-Aspen's historic May 5, 2009 IRV election audited as single ballots- 5/5/09 Aspen CO held an instant runoff election (IRV) for mayor and 2 council members. Interpreted contents of each ballot, scanned by True Ballot, were publicly released. Open records requests for a CD of image scans were denied. Aspen has been sued to protect records from destruction and to allow inspection of the scanned ballot files. A Court of Appeals ruling holds that unidentifiable ballots are public records.

Search this and related blogs

Sunday, April 18, 2010

Discussion of ballot text for possible future Aspen voting methods for November ballot question: Aspen Election

A discussion and questions about the presentation of voting methods for future Aspen elections on the ballot for the November coordinated election.

From: Harvie Branscomb [mailto:harvie@media.mit.edu]
Sent: Friday, April 16, 2010 1:36 AM
To: 'Mick Ireland', Aspen City Council, Aspen Election Commission
Subject: deciding the method of voting

Several voting systems will be identified to be placed on the ballot.  One of them will be similar or same as the IRV used in 2009.
Another one is likely to be plurality/winner takes all, no majority required
Another one is likely to be vote for one, majority required, runoff if necessary (or vote for 2 if CC)
Another one could be e.g. approval voting (each candidate can receive one vote per voter, voter votes for as many as preferred, majority required, runoff likelihood reduced
Another could be e.g. district or party primary, followed by general election, etc.

Will the IRV choice be represented by exactly what was done in 2009 or by an improved description? This seems to be a key question. I now think that calling rank choice voting “IRV” is a mistake.  I now see it as “RRV” or reduced runoff voting- a system that makes runoffs much less likely but would call for a runoff in case a true majority of ballots was not achieved by the winner(s) of the rank choice vote.

Each of these will appear as a separate contest on the ballot such as:
“Do you wish Aspen to elect its Mayor and City Council by approval voting in which each voter may vote for each, any or all candidates, the one candidate (mayor) or 2 candidates (cc) with the most votes exceeding a majority plus one among all ballots cast in the contest (including undervotes) wins, and a runoff is held among the top 2 (mayor) or top 4 (city council) if a majority is not reached in the first election:   Yes   No”

When the votes on these contests are tabulated, the contest with the highest number of “Yes” votes becomes the new voting method if that number also represents a majority of the total votes cast  in a contest (including undervotes).

If no method reaches a majority of all votes cast including undervotes, then?
This is Neil Siegel’s very good question.  If the description of IRV on the ballot will differ from the current implementation in order to make it consistent with law, then clearly Aspen would like to see one of the methods supported by a real majority of voters voting. His method is one way to handle this.  A runoff election would be another and perhaps more clear way to handle the lack of a majority.

I believe that the system I have described is virtually the same as approval voting- one vote for each method, vote for as many as you like- the choice with the most votes wins, as long as the choice also achieves a majority.  In other words, another way to put this on the ballot with the same results and a bit less confusion might be as follows:

The following are 5 methods of voting proposed for use in Aspen’s future elections.  You may vote for any and all methods of your choice.  Any method achieving the largest number of votes where that number represents a majority of the voters voting will become Aspen’s future voting method. In the event none of the 5 proposals receive a vote of approval from a majority of electors voting, a runoff election among the top two will be held. Vote for as many as you like:

Yes, I approve: The IRV (rank choice voting) system employed in 2009 with the following amendments: (a real majority of ballots cast must show votes for the winning candidate(s) otherwise a runoff election will be held)… etc.
Yes, I approve: Each voter may vote for one mayoral and two CC candidates. The 1 mayoral candidate and 2 CC candidates with the most votes win; no majority required
Yes, I approve: Each voter may vote for one mayoral and two CC candidates. The  1 mayoral and 2 CC candidates with the most votes in excess of a majority of ballots cast win, and lacking a majority participate in a runoff election
Yes, I approve: Each voter may vote for each mayoral and each CC candidate.  The  1 mayoral and 2 CC candidates with the most votes in excess of a majority of ballots cast win, and lacking a majority participate in a runoff election
Yes, I approve: Each precinct with contestants shall conduct a primary. Each winner of a precinct primary shall participate in a general election for 1 mayoral and 2 city council candidates.

Note that if in the first above scenario (each choice a contest, Yes and No choices for each) if you do not include the undervotes in the calculation of majority, then each contest will have its own different majority threshold.  This will generate strange non-monotonic effects such as a choice with a lower “Yes” vote count achieving a majority in its contest while a higher “Yes” vote count might not (because of a larger number of “No” votes and fewer undervotes.  I doubt that Aspen would want this outcome to be possible.  Therefore I suggest you use the more obvious definition for majority and count the undervotes.  I also suggest that the implementation of “IRV” be similarly corrected.


Dear Neil:

IRV is in the charter.  The charter itself cannot be amended without a majority.  The alternatives will be presented individually with a yes or no on each of them.  The alternative that gets the most votes and that number is a majority, will replace IRV.  But we cannot put five alternatives in the same question and say that the charter is amended because one of those alternatives got 21% of the vote.

Mick Ireland
515 Indepndence Place
Aspen CO 81611

No comments: