Maine supporters of ranked choice voting are conducting a referendum petition drive, to fight last month’s new Maine law that suspends ranked choice voting until 2021. If the supporters obtain approximately 61,000 signatures in the next 80 days, the October 2017 law will be suspended until the voters vote on it in November 2018. If the law is suspended, then Maine will use ranked choice voting in the June 2018 primary.
The first day the petition could circulate happened to be November 7, 2017, which was local election day in Maine. Supporters of ranked choice voting collected signatures at the polls, and obtained approximately 33,000 signatures that day.
In the last primary for the offices contested in 2018 (legislature and US representative, 2016; and US senator and governor, 2014), a total of 190 offices, 380 nominations:
There were 5 races with 3 candidates. In 3 of those 5, the lead candidate received a majority.
IRV would have mattered in at most 0.5% of the races.
Remember that in Maine, it is illegal for independents to vote in the segregated partisan primaries.
RCV/IRV ignores most of the data in a place votes table.
34 A-M-Z
33 Z-M-A
16 M-A-Z
16 M-Z-A
99
With RCV/IRV, M loses. A beats Z 50-49.
—————
Place Votes Table
— 1 — 2 — 3 — T
A 34 – 16 – 49 – 99
M 32 – 67 – 0 – 99
Z 33 – 16 – 50 – 99
T 99 – 99 – 99
i.e. RCV/IRV will cause even more extremist stuff due to rigged majority *mandate* stuff.
————
Head to Head (Condorcet) Math – from 1780s — repeat 1780s.
M beats A 65-34
M beats Z 66-33
P.R. and nonpartisan App.V. – pending head to head math – will need computer voting to do all the combinations.
Demo Rep,
Let’s look at more realistic elections:
90 A
10 –
100
With RCV, A wins.
SORRY- PLURALITY, TOP 2, RCV/IRV A-L-L HAVE the same sort of FATAL math defects.
in the example what if A = Stalin, Z = Hitler, M = George Washington.
These are SUPER-dangerous times {again] like 1773-1775, 1859-1861, 1914, 1939
— left/right statist extremist monarch/oligarch control freaks on the march — attacking Democracy.
How many 1st round minority/plurality EXTREMISTS have been elected via RCV/IRV rigged majorities in places like San Francisco ???
PR and AppV.
There was only one candidate. We don’t know whether she was Stalinova, Eva Braun, or Martha Washington.
Yeah. Very realistic — one candidate — esp in each rigged gerrymander district.
Just like the old one party commie/nazi regime *elections*.
Of the 380 primaries:
2.4% had 0 candidates.
89.5% had 1 candidate.
6.8% had 2 candidates.
2.3% had 3 candidates.
The cool thing about IRV in 2-candidate races is that if you like one candidate more than another, you can put a 2 next to the other candidate’s name rather than leaving it blank.
NO primaries.
PR and AppV
Things are really rough. Like USA Nov 2017.
49 A-M-Z
49 Z-M-A
1 M-A-Z
99
With RCV/IRV, M loses. A beats Z 50-49.
—————
Place Votes Table
— 1 — 2 — 3 — T
A 49 – 1 – 49 – 99
M 1 – 98 – 0 – 99
Z 49 – 0 – 50 – 99
T 99 – 99 – 99
————
M beats A 50-49
M beats Z 50-49
A = Stalin, Z = Hitler, M = George Washington
Note also the 50 Z in 3rd place — for reverse math.
Just a few months ago the partisan floor leader of the Maine state Senate wanted to be understood that he thought ranked choice voting was un-American. Given what happened to the socialist alternative candidate in the recent city council race in Minnesota, I’m wondering if he’d still maintain that.
I was looking at the Minneapolis mayoral results. Did the election in 2013 produce a weak mayor who faced a mob of challengers and was defeated. That is, did the instant runoff feature produce a lack of scrutiny that would have occurred under the Top 2 system that Minneapolis used to use.
57.7% of Minneapolis voters did not express a first preference for either of the final two.
After transfers were completed:
34.1% said they favored Frey.
28.1% said they favored Dehn.
37.8% said they had no opinion, disliked both, or found IRV too complicated.
The system proposed for Maine, Mainiacal Instant Ranked Voting (MIRV), is different than that used in Minneapolis. Under MIRV, there might be ranked primaries in the DFL primary, though many races would have two or fewer candidates, and then the general election would be by plurality.
One obvious reform — have the candidates rank each other in pre-election day public rank order lists.
If a candidate loses, then his/her votes go to the other highest ranked on his/her list. [with Condorcet math]
A voter votes for ONE candidate for a one person office.
Advanced Condorcet head to head math is like calculus compared to simple addition math.
PR and AppV — pending C math.
@DR,
That is similar to above-the-line voting in Australia. You could also permit a voter to rank more than one candidate, and if the ballot is exhausted, it would revert to the rankings of the candidate. One exploit of either system would be to recruit a bevy of celebrities (former athletes, etc.) who would in turn endorse the real candidate politician.
Another option would be to treat an exhausted as potentially being for the continuing candidate with the fewest votes. If that prevents further exclusions, then all the remaining candidates would appear on a new election ballot in a month. Repeat the process as necessary.
What percentage of IRV votes have been screwed up [so far] — in whole or part ???
— By ANY *extra* / option voting stuff by the armies of voters having under 150-200 IQs ???
— 2 or more *1* votes, *2* votes, etc — votes missing a number — no *1*, *2*, etc. ???
How about having a TOTAL crisis with an IRV vote for USA Prez with about 10 major candidates ???
In Minneapolis, they use a ballot that was designed for using three columns for different races (e.g. Governor in column 1, US Representative in column 2, Legislator in column 3. Instead they use Column 1 for 1st place choice, Column 2 for 2nd place choice, Column 3 for 3rd place choice. If the ballots are scanned at the polling place, more than one 1st place ranking would be detectable as an over vote. While multiple rankings for a candidate could be detected, they aren’t because the voting machine would treat that the same as voting for a governor and legislator who happened to be in the same “row” where there are not actually rows.
Minnesota counties are required to report election night results to the SOS. Hennepin County conducts the elections for Minneapolis, and then turns the IRV ballots over to the city for actual tabulation. The SOS reports the results for each choice.
In Minneapolis there were 104,297 1st place choices; 92,273 second place choices; and 78,188 3rd place choices, so 1/4 of 1st place voters did not have a 3rd choice (approximate, because some voters vote in the 3rd column, or duplicate choices, etc.) There were 16 candidates, plus write-ins so many voters did not understand the purpose of rankings, were confused, were too fatigued to rank so many candidates, or (LWV version) actively chose not to rank more candidates.
Voter: They said it was as easy as 1, 2, , but I forget the third number.
If we consider a ranking of 1, 2, or 3, to be approval, and ignore that some voters may have multi-ranked candidates, the 2nd place IRV candidate was the 5th place approval candidate.
The top 5 (of 16) candidates received 94.7% of the first preferences. The other 11 candidates were eliminated in the first round. Those also-rans had 55061st preferences, 10,075 2nd preferences, and 16,007 third preferences.
In other words, voters voted for Trump or Clinton, and then expressed a 2nd preference for Harold Stassen, or Rocky de la Fuente.
In St.Paul, voters could rank 6 candidates. When the initiative was being considered by the city council, it was reworded to permit voters to vote for 1st, 2nd, 3rd, 4th, 5th, or 6th choices, or so on. This appears to have been offered as a poison pill, and just offered on the spur of the moment, so they couldn’t think of “may rank all candidates”.
61,646 voters had a 1st preference, 46,962 a 2nd, 34364 a 3rd, 16981 a 4th; 9978 a fifth, 6984 a sixth. There were 10 candidates.
The winner in St.Paul received a majority of the votes, so can be considered a legitimate choice.
MN results — one more reason to have each candidate rank ALL of the other candidates
— pending Condorcet math.