[EM] Definite Majority incompatible with a burial-resistance property

[Chris Benham]

Kevin,
For a long time I had the idea that DMC had an anti-burial property that 
might be called
"Approval Dominant Mutual Third Burial Resistance", which says that if 
there are 3 candidates
X,Y,Z and X wins and is exclusively approved on more than a third of the 
ballots then changing
some ballots from Y>X  to  Y>Z can't change the winner to Y.

I've claimed this is true (including I think on EM), but have now 
discovered that this is incompatible
with Definite Majority (or at least with the combination of  DM and 
Condorcet).

34: A|B
17: C|A
49: B|A

A is the CW so DMC elects A.

34: A|B
17: C|A
26: B|C
23: BC|

B>C>A>B. Approvals:  B49,  C40,  A34.

The original  49 B|A voters have all buried A, dividing into 26B|C and 
23BC|.  The  DM set is {B}, but
the anti-burial property says "not B".

I think AWP (which is Schulze or River or Ranked Pairs measuring defeat 
strength by "Approval Against")
does have the anti-burial property.  (In the above example in AWP the 
burying fails, changing the winner from
A to C.)

Among methods that use ranked ballots with an approval threshold 
(allowing voters to rank among unapproved
candidates) this discovery has had the effect of somewhat raising my 
regard of both  ASM and AWP relative to DMC.

Chris Benham