AncestryDNA is at it Again – Missing Shared Matches
For those of you who wonder if I’ll report the error instead of just complaining about it, I will. But I want to get my 2 cents worth in first, and I expect no results from the bug report. But, I’ll do it anyway.
I haven’t trusted the “Shared Matches” feature since it was introduced. The count seemed low – WAY low. Now that my paternal half-sister (“Sissie”) is in the system, it is obviously too low. On AncestryDNA, I have 3,569 matches in total, but only 19 shared matches with Sissie. Common sense itself says that is not right. It can’t be right. I mean, even if most of the DNA she inherited came from our common ancestors whose DNA I didn’t inherit, it still can’t possibly be right. It’s only 0.53%. So I don’t know what criteria Ancestry is using for these matches, but it is not right.
For comparison’s sake, at Family Tree DNA (FTDNA), I have a total of 1,001 matches, and 194 in common with Sissie. That is 19.38%. Now that’s more like it. I am biased for FTDNA. With good reason.
You know why AncestryDNA bothers me so much? The lack of transparency. The errors. The rigidity. The expense.
Back to the situation.
I’ll use just one common match with Sissie and me on AncestryDNA. SPS, a female, is a descendant of James RENFRO and Margaret JACKSON, from their son Jesse J. RENFRO. Sissie and I are descended from their son Absalom RENFRO who married Elizabeth CORMACK/McCORMACK in 1830 and died in 1855.
SPS and her brother SLS are our 5th cousins, 1x removed. On AncestryDNA, both SPS and SLS show up as matches to both Sissie and me, but they do not show up as Shared Matches between the two of us. So what gives? How am I to trust that feature? How am I to trust the results?
It is not hard to write a matching algorithm between two data sets. I’ve done it myself.
So how many hundreds of “Shared Matches” am I missing?
4.^ Elizabeth Wilson Ballard, “Dear Kenny – Any Suggestions for Our New Ancestor Discoveries?,” Diggin’ Up Graves (https://digginupgraves.wordpress.com : accessed 23 May 2016).