* New Ancestry longest segment

The place to post news about genealogy products and services that might be of interest to other Family Historian users.
Post Reply
User avatar
gwilym'smum
Superstar
Posts: 302
Joined: 01 Feb 2016 16:28
Family Historian: V6.2
Location: South Cheshire
Contact:

New Ancestry longest segment

Post by gwilym'smum » 21 Aug 2020 07:11

Hi
I would appreciate members' thoughts on my Ancestry match. I logged on to Ancestry to look at the new longest match tool. Eventually, after searching, found it. Thought very interesting and could be helpful. At the same time I discovered a new quite high match - interesting. I emailed but par for the course have not heard back and she has no tree. I then decided to find her myself so started with shared matches. Great - I think I know which line she is on. I then looked at the longest segment for all the shared matches. (I know Ancestry have commented that the longest segment could be more than the total match - but I would not expect that to be by a large amount) Every match had a longest segment of 68 cms. My new match is 197 cms and one second cousin is over 200 but the rest of shared are in the 20 cms. I think something has gone wrong with their algorithm - what do FH members think please?
Stay safe and well
Ann
Researching Mayer, Parr/Parr, Simcock, Beech and all related families

User avatar
ColeValleyGirl
Megastar
Posts: 4853
Joined: 28 Dec 2005 22:02
Family Historian: V7
Location: Cirencester, Gloucestershire
Contact:

Re: New Ancestry longest segment

Post by ColeValleyGirl » 21 Aug 2020 07:57

I believe the longest segment is calculated before the TIMBER algorithm is applied, so conceivably that discrepancy is 'expected'.

User avatar
gwilym'smum
Superstar
Posts: 302
Joined: 01 Feb 2016 16:28
Family Historian: V6.2
Location: South Cheshire
Contact:

Re: New Ancestry longest segment

Post by gwilym'smum » 21 Aug 2020 09:04

Hi Helen
I know Ancestry had said that the fact there may be a longer largest segment than the total match because of residual DNA but I hadn't connected that they have worked out the longest segment before the TIMBER algorithm - thank you. However the fact that every shared match contains exactly the same number for the longest segment is really strange.
Ann
Researching Mayer, Parr/Parr, Simcock, Beech and all related families

User avatar
ColeValleyGirl
Megastar
Posts: 4853
Joined: 28 Dec 2005 22:02
Family Historian: V7
Location: Cirencester, Gloucestershire
Contact:

Re: New Ancestry longest segment

Post by ColeValleyGirl » 21 Aug 2020 09:24

If it's a known pile-up region, I can see how it might happen.

avatar
sueherrington99
Platinum
Posts: 30
Joined: 01 Mar 2010 16:23
Family Historian: V6.2
Location: Nottinghamshire

Re: New Ancestry longest segment

Post by sueherrington99 » 26 Aug 2020 11:13

Apparently this is a known bug that has been reported to Ancestry many times, & which they are allegedly working on. It happens when you view a match from the shared matches list. There are some workarounds to getting the correct value -

1 From the shared match list you can click on a match and then just refresh that page.

2 You can click on the amount of DNA instead of the match to get a pop-up with the largest segment information.

3 You can just search for the match in your main match list.
Sue Herrington
Researching Hall, Hemingway, Furnandiz, Rymer

Post Reply