8 cM

[Phoenix]

When Ancestry say 8 cM, do they mean:

8.0001 -8.9999 cm or
7.5000 -8.4999

It makes quite a difference and probably 70% of matches will fall in the lower half of the range.

[kiterunner]

I believe it is the second one.

[Phoenix]

That is certainly my gut feeling. I am wrong, or used they to tell you, before they last improved the system, the figure to several decimal places?

[kiterunner]

They did used to, yes.

[Phoenix]

And they have done again! At least we can see which ones are for the chop now.

[kiterunner]

Okay, when I filter by amount of shared DNA now and put 8 to 8, it is showing me 8.0 to 8.9, but it must have been showing 7.5 to 8.5 before, because I had started adding the "8 to 8"s with linked public trees into a group, and none of the 8.9 ones are in that group.

[Phoenix]

Quite right, Kite. The 8S that I worried about are showing as 7.5-7.6. I didn't manage to group any more.

[Kit]

Anything up to 7.9 will go and over 8 will stay?

[Phoenix]

That's my understanding. Interestingly, 6cM previously seems to hae been 6 - 6.5 cM.

[kiterunner]

At some point before they started showing a decimal point, I had started adding those in the 8 - 8 category to a group, and now I see that none of those now shown as 8.5 are in the group, but all shown as 8.4 are. So I think they are showing the first digit after the decimal point rather than rounding, for instance if it is 8.499 it is showing as 8.4. In which case if it is 7.999 it should be showing as 7.9.

[Phoenix]

I have now looked at all my matches and all Best Mates' up to 7.9 cM. I have now hidden 30k of her matches and 26k of mine. It will be interesting to see how many remain in September. There is no way I have time to do the exercise on the other trees I manage.

[kiterunner]

I don't know what happens to hidden matches when they do the cull.

[Phoenix]

I am hoping that the ones that aren't marked do go. Though when I looked at my friend's hidden matches, there were about a dozen that I'd previously rejected who had put trees up and had common ancestors.

[Phoenix]

"Predicted relationship: 5th–8th Cousin
Shared DNA: 6.9 cM across 1 segment
Longest Segment: 8 cM"

So, do I lose a match like this, or keep it?
What does it mean?

[kiterunner]

I guess it must mean the segment was 8 cM's before the Timber algorithm was applied? Or it's a mistake.

[kiterunner]

Yes, if you click on the longest segment info, you will see that it says "In some cases, the length of the longest shared segment is greater than the total length of shared DNA. This is because we adjust the length of shared DNA to reflect DNA that is most likely shared from a recent ancestor. Sometimes, DNA can be shared for reasons other than recent ancestry, such as when two people share the same ethnicity or are from the same regions."

[kiterunner]

If you click on "Shared DNA" now when viewing a DNA match, you will see the figure for "Unweighted shared DNA" is given along with the other figures. This is the total amount before the Timber algorithm is applied. This only applies to matches with whom you share less than 90 cM's of DNA, as they don't use Timber on closer matches.

[Phoenix]

That's really useful, Kite - particularly for extra segments.