Ancestry new match list woes

[Glen TK]

Ever since she tested my half sib was predicted as my half sib and that is what appeared to everyone on protools. The new match list rolled out to me today and suddenly she is my niece on both the match list and also to everyone using protools **EDIT** Some matches are now seeing S reported as my niece and others as my grand daughter, considering I'm 19 years younger than S you really have to ask why they suddenly dropped the half sib offering which I'd confirmed previously as correct. I can edit the relationship but it reverts back to niece as soon as I return to the match list.

I've noticed a lot of the relationship predictions have changed when scanning through my matches and checking tests I manage or can view. I'd already cancelled the protools renewal anyway but while my half sib appears as my niece I won't be renewing.

[Phoenix]

I don't think Ancestry can cope with the amount of data it has to manage. Despite sib, sib's child and my late cousin all being part of the same tree, they only appear on one branch of Thrulines. We apparently only share one great grandmother in common. These corruptions of data appear to happen randomly, presumably on updates, but don't get subsequently sorted.

[kiterunner]

I saw on Facebook that this is happening to other people, Glen - I daresay it is a temporary glitch which will be fixed soon.

[Glen TK]

It's getting even sillier as the day goes on.
I manage my test as well as S's (my half sib), along with viewing rights to a shared match we both have called E.
My test as of about 6pm predicts half sib again
S's test now says I'm her uncle
When viewing E's test and clicking my name, S is reported as my grand daughter. Click on S's name and I'm reported as her uncle.

[marquette]

One of the six kits I manage has lost all the Common Ancestors and ThruLines. Its really annoying.

Today, my father, my daughter and I have a new match - dad at 12cM in one segment (13cM unweighted, maternal), me at 18cM in 2 segments (19cM unweighted, paternal) and daughter at 20cM over 2 segments (22cM unweighted, maternal), none shared with other daughter. Tree of 10 people, all private or living.

No shared DNA with any of the other of the array of cousins, from 1st to 6th cousin, all over the world. Marked it with a query and notes so it does not get lost.

Maybe I should have a go with Protools and see if thats any help!

[Mary from Italy]

It's worth a try. I've been using it, and I quite often find that someone who has no tree turns out to have a shared match with a close relative of theirs who does have a tree.

[Glen TK]

Originally Posted by Mary from Italy

It's worth a try. I've been using it, and I quite often find that someone who has no tree turns out to have a shared match with a close relative of theirs who does have a tree.

I've found a few like that including one who removed their username but left their results up. The close relative they have is estimated as a half sibling so that might explain why but I can't be sure. Perhaps the half sibling was a shock discovery they would rather not let the world see, I can imagine a scenario where if they are a casual user then someone with protools could have been aware of it long before they knew themselves.

It's also been useful with a recent small group of matches that popped up over a two day period, I think someone purchased several kits for family members and opened up accounts for them but has potentially assigned the kits wrongly. Male icons with female names but predicted relationship as a father or son to another with mismatched icon/name. If they have also entered ages/dob's that are wildly inaccurate then it has an impact on predicted relationships too. Maybe someone gave a deathbed confession and they are trying to prove it without realising the implications that the wrong data has further afield. If nothing else it proves my aging mind is still capable of far fetched ideas anyway.