The entire idea of “votes not being wasted” sounds good at the first look, but is it really more than just a catchphrase?
For the top-to-bottom distribution, my specific question is:
In some made-up STV vote a candidate needs to reach 3 votes in order to get elected.
Four voters voted for Candidate A as their top choice:
Voter 1: A, B, C, D, E
Voter 2: A, C, D, E, B
Voter 3: A, D, E, B, C
Voter 4: A, E, B, C, D
This means that Candidate A is elected and 1 vote is in excess and thus will get transferred.
Question: who gets that vote, Candidate B, C, D or E?
It seems that the problem lies in the pre-fixed number of seats. One could instead have a quota of voters (5% or so) with a minimum number of seats.
The other problem seems to be how majorly important the order of any voters list is. I don’t think it is clearly advertised, that once your first on the list vote counts because the candidate is neither too high nor too low, your other votes won’t. While if your first choice drops out of the race due to winning a seat or being the lowest, you have a 2nd vote. That’s weird.
Voters whose first choice is a candidate that would compete for 10th and final place could effectively not be represented because the rest of their lists would never be looked at.
In other words, isn’t it possible to massively influence the outcome of the elections, if you were a large block and knew about several other blocks voting lists? This would allow to develop mathematical models which could help you to seize opportunities in influencing the dogfight for the last CSM seats.
For instance, you could ask some part of your block-members to strategically vote #1 for people you don’t support and who will surely drop out of the race. This would stop them from dropping out early and thus other people who vote them first, won’t have their 2nd choices counted before their 1st choice drops out of the race. At which point it could be that their 2nd choice already dropped out of the race, despite their accumulated #1 and #2 choice voters would have been enough to keep them in longer.
Given that you had enough reliable data about opponents voters behaviour, you could make educated guesses on how to manipulate the outcome of the final seat fillings, by playing the algorithm. Such knowledge would not be realistic in the real world, but it seems it could be in EVE.
TL;R you cannot exclude people who immediately hit the quota, but you can try to time when 2nd and 3rd choices are counted in order to get the upper hand for those last vacant seats