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

[Richard Lung]

KM

BTW my ipad doesn't keep a lot of emails, including yours.

My apologies for not answering sooner (privilege of old age).

Your calculation criticism is perfectly fair. I think the best answer I 
can give is that C is elected on over the quota. And that literally 
No-one or Nemo is excluded, on an even larger exclusion quota.

If Binomial STV implementation progresses, there would no doubt be a 
good many other second guesses -- it would be astonishing if there weren't.

My main interests are to see things like the effect of exponential decay 
of preferences, in realistic elections, to assess  the balance of power 
between election and exclusion counts. And to see how much significance 
tests are needed on candidate proximity to the election quota, when that 
candidate has not been excluded -- which is not the same as being elected.

Regards,

Richard Lung.


On 21/05/2022 08:57, Richard Lung wrote:
> Thank you, Kristofer,
>
> I refer you to the post as a whole, not just the first couple of 
> lines, for my answer.
>
> It's not just a matter of "shut up and calculate" to quote a famous 
> grouse of hapless quantum theory students.
>
> (You make me suspect you are an instructor. However, I appreciate your 
> consideration and competance.)
>
> Regards,
>
> Richard Lung.
>
>
> On 16/05/2022 10:13, Kristofer Munsterhjelm wrote:
>> On 16.05.2022 08:20, Richard Lung wrote:
>>>
>>> Binomial STV is later no harm, unlike Borda count, because uses keep
>>> values , equivalent to Gregory method, for both election and exclusion
>>> counts.
>> I would like to check that for myself. That's why I've asked (three
>> times) if you could give me the concrete keep and exclude values, and
>> the winners, for particular example elections involving truncation.
>>
>> Could you please do that?
>>
>> -km
> ----
> Election-Methods mailing list - see https://electorama.com/em for list 
> info