3

Original question: Oregon will be voting on the use of Ranked-choice voting in 2024, and I am preparing a presentation for interested local community groups to demonstrate how the process will operate. However, after doing my research, I still have a question about what I believe are called "exhausted" ballots.

If someone does not rank every candidate on their ballot and all of their votes have been "exhausted" in the tallying process, is that ballot then subtracted from the total number of ballots in subsequent run-off tallies? Or does it remain in the total count of ballots when trying to calculate if another candidate - one the person did not rank on their ballot - has reached a majority.

If it is the latter - that the exhausted ballot remains in the total count of ballots - doesn't it make it harder for another candidate to reach a majority+1 result?

Revision 1: Say I have an election with 5 candidates: White, Yellow, Orange, Green and Purple.

Here are the ballots:
Ballot(Yellow, Green, Purple, White, Orange)
Ballot(Purple, Green, Yellow, White, Orange)
Ballot(Yellow, Purple, Green, White, Orange)
Ballot(Green, White, Purple, Orange, Yellow)
Ballot(Green, White, Orange, Yellow, Purple)
Ballot(Yellow, Green, Purple, Orange, White)
Ballot(Orange, Yellow, Purple, White, Green)
Ballot(White, Orange, Green, Purple, Yellow)
Ballot(Orange, Green, White, Yellow, Purple)
Ballot(Purple, Yellow, Green, White, Orange)
Ballot(White, Green)
Ballot(Purple, Green)
Ballot(Green, Yellow)
Ballot(Purple, White)
Ballot(Yellow, Orange)
Ballot(White)
Ballot(Yellow)
Ballot(White)
Ballot(Green)
Ballot(Purple)

This type of voting is relatively new here, so some people did not rank all of the available candidates.

Here is the result of tallying the votes.

Round 1

Candidates, Total Votes, % (Total Votes / Total Ballots Count), % (Total Votes / Votes Cast in Round), Status
White, 4, 20%, 20%, Moves to next round
Yellow, 5, 25%, 25%, Moves to next round
Orange, 2, 10%, 10%, Dropped
Green, 4, 20%, 20%, Moves to next round
Purple, 5, 25%, 25%, Moves to next round

Votes Cast in Round: 20
Exhausted Ballots Count: 0
Total Ballots Count: 20

Round 2

Candidates, Total Votes, % (Total Votes / Total Ballots Count), % (Total Votes / Votes Cast in Round), Status
White, 4, 20%, 20%, Dropped
Yellow, 6, 30%, 30%, Moves to next round
Green, 5, 25%, 25%, Moves to next round
Purple, 5, 25%, 25%, Moves to next round

Votes Cast in Round: 20
Exhausted Ballots Count: 0
Total Ballots Count: 20

Round 3

Candidates, Total Votes, % (Total Votes / Total Ballots Count), % (Total Votes / Votes Cast in Round), Status
Yellow, 6, 30%, 33%, Moved to next round
Green, 7, 35%, 39%, Moved to next round
Purple, 5, 25%, 28%, Dropped

Votes Cast in Round: 18
Exhausted Ballots Count: 2
Total Ballots Count: 20

Round 4

Candidates, Total Votes, % (Total Votes / Total Ballots Count), % (Total Votes / Votes Cast in Round), Status
Yellow, 7, 35%, 44%, Lost
Green, 9, 45%, 56%, Elected?

Votes Cast in Round: 16
Exhausted Ballots Count: 4
Total Ballots Count: 20

It seems that in Round 4, Green won the election with 9 votes, which was 45% of the total ballots (or 56% of the total votes cast in the round).

So I am confused as to which percentage is used in ranked-choice voting, the (total votes / total ballot count) or (total votes / votes cast in round). If it is the (total votes / total ballot count) version, then it seems to me that the number of exhausted ballots continue to have an outsize effect on the percentage and make it harder for a candidate to get a majority+1 of the votes cast as the process goes through more rounds.

Apologies if this question seems simplistic, but I'm just trying to get my head around it... Thanks!

Improve this question
2
  • I'm not sure i understand your question, but only the winner and runner up totals matter at the end. If you voted for one of those, the ballot will count, and if you didn't, it won't. You don't specify the "round", just who you like.
    – dandavis
    Commented Dec 5, 2023 at 6:46
  • //"... but only the winner and runner up totals matter at the end."// The problem is when they announce the percent vote that the winner gets, that fraction (which RCV advocates claim is more than 50%) is exaggerated upward because the denominator is too small and excludes all of those exhausted ballots.
    – robert bristow-johnson
    Commented Mar 25 at 1:46

2 Answers 2

Reset to default
2

The full text of the measure can be found on Ballotpedia

There are two cases.

1. Elections with a single winner (Section 4 Subsection 2)

(A) If an active candidate has a majority of votes in a round, the candidate with the greatest number of votes is nominated or elected and the tallying of ballots is complete for that office.

(B) If no active candidate has a majority of votes in a round:
(i) The active candidate with the fewest votes is defeated and is no longer an active candidate;
(ii) Votes that had been counted for the defeated candidate are transferred to each ballot’s next highest-ranked active candidate; and
(iii) A new round of vote tallying begins with ballots retallied in the manner described in this subparagraph

2. Elections with multiple winners (Section 4 Subsection 3)

(3) Notwithstanding subsection (2) of this section and except as otherwise expressly provided by law, when an election to an office is determined by ranked choice voting, and more than one person is to be elected to a single office, the people elected to the office shall be determined by a proportional methodology adopted by rule by the Secretary of State. Any rules adopted under this subsection shall provide that candidates are elected to office by:
(a) Receiving more votes than a threshold determined by dividing the total votes counted for active candidates in the first round of tabulation by the sum of the number of people to be elected plus one, with all votes that are received by a candidate that are in excess of the minimum number of votes required to be elected to office being transferred to lower-ranked active candidates in the manner set forth in the proportional methodology adopted by the secretary under this subsection; or
(b) If the number of active candidates is less than or equal to the number of seats remaining to be filled, by being one of the active candidates.

In the first case, exhausted ballots are removed from the total for each round. If one candidate has a majority of the remaining votes, they are elected.

In the second case, a threshold is calculated in the first round with candidates above that threshold being elected. Exhausted ballots do not lower this threshold but if no candidate reaches it, the lowest ranking candidates are eliminated until there is a seat for every remaining candidate. At that point all remaining candidate are elected.

Improve this answer
6
  • In other words, if people don't rank all the choices then it's possible for someone to be elected without having been ranked by a majority of voters, for example if 49% rank candidate A higher than B and 48% rank B higher than A and 3% do not rank A or B.
    – phoog
    Commented Dec 5, 2023 at 11:23
  • The implication of this is that here in the second case the "threshold/quota" does not change for exhausted ballots (it does in more complicated systems), while in the first case transfers which will not change the result (because of exhausted ballots) are not made.
    – Henry
    Commented Dec 5, 2023 at 11:52
  • @phoog Of course, given turnout, that's true in many elections around the world regardless, depending on whether voters means "people eligible to vote", or some other measure. After all, ranked choice voting systems are also called instant run-off, among other names.
    – origimbo
    Commented Dec 5, 2023 at 14:12
  • @phoog The more choices that are not ranked, the closer this system comes to plurality voting. If voters only indicate their first choice, the result is just that.
    – xyldke
    Commented Dec 5, 2023 at 14:25
  • 1
    Section (3)(b) implies that you could have more remaining seats than remaining candidates. This is, of course, normally impossible, unless there were fewer candidates than seats from the start.
    – OpenAI was the last straw
    Commented Dec 5, 2023 at 16:52
-1

There is some serious dishonesty coming from FairVote and RCV advocates in that they over-sell the product they're marketing. They claim it can do things that it can't. Like guarantee that the candidate elected has majority support.

At the end, they claim that the candidate elected has over 50% of the vote. That's sometimes is not true and was not true in the Burlington 2009 nor Alaska 2022 (August special election) cases.

By excluding the exhausted vote from the denominator of the fraction of the total vote that votes are counted for the winner, they are fudging the numbers.

Improve this answer
4
  • This doesn't seem to be a real attempt to answer the question but just a convenient place for you to complain about RCV
    – James K
    Commented Mar 25 at 6:09
  • @JamesK , try not to confuse or conflate "IRV" with "RCV". The former is a subset of the latter. This is about the fact that "In [Instant-Runoff] voting, ... exhausted ballots [are] removed from the total vote count for subsequent rounds". And I am saying that they shouldn't be, for the purposes of computing the percentage of the vote that the IRV winner gets. We get dishonest reporting regarding the performance of the IRV election and the nature of the victory of the winner of the IRV election. Sometimes they win without a majority. Not even a Simple Majority.
    – robert bristow-johnson
    Commented Mar 25 at 16:37
  • I don't, (not least because the terminiology and politics is rather different in the UK.) But your answer only mentions RCV, not IRV. And in the context of the USA, the comparison is not the choice between IRV, STV, various Concordet systems or other versions or RCV, but with a version of RCV and FPTP. Anywho, the question was "are exhausted ballots removed from the vote? xyldke seems to have answered that in the affirmative. Your answer adds nothing. I don't see it as "fudging the numbers, just how the system works.
    – James K
    Commented Mar 25 at 20:32
  • The title of the question says "Ranked-choice voting", but as soon as it mentions (second word later) "exhausted ballots", we know it's about Instant-Runoff Voting. Now in both Burlington 2009 and in Alaska in August 2022, both of those elections were won by a candidate that did not receive a "majority" or more than half of the vote. But they repeatedly advertise that the candidate who wins must receive more than 50% of "the vote". But it's not true, and it's not true because of not counting the exhausted ballots in "the vote".
    – robert bristow-johnson
    Commented Mar 27 at 2:01

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .