Unexpected number of cMs of DNA matches

[Blackwater Park]

My daughter and I have both done the Ancestry DNA test. Needless to say we share many DNA matches, but what is unexpected is that the number of cM for those matches is often more for my daughter than for me. I would have expected (maybe naively) that the number of cMs matched would be either the same as for me, or less. The result has been that where I have sorted the matches into groups on the basis of their shared matches, it is sometimes the case that there are several members of the group that have shown up on her list but not on mine as presumably they are below the 8cM limit for me.

Is this just down to random errors? Does anyone know what the error bar is on DNA matches? Is this a common state of affairs? Or is my test somehow flawed?

[whoamitoday]

I've seen this across the generations of my family. Generally, for us, it's only for the very distant matches and the discrepancy is only on the first look at the size of the match - when you click on the match size and view the unweighted size, it's makes more sense. Ancestry often adjusts a match size according to it's "Timber" algorithm, that doesn't always adjust consistently, in my experience.

[Blackwater Park]

Didn't know about the unweighted data (thank you!) so I've just tried that out: the first one I tried had 18 cM unweighted for me, 27 for my daughter. Second one from another group: 23 and 29 cM respectively. I suppose what I'm asking is, is that a statistically significant difference?

[whoamitoday]

I'm not certain, but my instinct says that it might be significant. Hopefully someone with more expertise will be able to rationalise it for you.

[Ruskie]

This happens quite a lot with myself and my father, and also with my daughter and my father, her grandfather.

There are numerous examples of my daughter and her grandfather sharing similar cMs with their matches. This is on My Heritage.

I think it is just down to the randomness of DNA inheritance.