Key Electoral Reform Hearing Tomorrow in California Assembly on AB 1294

The Dickinson article makes these false statements:
1. "IRV ensures that the winner of a single-winner election has the support of the majority of voters in a single election."
COUNTEREXAMPLE SHOWING HOW IRV REFUSES TO ELECT A CANDIDATE WHO WOULD BEAT ANY OPPONENT BY A HUGE SUPERMAJORITY):
http://rangevoting.org/rangeVirv.html#nasty

2. "By eliminating the need for a costly runoff election it saves local governments a lot of money -- about $2M per election in San Francisco alone."
ACTUALLY, instant runoff can cost more money than it saves. University
of Vermont report with cost analysis:
http://rangevoting.org/VermontIrvCost.html

3. "IRV also eliminates vote-splitting and spoiler effects."
COUNTEREXAMPLES SHOWING IRV SPOILER EFFECTS:
http://rangevoting.org/IRVpartic.html
http://rangevoting.org/IRVcs.html
http://rangevoting.org/IRV1519.html

4. San Francisco has used Instant Runoff Voting extremely successfully... survey research shows that the results have been excellent. San Francisco voters understood IRV extremely well...
ACTUALLY: the ballot spoilage rate for SF voters in IRV races is 7 times higher
than in ordinary plurality races, see
http://rangevoting.org/SPRates.html
for data from SF election office.
Also, there was a software problem SF experienced ("extreme success"?)
http://rangevoting.org/SanFranDelay.html

Range Voting is a simpler, cheaper, and better voting method than "instant runoff." See
http://RangeVoting.org

In the vain hope of advancing range voting (vain because it is such an inadequate system), Smith and his handful of similarly sectarian allies like to make these kinds of posts, particularly when hard-working grassroots activists have placed IRV on the ballot or organized support for legislation like the admirable AB 1294 (a bill that is about choice voting as well as IRV, by the way). It's troubling, particulary given how distorted and misleading is their "analyis."

Let me touch on each point:

1. Smith's "spoiled ballot" rate is almost all tied to "undervotes" -- e.g., people who vote in a higher level race but choose to skip over a "down ballot" race. This is common when people are drawn to the polls to vote for high-profile offices like president and governor, but aren't familiar with the down-ballot races. The undervote in some down ballot races around the country will approach 50%. It's absurd to call these "spoiled ballot" because people are deciding to skip over that race.

Nationally the undervote in U.S. House races in presidential years will approach 5%, and in the San Francisco races Smith cites from the presidential year of 2004, the undervote was higher in some state legislative races than the Board of Supervisors races.

2. The "software problem" was a one-time event in the first IRV race in 2004 that simply involved the vendor adding a single line of code that provided more memory when they combined data from the absentee ballots and polling place ballots. The board planned to announce preliminary IRV tallies on Wed. afternoon. They got the change in code certified on Thursday and released the full IRV results to no controversy on Friday. There have been no problems in the city's two IRV elections since and no problems with ballot-counting in cities like Burlington (VT) and Takoma Park (MD).

3. On costs, Smith cites a report about a prospective IRV election that we and others dispute, but conveniently overlooks the fact that IRV is saving San Francisco millions of dollars and was very inexpensively implemented in Burlington and Takoma Park. Once more vendors are certifed to run IRV elections, it will be a relative snap to implement it, and from the first election on it will absolutely save jurisidctions money when going from two elections to one.

4. Smith likes to come up with various theoretical examples of how IRV doesn't generate majority winners nor eliminate the spoiler problem. Relating to majority winners, his concern is for candidates who are weak candidates, but relatively broadly supported as a second or third choice. For most people, a measure of majority support is having substantial first choice support. You're an unlikely "majority winner" if you never could win a pluraly voting race (plurality being where the candidate with the most votes wins). His alternative would alllow someone to win as the "majority candidate" who in fact was nobody's first choice candidate.

The concept of "spoilers" takes on meaning when it affects voter behavior. IRV has been used in literally thousands of govermental elections, is recommended by Robert's Rules of Order for postal elections and is used by more than half of our nation's 30 most highly rated universities. I'd like to see range voting advocates find any examples of news articles during a campaign or candidate statements bringing up their alleged "spoiling."

Let me respond to Richie.

1. "Smith's 'spoiled ballot' rate is almost all tied to 'undervotes' ... it's absurd...
voters decided to skip over this race."
RESPONSE: the overvote rate was typically 7 times higher for IRV than
non-IRV races.
As I already said in the web page I had posted:
http://RangeVoting.org/SPRates.html
It did not count undervotes as spoiled.
As I also said there. Richie simply is wrong.

2. "The 'software problem' was a one-time event in the first IRV race in 2004..."
RESPONSE: Richie in private email to me informed me that here I was "absolutely distorting certain things like the record in San Francisco."
I deny that. I simply quoted the front page article from the pre-eminent
San Francisco newspaper, in my web page
http://rangevoting.org/SanFranDelay.html
I will say, however, that SF did not delay reporting the screwed up
IRV election results for 2 weeks, as
the newspaper article expected, but in fact only a few days.
We actually have a lot of additional interesting data about this story but never fully turned it into a ready-to-post webpage.

3. "On costs, Smith cites a report about a prospective IRV election that we and others dispute, but conveniently overlooks the fact that IRV is saving San Francisco millions of dollars..."
RESPONSE:
This is an interesting claim. Can you back up the claims of $million savings?
A priori, I find that $ amount dubious.

In neighboring Oakland, which is of comparable size
to San Francisco, I believe the pro-IRV group pushing for IRV in Oakland exaggerated predictions of cost savings on their web site in 2006.
They said each runoff election under the old delayed-runoff scheme cost Oakland "hundreds of thousands of dollars," which was their misleading
way of saying $200,000. With IRV that cost would be avoided.
However, they did not mention that Oakland's annual budget is over $1 billion so that the "cost savings" they were lobbying for was of order 0.02% fractionally. Surely there are superior ways to save Oakland's money! Also they did not mention that it cost (neighboring, comparable size) San Francisco $1,600,000 to upgrade its voting machines to run IRV two years before. So
assuming $1.6M up front expense and $200K savings each runoff-avoided, the payback time required to justify this cost "savings," as you can see, would be very large, perhaps 30-40 years assuming elections every 2 years and runoffs required (but now not) half the time. Quite probably Oakland would be re-replacing its machines before that time, in which case the up-front costs NEVER would be repaid, i.e. the "savings" would be a "deficit."

4. "Smith likes to come up with various theoretical examples of how IRV doesn't generate majority winners nor eliminate the spoiler problem..."
RESPONSE: I dislike Richie's use of the word "theoretical"
to try to imply my examples would not occur in the real world. They would.
They in fact already have. It is simply false to claim IRV "eliminates" them. Further, there are analyses on the RangeVoting.org web pages of how often various kinds of nasty examples arise.
For example, the "spoiler" problem, which we call an "FBLE scenario"
(Favorite-Betrayal Lesser Evil) arises in 3-candidate IRV elections 19.6% of
the time in one probabilistic model, and 20.2% in another, but in a sense
occurs 100% of the time. That sense is, that IRV voters acting on imperfect information can be strategically well-advised 100% of the time, to
"betray their favorite" third-party candidate and strategically vote
for their non-favorite "lesser evil" major party candidate. The point is
that although this strategic move will only pay off for them about 20% of the time, it will hurt them a far smaller percentage of the time since third-party
candidates win IRV elections in IRV countries well under 1% of the time, historically. So it's a good gamble. This effect of lots of voters acting that way, of course, will cause third parties to die out under IRV, and 2-party domination to set in, just like now. ("Duverger's law applies to IRV
voting.") And that is in fact exactly what has happened in every IRV country in the world in IRV seats. Australia, the country that has used IRV the most and longest, is particularly revealing. The IRV seats in Australia are 2-party dominated, but the non-IRV (senate) seats in Australia have quite a few third-party members sitting in them. All this is well known to Australian analysts.
For more on this, see
http://www.rangevoting.org/AusIRV.html
http://www.rangevoting.org/USpartyComp.html
http://www.rangevoting.org/AustralianPol.html

Warren D Smith
http://rangevoting.org

Sorry, comments are temporarily disabled. We're doing a bit of server maintenance on the commenting area. We'll be back up and running shortly. Thank you for your patience.