[EM] Notes on a few Later-no-harm methods

[Kristofer Munsterhjelm]

On 11.05.2022 19:32, Richard Lung wrote:
> 
> Binomial STV is a later-no-harm method. It is also monotonic. It is
> a both elective and exclusive rational count.

In the full preferences case, single-winner Binomial STV in effect works
by letting each candidate's score be the number of first preferences
divided by the number of last preferences, and then electing the
candidate with the highest score.

Later-no-harm is about what happens when you add rankings to an
incomplete ballot, e.g. going from

A>B

to

A>B>C.

Now suppose we have a three-candidate single-winner election, and
suppose the ballots are, just as an example:

3: A>B>C
5: A>C
13: C

How many last preferences does Binomial STV count A, B, and C as having?

And for the sake of completeness, in the election:

3: A=B>C
5: B=C>A
13: C=A>B

How many first preferences do A, B, and C have according to Binomial STV?

-km