizing the companies that her family members have tested with and where the results have been transferred. Making a chart like this seems like a good idea when you are comparing test results from several people.
This section has links which specifically address the concerns in the current question title, "Generational Loss of Data with DNA Testing".
The following quotes are from the Learning Center at Family Tree DNA.
The Family Finder is Family Tree DNA's name for their autosomal test.
Q: How many generations does Family Finder analyze or predict?
The Family Finder program determines relationships for up to five
generations. We will return data and matches that are more distant.
However, we established the confidence interval based on the level of
quality to which Family Tree DNA customers deserve and expect.
In their page explaining autosomal tests, they say:
If you are trying to confirm a relationship with someone else who is a
3rd cousin or closer, the Family Finder test is recommended. We will
provide you with names, email address, and the genealogical
information your matches have shared. You will be able to communicate
with them freely to find your common ancestors.
Please note this test cannot distinguish between matches from your
mother’s side versus your father’s side.
Mitochondrial DNA tests:
Q: How many generations back does mitochondrial DNA (mtDNA) testing
Mitochondrial DNA (mtDNA) testing covers both recent and distant
- Matching on HVR1 means that you have a 50% chance of sharing a common maternal ancestor within the last fifty-two generations. That
is about 1,300 years.
- Matching on HVR1 and HVR2 means that you have a 50% chance of sharing a common maternal ancestor within the last twenty-eight
generations. That is about 700 years.
- Matching on the Mitochondrial DNA Full Genomic Sequence test brings your matches into times that are more recent. It means that you have a
50% chance of sharing a common maternal ancestor within the last 5
generations. That is about 125 years.
Mitochondrial DNA testing at Family Tree DNA also includes haplogroup
testing. Your haplogroup represents your ancestral origins thousands
and tens of thousands of years ago.
Y-DNA and paternal line testing:
Paternal line DNA testing uses STR markers. STR markers are places
where your genetic code has a variable number of repeated parts. STR
marker values change slowly from one generation to the next. Testing
multiple markers gives us distinctive result sets. These sets form
signatures for a paternal lineage. We compare your set of results to
those of other men in our database. The range of possible generations
before you share a common ancestor with a match depends on the level
of test you take. A match may be recent, but it may also be hundreds
of years in the past.
More information is available at their Learning Center and in their Recommended Reading section. Also, there may be information that can help you make a decision in the other questions here tagged dna.
Other posts of interest by by Judy G. Russell on The Legal Genealogist:
- For a case study, see Time to MRCA: Proving John Locke, posted 22 Nov 2015
- For the problem of why you might get a match for Y-DNA or mt-DNA tests but not see the person in an autosomal match list at Family Tree DNA, see Why is the chair empty? posted Dec 3, 2017 (in which she discusses the differences in 'reach' between autosomal and Y-DNA and mt-DNA tests).