How do transfers in our parliamentary elections work?

amacachi

It's something I've not been able to find an answer to, suppose someone has a surplus of 1,000 votes which are then distributed, which of the votes cast for them are used, is it just the ones closest to hand or what?

Sulmac

All of that candidate's votes are checked for their second preference and redistributed proportionally.

From CitizensInformation.ie:

Surplus votes

If a candidate receives more than the quota on any count, the surplus votes are transferred to the remaining candidates in proportion to the next available preferences indicated by voters (i.e., the next preference on each vote for a candidate who has not been elected or eliminated). For example, if candidate A receives 900 votes more than the quota on the first count and on examining all of his or her votes, it is found that 30% of these have next available preferences for candidate B, then candidate B does not get 30% of all candidate A's votes, candidate B gets 30% of his/her surplus, i.e., 270 votes (30% of 900).

Where a candidate is elected at the second or at later count, only the votes that brought him/her over the quota are examined in the surplus distribution, i.e., the parcel of votes last transferred to the elected candidate.

If two or more candidates exceed the quota at the same time, the larger surplus is distributed first. The distribution of a surplus is prohibited if it cannot materially affect the progress of the count, either by electing a candidate or saving the lowest candidate from elimination.

Where there is no surplus for distribution or the distribution of the surplus is prohibited, the next step is the elimination of the lowest candidate. Two or more of the lowest candidates must be excluded together where it is clear that they cannot possibly be saved from elimination in the long run. Where a candidate is eliminated, all of his/her votes are transferred to the next available preferences on them.

Counting continues until all the seats have been filled. The last seat can be filled either by a candidate(s) exceeding the quota or by a candidate(s) being elected without reaching the quota because it is clear that he/she is ultimately going to be elected. Thus, if the number of seats left to be filled is just one less than the number of candidates still in the running and an available surplus cannot bring the lowest candidate level with or above the second lowest candidate, all the candidates, except the one with the lowest number of votes, are deemed elected even though none of them have actually reached the quota.

amacachi

Sulmac wrote: »

All of that candidate's votes are checked for their second preference and redistributed proportionally.

From CitizensInformation.ie:

That's what I thought would be fairest, also explain how it takes so long, thanks.

nesf

Sulmac wrote: »

All of that candidate's votes are checked for their second preference and redistributed proportionally.

No. That is only on the first count. For all later counts it's the last votes counted that are passed forward as surplus.

Where a candidate is elected at the second or at later count, only the votes that brought him/her over the quota are examined in the surplus distribution, i.e., the parcel of votes last transferred to the elected candidate.

amacachi

nesf wrote: »

No. That is only on the first count. For all later counts it's the last votes counted that are passed forward as surplus.

So the same votes counted properly 10 times could give 10 different results?

nesf

amacachi wrote: »

So the same votes counted properly 10 times could give 10 different results?

No. Only the original batch forwarded is used in any recounts.

Eliot Rosewater

Making the second and later counts proportional could only be done feasibly on a computer. I once designed a program to calculate PR-STV election results that, I think, retained proportionality all the way through.


The program gives each vote a weight, with all votes starting with a weight of 1. When it comes to redistributing votes from excess quotes every single vote is redistributed, but the weights are amended. For example, if the quota is 900 and Mr A gets 1000 votes, all of his 1000 votes are redistributed but they are weighted so that their sum (their value) comes to 100. So each ballot paper becomes a tenth of a vote for the next candidate.

If, say, one of these votes then goes to a candidate Mrs B who also gets elected with a surplus, the weight of the vote is reduced again (for simplicity, by another 90%, giving the vote an overall weight of 1/100).

What now happens is that the redistributed votes have different weights. The vote X which had Mr A as first preference, and Mrs B as second, will have weight 1/100, while the vote Y which had Mr B as first preference will have weight 1/10 (ten times more valuable than the other vote).

But I think this is actually fair and proportional. 90% of vote X was used to elect Mr A, and 9% used to elect Mrs B, and so it has 1% leftover. Only 90% of vote Y was used, so it rightly has 10% leftover.


The problem really, is if a vote is redistributed to a candidate who has already been elected. The vote won't change but by rights it should. If vote Y is supposed to go to Mr A, but couldn't because he was already elected, it will go straight to Mr C with weight 1/10. Vote X will go to Mr C with weight 1/100 even though the top two preferences of X and Y are the same.

A solution would be to knock the percent off of Vote Y for electing Mr A as if it were present when he was elected OR adjudicating the weights altogether, which would be messy and potentially disastrous: you could get an infinite loop.

ISAW

Eliot Rosewater wrote: »

Making the second and later counts proportional could only be done feasibly on a computer.

I disagree. Say a quota is 8000 and nobody reaches it and a candidate is eliminated with 150 votes it is very easy just to give them to the secomnsd preference. and with pidgeon holes very easy to track all the following no 3 ,4,5 etc.

The only complication is as stated distribution of a surplus.

And I also disagree that only the last packets are taken in a recount. AFAIK a "full recount" involves recounting the lot.

I once designed a program to calculate PR-STV election results that, I think, retained proportionality all the way through.

Good for you.

Eliot Rosewater

ISAW wrote: »

The only complication is as stated distribution of a surplus.

Yeah. Hence why you need a computer.

ISAW wrote: »

Good for you.

Are you suggesting that the computational method I devised for retaining proportionally is somehow irrelevant to this thread and was only posted for my ego? I wouldn't do that, my ego is over-inflated enough as it is. :pac: