Jump to content

Talk:2022 Alaska's at-large congressional district special election

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Orphaned references in 2022 Alaska's at-large congressional district special election

[edit]

I check pages listed in Category:Pages with incorrect ref formatting to try to fix reference errors. One of the things I do is look for content for orphaned references in wikilinked articles. I have found content for some of 2022 Alaska's at-large congressional district special election's orphans, the problem is that I found more than one version. I can't determine which (if any) is correct for this article, so I am asking for a sentient editor to look it over and copy the correct ref content into this article.

Reference named "politics1":

I apologize if any of the above are effectively identical; I am just a simple computer program, so I can't determine whether minor differences are significant or not. AnomieBOT⚡ 15:40, 19 March 2022 (UTC)[reply]

 Done ---CX Zoom(he/him) (let's talk|contribs) 16:42, 19 March 2022 (UTC)[reply]

This article's creation

[edit]

Wikipedia is not a news site, or at least as far as we continue to be told in project space. Was there any real need to create this article before Dunleavy's proclamation? I couldn't help but notice that the article's creators are the usual gang of SPAs who specialize in excessively puffed-up content, often dubiously-sourced, about elections which haven't yet taken place. This means that facts and sources aren't settled. They never bother to stick around any of these articles once that is the case. Giving them such free reign to do this means that we are effectively creating a news site and not recognizing the difference between that and an encyclopedia. Gutting the article's content certainly appears to be justified, but it's insulting to look at a bunch of subject headers and templates containing no meaningful content. It's even more insulting to give this higher a priority than long-notable topics with long-settled facts and sources. RadioKAOS / Talk to me, Billy / Transmissions 22:57, 19 March 2022 (UTC)[reply]

I agree with you. This is not supposed to be news. All the frenzy to document daily changes before the election just clutters up coverage of the actual process and event. Your concerns here seem also to apply to the article on 2022 United States House of Representatives election in Alaska. There is so much info in that on people who were rumored to be filing, were suggested or encouraged, but then are listed as 'withdrawn', declined, or otherwise did not even participate in the blanket primary. I don't think these articles should spend time on people who did not even run in the primary. (Let people writing on those individuals include info if they think it significant.)Parkwells (talk) 21:20, 10 June 2024 (UTC)[reply]

But wait, there's more

[edit]

John Coghill filed Monday morning. Curiously, I'm finding no coverage. I learned about it through a mutual friend on Facebook, which seems to be just as valid as some of the sources I'm currently seeing in the article (e.g. blog posts and tweets). RadioKAOS / Talk to me, Billy / Transmissions 05:09, 29 March 2022 (UTC)[reply]

@RadioKAOS: Forgot to reply, but there is news coverage for this, which I've added. Also, I don't see blog posts and tweets being used as sources in this specific article, which would be unreliable. —twotwofourtysix(My talk page and contributions) 10:48, 29 March 2022 (UTC)[reply]
As far as my Google search of several minutes ago is concerned, the headline in the Anchorage Daily News reads "Gross, Coghill say they plan to run for Alaska US House seat". This story was published 16 hours after Coghill actually filed. While many of us are acutely aware of how far local newspapers have fallen from grace, this is not very "reliable" from a regional newspaper of record in a world where so many are obsessed with 24/7, up-to-the-minute news (with Wikipedia doing its best to play along).
Since you say "I don't see", I'll happily critique the sources and the development of the article in general. The first issue is the separate "Declared" and "Filed paperwork" sections. Declared according to whom? There's a severely abbreviated filing window and a recent browsing of the FEC website turned up nothing specific to this election. Without that verification, the "declaration" is solely in the mind of the individuals listed and/or journalists friendly to them, using the mere existence of reliable sources to justify inclusion. In other words, I could declare myself to be a candidate and it would be just as valid if I conned a media outlet into taking it seriously, even if I filed no paperwork. It smacks of WP:SYNTH to me. More importantly, the brevity of the filing window means the former section will cease to be of significance in just a matter of days. Quit being a slave to the same old MOS when it's not doing the topic justice. With that in mind, why are we creating a gallery for "declared" candidates but not affording the same courtesy to actual verified candidates, those who appear in the following section? Sounds promotional and a serious NPOV violation to me. The article is riddled with factual errors throughout, including those easily verified by actually reading the sources present in the article.
Now, on to the sources:
  • Source 1 — While Fox News is nominally a RS, if one browses the revision history and talk page of Young's article, you'll realize Fox initially reported that Young died at the terminal at LAX, while everyone else reported that he died on the flight to SEA or was declared dead after the plane landed. It's being used solely to source the fact that Young died and there exists a ton of higher-quality sources for that.
  • Source 4 — This truncated AP story says nothing that isn't already covered in source 2, the ADN story by Brooks and Herz. I suppose it would suffice if it were tacked on as an additional source, though.
  • Source 6 — As I refer to above, this AP story published by the SF Chronicle is the same story the ADN published today. It appears the idea is to prove that the topic is receiving coverage from a broad range of sources, even if that coverage can be traced back to the same small handful of Alaska-based journalists. It's doesn't appear to be an effort to reflect the highest-quality coverage.
  • Source 7 — The list from the Division of Elections was last updated on Friday. Brelsford filed but that isn't reflected in the article. Coghill is known to have filed but that isn't reflected in the list. There's no coverage about Revak actually filing that I could find. Which brings us to...
  • Source 8 — Must Read Alaska is a "news blog", of which there are many in Alaska (The Alaska Landmine, The Alaska Watchman and Midnight Sun AK are other popular examples). There may be credentialed journalists involved, but they are considered blogs and exist to influence elections and the political process. As I recall, WP:RSN declared several years ago that subcommunities of editors simply can't declare something to be a reliable source without vetting it through them. As MRAK is used dozens of times throughout the encyclopedia, it's clear they're not doing their job in that respect.
  • Source 9 — This story written by Herz mostly rehashes the earlier story he co-wrote.
  • Source 10 — Move along, nothing to see here. The same as numerous other instances over the past decade of journalists prodding Palin for a quote about a prospective campaign which never materializes. We're supposed to treat this one as if it exists in a vacuum?
  • Source 11 — Paywalled and not marked as such. Looks like there's statements which can't be verified without a subscription.
  • Source 12 — Reads like an editorial and not a reporting of facts.
  • Source 13 — Same as source 11.
Hope this helps. I saw multiple tweets in earlier revisions, and the same editors who've been adding such "sources" to elections articles for quite some time. I guess they've already been dealt with. RadioKAOS / Talk to me, Billy / Transmissions 21:04, 29 March 2022 (UTC)[reply]

Rep. Young under candidates

[edit]

Rep. Young was running in the regular election, not this special election that was created due to his passing. He wouldn't have contemplated running in a special election created by his passing, so I don't think he should be listed at all under "Candidates"(even if as decesased). 331dot (talk) 15:27, 2 April 2022 (UTC)[reply]

A Commons file used on this page or its Wikidata item has been nominated for speedy deletion

[edit]

The following Wikimedia Commons file used on this page or its Wikidata item has been nominated for speedy deletion:

You can see the reason for deletion at the file description page linked above. —Community Tech bot (talk) 19:53, 5 April 2022 (UTC)[reply]

The less I have to do with ANI, the better. However, it's obvious that some of you have been given free reign to do your own thing, escaping the notice of the community at large, for long enough. First off, people are showing their true commitment to NPOV by repeatedly inserting copyvio photos of Nick Begich and Al Gross into this article, all the while there has been a properly-licensed photo of John Wayne Howe on Flickr for nearly a year and a half. Now that it's on Commons, you still are going nowhere near it. I'll be as clear about this as I can: NPOV MEANS WE ARE NOT HERE TO PROMOTE CERTAIN CANDIDACIES SIMPLY BECAUSE MONEY IS BEING RAISED AND SPENT. That's how articles on current elections have been built on Wikipedia for years and years. Ridiculous. Secondly, this should have been tagged with {{Historical election article}} from the start. The fact that it wasn't further shows evidence of some people's true commitment to honoring Wikipedia's core principles. What was the point of adding names to this article in the beginning which are no longer present? Please answer in a way which respects WP:NOTNEWS and WP:OR/WP:SYNTH. RadioKAOS / Talk to me, Billy / Transmissions 20:05, 5 April 2022 (UTC)[reply]

A Commons file used on this page or its Wikidata item has been nominated for deletion

[edit]

The following Wikimedia Commons file used on this page or its Wikidata item has been nominated for deletion:

Participate in the deletion discussion at the nomination page. —Community Tech bot (talk) 20:37, 7 April 2022 (UTC)[reply]

So.. Sattler or Peltola?

[edit]

@Thomascampbell123: Many news outlets, especially during the campaign, say that her surname is Peltola, including her FEC filing and the Alaska Division of Elections. More about this at Talk:Mary Sattler#Requested move 17 June 2022 and section above, please join the discussion there. I just don't wanna cause confusion to readers looking for "Peltola" only finding "Sattler", including my map only having "Peltola" in it. —twotwofourtysix(My talk page and contributions) 11:18, 18 June 2022 (UTC)[reply]

+1 Peltola: Sources on this article are using Peltola to refer to her, when using only last name. —CX Zoom[he/him] (let's talk • {C•X}) 08:47, 20 June 2022 (UTC)[reply]

I think that Sattler should by used unless both surnames are added to her Wikipedia page. Thomascampbell123 (talk) 00:42, 21 June 2022 (UTC)[reply]

Ordering of Candidates

[edit]

Hi all,

I'm new to Wikipedia, so I just have one quick question. How are candidates ordered, particularly in the section listing candidates who advanced to the primary? It seems alphabetical, and is this always the standard across wikipedia pages?

Thanks, Vergilreader (talk) 02:01, 19 June 2022 (UTC)Vergilreader[reply]

Yes, lists of candidates in US election articles, and indeed election articles for many other countries, are usually sorted by last name. —twotwofourtysix(My talk page and contributions) 06:00, 19 June 2022 (UTC)[reply]

Gross withdrawal + Sweeney advancing update

[edit]

While there's little doubt that Al Gross won't appear on the general election ballot. This AP article quoted a spokesperson from the Alaska Division of Elections saying that she was still looking into whether the fifth-placed candidate (currently Tara Sweeney) would actually advance into the general election and had no answer. Not to mention that Sweeney can/may be overtaken by another candidate, eg. Santa Claus, although unlikely. However, the Alaska Landmine that first broke the story seems to be very sure about the things I mentioned here. Thoughts? —twotwofourtysix(My talk page and contributions) 04:44, 21 June 2022 (UTC)[reply]

@Stroopwafels, 02rufus02, and David O. Johnson: Courtesy ping to interested editors —twotwofourtysix(My talk page and contributions) 04:57, 21 June 2022 (UTC)[reply]
Can we add a note that while reliable sources have indicated Sweeney is likely to be on the ballot, the Alaska Division of Elections is yet to clarify if a fifth placed candidate may advance into the general when one of the top four vote getters withdraws? TulsaPoliticsFan (talk) 05:38, 21 June 2022 (UTC)[reply]
Added a note here to that end. Stroopwafels (talk) 07:43, 21 June 2022 (UTC)[reply]

Ordering of Endorsements

[edit]

Hi, I have a question regarding Sarah Palin’s endorsers. Shouldn’t we have Trump’s endorsement first before Haley’s? Or is this in chronological order? Cheers. -Conservative Alabamian (talk) 21:13, 22 June 2022 (UTC)[reply]

It is in alphabetical order by last name, so Haley comes before Trump. Cheers! —CX Zoom[he/him] (let's talk • {C•X}) 21:16, 22 June 2022 (UTC)[reply]

Round 2 results map

[edit]

So it looks like we're not going to get any results breakdown by House district or precinct for round 2, ie. just a matchup between Peltola and Palin. If that's the case, should we keep it to the first round map only or should I add a simulated second round map based on uniform preference flows applied to all district (which might be speculative)? I'll make a first round results map by precinct soon. —twotwofourtysix(talk || edits) 09:12, 4 September 2022 (UTC)[reply]

Don't do a uniform flow map, that could be factually wrong, and even if it weren't WP:NOR probably applies. That said, I'm perfectly okay with general election's first round only results if second round results are not yet available. —CX Zoom[he/him] (let's talk • {C•X}) 15:12, 4 September 2022 (UTC)[reply]
So I just uploaded the more official (?) second round map. Explanation's over on the file page, if you're ok with it. —twotwofourtysix(talk || edits) 15:33, 11 September 2022 (UTC)[reply]
Looks great to me :-) —CX Zoom[he/him] (let's talk • {C•X}) 17:06, 11 September 2022 (UTC)[reply]

A Commons file used on this page or its Wikidata item has been nominated for speedy deletion

[edit]

The following Wikimedia Commons file used on this page or its Wikidata item has been nominated for speedy deletion:

You can see the reason for deletion at the file description page linked above. —Community Tech bot (talk) 21:22, 18 September 2022 (UTC)[reply]

Edit reversion of results table

[edit]

@CX Zoom

You have reverted my edit to the results table. Please see earlier discussion at https://en.wikipedia.org/wiki/Talk:2022_United_States_House_of_Representatives_election_in_Alaska#Total_votes_tallly for earlier discussion on the format of results tables for Alaska's RCV races.

Note in particular for this article, the "Election Summary Report" differs from the Round 1 results on the RCV Tabulation, so it is inaccurate to report the summary results as the round 1 results. If you have objections to this format, please bring them up at the other talk page (to keep discussion centralized) so we can reach a consensus. Thanks, 71.162.7.170 (talk) 16:51, 12 February 2023 (UTC)[reply]

On IRV pathologies

[edit]

@108.30.55.246 Just wanted to say that if you feel there's a need for edits to the paragraph discussing the IRV pathologies in this election, best to discuss first on the Talk page. –Sincerely, A Lime 16:43, 28 May 2024 (UTC)[reply]

As I’ve said, there’s no issue with using mathematical terminology, though I will caution that this is Wikipedia so it should be understandable to most readers. What I do have a problem with is only criticizing ranked choice voting (or this version of it) without providing a balanced discussion. That’s an opinion that shouldn’t be framed as fact. Stormy160 (talk) 22:10, 28 May 2024 (UTC)[reply]
I'm happy to add any additional analyses you'd like from mathematicians or social choice theorists, but FairVote almost-certainly falls under WP:PARTISAN and wouldn't be considered a WP:RS. –Sincerely, A Lime 22:12, 28 May 2024 (UTC)[reply]
We can use a news source instead of FairVote. Additionally, saying “it generated discussion” summarizes something that happened as a result of the election (the discussion), as opposed to just presenting one opinion as if Wikipedia believes it. Stormy160 (talk) 22:14, 28 May 2024 (UTC)[reply]
Yes, but the mathematical pathologies weren't "discussion"—they're mathematical facts. We can see that subtracting 5% from Republicans' vote totals would have led to Begich winning, and we can also see that most ballots preferred Begich to Peltola (and an overwhelming majority preferred Begich to Palin). –Sincerely, A Lime 22:20, 28 May 2024 (UTC)[reply]
First of all, that’s based off of one analysis that is based on a survey it’s not a set in stone fact. Second, plenty would argue that it doesn’t matter if the majority preferred Begich because they didn’t express that in the first round. I don’t personally care in this instance which is correct or more mathematical but you can’t just bash the system. Stormy160 (talk) 22:23, 28 May 2024 (UTC)[reply]
It's not based on a survey, it's based on the cast vote records, i.e. the actual ballots. On ~52% of ballots, Begich was ranked higher than Peltola (and on ~60% he was ranked higher than Palin). This is the Condorcet matrix you can see at the bottom. It's not based on polling (though polls also confirm this). I believe I gave 3 citations, but if you want I can add more, since the analysis has been replicated a few times in different papers. (I just think it'd be overkill.)
I'm not also not "bashing the system". I'm pointing out that A) social choice theorists criticized this result (they did) and B) explaining why they criticized it.
Second, plenty would argue that it doesn’t matter if the majority preferred Begich because they didn’t express that in the first round.
I'm sure some people would; the argument that only the first-round preferences matter is basically the argument for first-past-the-post and all of its variants (including sequential loser plurality). I'm not sure how it would fit in here, though, because that sounds like an opinion. (Assuming I understand what you mean; preferences aren't expressed in rounds, they're expressed as soon as you cast your ballot). Still, that doesn't change the fact that social choice theorists criticized it for the reasons I laid out. –Sincerely, A Lime 22:42, 28 May 2024 (UTC)[reply]
That's actually not what I was thinking, I was actually referring to the argument that rewarding second and third place votes in the first round disincentivizes people to put a second or third choice since it could hurt their first choice. But there you go, more proof that there are many angles to this. The latest version is an improvement but it still needs to be framed as a summary of the discussion created by this election, not as an opinion. It still reads very negative, and IRV is not a common acronym it should just be referred to as "ranked-choice voting" because this is what most people think of when they think of ranked-choice voting. 108.30.55.246 (talk) 01:03, 29 May 2024 (UTC)[reply]
Also, there are some questionable things in there. You cannot say that a majority were opposed to Peltola is factual because she got a majority in the final round. Palin being a spoiler is a valid but debatable argument, but that doesn't automatically mean that every Begich voter who put Peltola second did it despite their disapproval. That's why I put it in terms of favorability ratings instead. And plurality voting with partisan primaries still would've most likely produced a Peltola vs Palin matchup. I also don't understand the last bit about voters who put Peltola last - that has no effect on the election. Stormy160 (talk) 01:10, 29 May 2024 (UTC)[reply]

Also, there are some questionable things in there. You cannot say that a majority were opposed to Peltola is factual because she got a majority in the final round.

Opposition refers to number of voters who would prefer some other candidate in one-on-one match; idea of a "final round" is a bit of an illusion (in IRV there are no rounds, it all happens simultaneously; "simulated runoff voting" would've been a better name). You can create a fake "majority" in a final round for any candidate, by eliminating all the popular candidates in early rounds; this is why standard definition of opposition in social choice is to take the worst-case opposition.

That's actually not what I was thinking, I was actually referring to the argument that rewarding second and third place votes in the first round disincentivizes people to put a second or third choice since it could hurt their first choice.

I'm not sure how that's related here, though I'll note (as discussed at later-no-harm criterion) that failing to reward 2nd/3rd place votes in the 1st round can also disincentivize putting them down. (Like here—for Palin voters, the 2nd preferences had no effect on the election and did not help Begich, so no reason to waste time ranking anyone besides Palin. So, many voters don't bother—30% in typical election I think.)

The latest version is an improvement but it still needs to be framed as a summary of the discussion created by this election, not as an opinion. It still reads very negative

I think it would be very hard to find positive commentary on this from any experts (social choice theorists). The commentary was all negative because this is the kind of race where IRV and FPP do very poorly, called a center squeeze. In polarized electorate, both systems tend to elect "extremists"—in quotes because Peltola is very moderate, but in low-turnout Alaska race, electorate is so conservative she might as well be communist. :p The median voter and majority went for Begich here, but he lost because the vote was split with Palin.

and IRV is not a common acronym it should just be referred to as "ranked-choice voting" because this is what most people think of when they think of ranked-choice voting.

Wikipedia still uses instant-runoff voting, since ranked choice voting is ambiguous and bit of misnomer. The IRV page notes sources discourage the term RCV because it . It also tricks people into thinking IRV is like other ranked voting methods (Condorcet methods), when it shares most of its properties and "flavor" with FPP (e.g. fails the median voter theorem).

And plurality voting with partisan primaries still would've most likely produced a Peltola vs Palin matchup.

Yep, that's correct. Criticisms of IRV from social choice theorists boil down to "doesn't go far enough". IRV falls in the same broad family of voting systems as FPP. Has all the same properties, e.g. lesser evil voting—for conservative Republicans, the honest vote for Palin let Peltola win (but voting Begich would have elected him).

I also don't understand the last bit about voters who put Peltola last - that has no effect on the election.

Palin voters let Peltola win: Palin eliminates Begich in first round. If not for Palin's supporters, Begich would defeat Peltola. Another way to put it is, if Alaska was more liberal, a Republican would've won the election. See Monotonicity criterion and Participation criterion. These are called negative voting weight events, because your ballot does the exact opposite of what you tell it to. You tell it to support Palin first and Begich second, but instead your ballot elects Peltola. This is what makes voting in IRV complicated or confusing: you can't rank candidates in order of preference, the order you put them on your ballot doesn't correspond with who benefits from your vote. –Sincerely, A Lime 02:50, 29 May 2024 (UTC)[reply]
Wow, I can tell that you detest ranked-choice voting/IRV. It's an interesting conversation and I'm not trying to dispute your points, but maybe because you are so opinionated it's not your place to be editing this. It is not Wikipedia's job to judge the system. If there was a discussion by others about whether the system worked, Wikipedia can describe what arguments were made. I believe the current version provides a nice survey of experts, pundits, and politicians with a range of views. It is not Wikipedia's place to make its own argument. Stormy160 (talk) 02:58, 29 May 2024 (UTC)[reply]
I'm not particularly emotionally involved, and I wouldn't really describe myself as detesting the system. It certainly has some advantages. You could even argue this center-squeeze property is good thing because extremists are the most passionate. This was a popular argument for political scientists until ~20 years ago, though it never really caught on in social choice.
I'm just trying to summarize the social choice commentary on this election, which is all very negative because it's a very good example of a center-squeeze. This is just the kind of setup where IRV does poorly (as every ranked system, including FPP, must in some situations). For IRV, that situation is when you have two candidates on the wings with a moderate in between. In other cases like the Maine elections in 2018, IRV did its job perfectly fine by eliminating minor-party spoilers. But with IRV, all the pathologies tend to crop up at once in this kind of race. So if you summarize how IRV performed in the election accurately, it looks almost like you have to be making it up, because everything goes wrong all at once :p
If you are interested in the topic, I recommend Moulin's textbook Axioms of Cooperative Decision Making or Fishburn's The Theory of Social Choice, which will go over all the commentary from voting theorists on this specific class of elections. –Sincerely, A Lime 03:47, 29 May 2024 (UTC)[reply]
Once again, this is very interesting, but not for this article. Stormy160 (talk) 04:08, 29 May 2024 (UTC)[reply]
I don't see how discussing things election scientists demonstrated happened under this new system wouldn't be relevant to this article. It would be like trying to remove commentary on the spoiler effect from the 2000 Florida election article.
On your request, I attempted to add contrary commentary by pundits as well. (Though I find its quality or relevance—on both sides—dubious, given their lack of expertise.)
However, my current concern is you keep attempting to replace the verifiable claim that a majority of voters opposed Peltola with the unverified claim that Peltola had a lower favorability rating than Begich (which we cannot determine from the ballots, and which is not stated anywhere in the sources supposedly cited).
What we know for certain is Peltola faced opposition from a majority of voters, under the common definition of opposition in social choice: 52% of voters ranked her either last or tied for last (i.e. a majority least-favorite), and more than half of voters would have opposed her in a runoff with Begich (pairwise opposition). Noting this is not particularly biased. The election did in fact spur substantial criticism of IRV from social choice/voting theorists.
If you can find any examples of positive commentary in a reputable mathematical or social choice journal, I'd be happy to add them. However, I was unable to find any in the usual social choice journals. –Sincerely, A Lime 17:26, 30 May 2024 (UTC)[reply]
One cannot say that Peltola lacked majority support because she got it in the final round. It’s simply a false claim. Saying that Begich had a higher favorablity rating strikes a middle ground between that bogus claim and the claim that Begich was closer to the center of opinion, which has merit but is harder to prove. Stormy160 (talk) 00:44, 31 May 2024 (UTC)[reply]

When the Alaska Repeal Top-Four Ranked-Choice Voting Initiative gets its own Wikipedia article, that paragraph should go right back in, because it is this election that was the impetus behind the repeal movement (just as the similar pathologies in the 2009 Burlington mayoral election was behind that repeal in 2010). I am looking at the numbers and I do not see non-monotonicity demonstrated in the Alaska election. I specifically do not see support for the statement that Begich lost because he got too many votes. Non-monotonicity means that specifically regarding a candidate who otherwise wins, because some number of voters change their vote to vote that candidate these additional vote actually causes the candidate to lose. In the Burlington 2009 election a hypothetical non-monotonicity scenario can be cooked up (if 741 Wright voters had changed their vote to Kiss, they would have caused Kiss to lose) but I surely do not see it in this election.

It would be appropriate to place this discussion in its own section of the anomalous election with numbers and citations. Can we get consensus for that? 146.115.164.202 (talk) 00:00, 9 July 2024 (UTC)[reply]

RFC: No-show paradox

[edit]

Should the article discuss the occurrence of a no-show paradox, up-is-down (monotonicity) paradox, and center-squeeze in this election? –Sincerely, A Lime 18:00, 30 May 2024 (UTC)[reply]

@Closed Limelike Curves: I can't work out what this is about. Have you exhausted WP:RFCBEFORE? Also, |Voting systems is not a valid Rfc category - did you not see the big red error message? --Redrose64 🌹 (talk) 20:42, 30 May 2024 (UTC)[reply]
(Summoned by bot) Comment - As with Redrose above, I'm not sure what this is referring to, although I'm not the most technical person. Have you discussed the matter with other parties? Bandit Heeler (talk) 23:05, 30 May 2024 (UTC)[reply]
You can see the talk above, and the most recent version I wrote, before I to try and put an end to the edit war after I got reverted a few times by Stormy160. And yes, very sorry about summoning y'all into a fairly mathematically-heavy discussion 😅
I'm trying to include information and commentary from mathematicians and social choice theorists on Alaska's new version of RCV (Instant-runoff voting), but Stormy160 keeps reverting my edits, referring to them as biased. I tried a few times to come up with a compromise, but it doesn't look likely to happen now so I figured I needed more comments.
The main place I tried to get comments from was in the Voting Systems WikiProject (sorry for misusing the template, I thought I could use a WikiProject as a category!). I think they're the people most likely to have the relevant expertise for this question. –Sincerely, A Lime 23:19, 30 May 2024 (UTC)[reply]
In short, this election generated a lot of commentary from mathematicians, because it shows a lot of the glitches found in Alaska's particular implementation of RCV. Basically:
1. A majority of voters ranked Begich above Peltola, but Peltola still won. (So Peltola won even though most voters preferred Begich; as the articles I had cited in earlier versions stated, this happened because of a spoiler effect where Palin split the Republican vote by failing to drop out.)
2. If Republican turnout had been lower, i.e. If Republicans got fewer votes, Peltola would have lost to Begich (a no-show paradox).
I think both these facts are important and interesting enough to be included in the article. However, Stormy continues to replace these summaries of the papers I cited with references to Begich having a higher favorability rating. So far as I can tell, this is not at all supported by the papers I cited, which say nothing about favorability ratings. The papers are exclusively about the no-show paradox and majority-reversal paradoxes in the election. –Sincerely, A Lime 23:29, 30 May 2024 (UTC)[reply]
@Closed Limelike Curves: I said I can't work out what this is about because this RfC does not provide any context whatsoever, see WP:RFCNEUTRAL. I have searched the whole of this page for the words "no-show" and "paradox", and the only place that either one of them occurs is in this thread. If there was previous discussion, you should have linked it. --Redrose64 🌹 (talk) 07:14, 31 May 2024 (UTC)[reply]
Very sorry about that—I was trying to get the attention of the voting systems WikiProject (who I assumed would know those terms). "No-show paradox" is another term for a participation failure. –Sincerely, A Lime 17:53, 31 May 2024 (UTC)[reply]
To attract the attention of a WikiProject, the correct thing to do is drop a note at their talk page. I see that you already did that at Wikipedia talk:WikiProject Voting systems#2022 Alaska's at-large congressional district special election has an RfC. --Redrose64 🌹 (talk) 20:59, 31 May 2024 (UTC)[reply]
I’ve reviewed the discussion above (more heat than light, I’m afraid) and the recent edit history. Suffice it to say that I believe the inclusion of information on the no-show paradox, monotonicity paradox, and center-squeeze would be beneficial. I don’t believe it’s appropriate that this information is removed from the article.
I’m going to edit the article and hope that the editors can behave and work out a better version of the article from here. I don’t want to have to protect the article; we can be civil. In particular, Stormy, if you have information showing beneficial aspects of IRV as concern this election, by all means add them in. (I’m not even too worried about sourcing; if some good aspect of IRV was apparent in this election, a paper describing the effect itself without reference to this election would be fine with me, as long as its application falls within WP:CALC.)
CRGreathouse (t | c) 01:21, 31 May 2024 (UTC)[reply]
As I’ve said all along my problem was never providing information on those concepts, it’s the use of a Wikipedia article to argue a certain perspective. I strongly feel it should be framed as “the election generated discussion on instant runoff voting” because that’s neutral language that doesn’t suggest anything. And that’s why I added in what others said about voters crossing partisan lines by voting Begich first and Peltola second. That way there is a survey of the resulting discussion as opposed to just one side. That cannot just be discredited, that’s a clear instance of bias. I condensed the mathematical stuff only for clarity so that the lede isn’t too long and technical, I tried to keep the point being made the same as before. Stormy160 (talk) 01:30, 31 May 2024 (UTC)[reply]
I think that's a reasonable goal. However, the newer versions didn't accurately summarize the main objections experts raised to IRV in this race, which were:
  1. Peltola won with a minority of the vote—52% of voters preferred Begich to Peltola, and a similar number (52%) refused to support her in the final round.
  2. Palin spoiled the election, by knocking out the majority-preferred winner (Begich).
  3. Begich lost because he got too many votes.
"Generated discussion" feels like an attempt to gloss over some very harsh criticism election scientists gave. "Generated controversy" seems like an accurate and neutral description. (Many people were definitely upset about the outcome!) I made some efforts to compromise on this by including commentary by politicians or pundits, although it feels like the result was to make the lead a bit bloated. However, WP:NPOV doesn't mandate equal coverage of both sides of a discussion: it requires coverage to reflect the views of experts in the field and other WP:Reliable Sources. In this case, while many political offered some kind of spin on the election, the consensus of academics was that this election went very badly for IRV. That doesn't mean IRV is worse than FPP or traditional primaries, or that IRV is bad across all elections. However, it does mean the Wikipedia article for this particular election has to reflect that consensus. –Sincerely, A Lime 02:42, 31 May 2024 (UTC)[reply]
Based on what is sourced, I don't think the criticism is that harsh. Peltola was also a moderate candidate, I don't think anyone argues that she is that far off from the median voter. I also think discussion is the right word, because controversy kind of implies notoriety. Sure, there were some comments made particularly by Republicans about the outcome, but I don't think this was some widespread thing that many people talked about. It was something discussed by a group of academics, who provided criticism. It was also discussed by others with a mix of praise and critique. And finally, my aim wasn't balanced coverage per say (there is more criticism than praise in there), just a neutral survey of what was said. Stormy160 (talk) 03:11, 31 May 2024 (UTC)[reply]
But on the spectrum of the three candidates in the general election, Peltola was clearly the candidate on the Left and Palin was clearly the candidate on the Right. Leaving exactly one space for Begich to occupy and that is of the center, between Peltola and Palin. This is not subjective. Begich got far more second-choice votes from Palin voters than did Peltola. And Begich got far more second-choice votes from Peltola voters than did Palin. Stormy, I realize you don't want this information in there, but this is salient because of the 2024 repeal question. 146.115.164.202 (talk) 00:07, 9 July 2024 (UTC)[reply]
The 2024 repeal could be just as much about the 2022 senate race. And strongly challenge that I have shown any bias here. I have consistently criticized how this has become like a research project as opposed to a factual Wikipedia summary and have repeatedly incorporated edits that have sought to increase the accuracy of what is said. And on Peltola being moderate, I'm not sure I've seen anyone argue that Peltola is left wing in the same way that Palin is right wing. There is just no truth to that, and Begich is not solely a centrist, but as the article already says he is probably closest of the three to Alaska's right of center views. I think the idea behind Yang's quote is that Peltola is more moderate compared to Palin, and that there is no consideration for Begich because he did not reach the final round. Stormy160 (talk) 00:22, 9 July 2024 (UTC)[reply]
Yeah, I could see the massive favorite-betrayal/tactical voting campaign for the 2022 Senate race turning voters off of IRV too, once they saw the "Vote your conscience" claims about IRV weren't true. (Democrats were forced to abandon their candidates to make sure RCV didn't eliminate Murkowski in round 1 like it did Begich, making it a one-party race.) I'd be interested in seeing that covered too for sure if you have sources on it, but that definitely doesn't belong in this article.

There is just no truth to that, and Begich is not solely a centrist, but as the article already says he is probably closest of the three to Alaska's right of center views.

Yeah, Peltola happens to be the "moderate" in this race by the common-language definition, but RCV didn't deliver any improvement in this race over FPP-with-primaries. Closed Limelike Curves (talk) 00:37, 9 July 2024 (UTC)[reply]
That's an opinion. Stormy160 (talk) 00:42, 9 July 2024 (UTC)[reply]
I'm not sure what you're referring to here. This is just a fact about the ballots and candidates. If you limited the runoff to one candidate of each party, as in a traditional primary, the winner would have been exactly the same, and the cited papers show this. The only exception is I suppose it's theoretically possible that the higher turnout (compared to the general) hurt Begich, in which case RCV would be the cause of these pathologies, instead of just having failed to stop them. That's definitely reaching into speculation, though. Closed Limelike Curves (talk) 01:00, 9 July 2024 (UTC)[reply]
See, the discussion wasn’t about whether these concepts should be in the article. It was about not letting one’s opinion be the driving force behind the text. Lime’s edits painted an overwhelmingly negative view of instant runoff voting that wasn’t encyclopedic, and any mention of something positive about it (such as Yang’s commentary, which I added in) was immediately followed by something meant to discredit it. Again, I have no issue with introducing mathematical concepts in an easy to read way, but the purpose of this article is to summarize this particular election, not to formulate an opinion on instant runoff voting. The election generated discussion on the efficacy of the system, and that’s something worth noting because it’s a direct result of this election. But we should not be turning this into an opinion piece or one sided analysis. Also, center squeeze is already mentioned in the current version of this article, and it provides a link to the relevant page so that we don’t have to delve into all the details. This page already applies the theory to this specific election by labeling Palin as the spoiler. Stormy160 (talk)

RFC: Majority opposition

[edit]

Is it correct to refer to Peltola as having majority opposition, on the basis of having a majority of voters opposed in a runoff with Begich; as well as being left either unranked or ranked last on a majority of ballots? –Sincerely, A Lime 18:02, 30 May 2024 (UTC)[reply]

Same as above. --Redrose64 🌹 (talk) 21:10, 30 May 2024 (UTC)[reply]
@CRGreathouse An absolute majority of voters gave Peltola no support (i.e. unranked or ranked last) and a relative majority ranked her beneath Begich. It would be typical to describe this as majority opposition, no? –Sincerely, A Lime 15:48, 2 June 2024 (UTC)[reply]
Ranked last could be support, it's entirely plausible for a voter to like all three candidates. Additionally, there is a jump between not voting for someone and opposing them. Stormy160 (talk) 17:21, 2 June 2024 (UTC)[reply]
It's entirely plausible for a voter to like all three candidates, but liking a candidate is not the same thing as giving support to a candidate. A candidate who is ranked last or tied-for-last (truncated) on a ballot is one who the voter believes should not receive any kind of support from them, regardless of who their opponent is. –Sincerely, A Lime 18:22, 2 June 2024 (UTC)[reply]

(invited by the bot) This RFC is written "for insiders" in a way that any respondents would need to take a deep dive in reading the article and talk page to fully understand the question and give a thoughtful answer. If you are seeking more outside participation, suggest adding more of an explanation. Sincerely, North8000 (talk) 14:25, 5 June 2024 (UTC)[reply]

In terms of the RFC, I believe that it is "incorrect" and a misstatement to say that Peltola had majority opposition. I agree with User:Stormy160 that it is a jump "between not voting for someone and opposing them." It is not typical to describe early Republican voter preferences for their own party candidates strictly as "opposition" to another. A Lime appears to consistently overlook the final result of this election. According to the state's official election results, which are the only ones that count, Peltola received 51.5% of the vote, defeating Palin. Arguing about how things stood in the first round of the runoff is beside the point. That seems an attempt to pick apart the process and look at an early part of the election, rather than the final result.Parkwells (talk) 22:42, 8 June 2024 (UTC)[reply]
The final result was Peltola receiving 48.4% of the vote, whereas Palin receive 44%; the remainder was held by voters who supported Begich and refused to support either Palin or Peltola.
There is a huge difference between not voting for someone and opposing them, but if there's still objections to this terminology, I'll rephrase it to the more-precise "ranked last". –Sincerely, A Lime 00:59, 9 June 2024 (UTC)[reply]
@North8000—I actually wasn't looking for comments from politics observers, which is why I condensed this into technical terminology. The terms I used should be familiar to people with experience studying voting systems. The relevant category for this is Mathematics—or at the very least economics and mechanism design—rather than Politics/Government/Law. –Sincerely, A Lime 00:55, 9 June 2024 (UTC)[reply]

I was not going to participate in this discussion, but read enough of it to go further and try to understand more. I have made some copy edits to the article in an effort to express a more neutral and factual tone. In my experience, it is unusual to have articles about elections express voters' preferences for a candidate(s) by stressing their "opposition" to another. Have tried to make my way through this article, the Talk page, and associated articles about alternative election systems in order to get more perspective.Parkwells (talk) 22:42, 8 June 2024 (UTC)[reply]

(cont'd) I was concerned that both the Mansky/Foley and the Atkinson/Ganz critiques cited in the article (and which A Lime repeatedly refers to also on the Talk page) were expressed in OpEd or Opinion pieces in their sources, the Washington Post and The Hill. The scientists' viewpoints were not reported or discussed by reporters of either source as part of an overview of specialists of the election, nor were they published in scholarly journals under peer review. (I added the Opinion designation to each cite). Since this was the first election in AK under the ranked-voting system, perhaps other sources have reported further on these and other scholars in a way to give broader insight.Parkwells (talk) 22:42, 8 June 2024 (UTC)[reply]

The news sources filed them as opinion pieces, but I cited scholarly journals that support Maskin and Atkinson's commentary, which is basically the consensus of experts in social choice theory and mechanism design. (Actually, some of them have much harsher words... Doron and Kronick called results like these "perverse", and Kenneth Arrow's first definition of a voting system accidentally excluded RCV because he didn't realize this kind of scenario was a possibility.) –Sincerely, A Lime 01:09, 9 June 2024 (UTC)[reply]
I don't have the ability to review the edits to this controversial section thanks to disability, but *good lord* the way it's currently written tries to make these critiques sound like they're coming from some poindexters with purely theoretical gripes. Like, to the point where it gets in the way of reading ease and comprehension. I get that this is a contentious topic, because folks have emotional ties to their favorite voting system, but that part is currently trying to make it seem like imputant nerds were mad about technicalities.
The critiques of the system are straightforward and this election provided a clear and notable example of how RCV failed independence of irrelevant alternatives and failed to elect the Condorcet winner. This election is currently the go-to example of these failures due to recency and the notoriety of the election itself.
Seeing as how you guys are going back and forth at it, how about I give it a go and try to make those last three paragraphs as neutral as possible while still pointing out the interesting results from a criteria failure perspective. As a first demonstration, see my most recent edits. If I believe a change could start a significant argument, I'll come back to this thread and ask for opinions. I plan on mostly changing tone. I will explain each edit as I go in change summary and keep each edit small. Jasavina (talk) 05:56, 6 July 2024 (UTC)[reply]
Alright so I ended up making all the edits at once with explanations in the summary. I have just one thing to point out, though I don't have a strong opinion on it.
Yang may have said RCV elects moderates, but aside from being mathematically wrong, as explained in the next paragraph (center squeeze literally squeezes out centrists), it seems plain that the moderate candidate between Palin, Begich, and Peltola is not Peltola. I would advise removing his comment on account of being obviously incorrect. Jasavina (talk) 06:37, 6 July 2024 (UTC)[reply]
I think it's still notable enough to be included; politicians saying incorrect things is something we can report on, even if they're wrong, as long as we also provide commentary by experts. Closed Limelike Curves (talk) 17:29, 6 July 2024 (UTC)[reply]
@Closed Limelike Curves I think your recent edit to the second to last and third to last paragraphs go a bit further into the "RCV bad" tone that others can read into it, if they're interested in doing so. It might be helpful to somehow point out that even mathematical criticisms are somewhat of an opinion, given that no system can satisfy all criteria.
Also that second to last paragraph has two sentences back-to-back that say the same thing. I don't know which version you prefer, and they're tonally similar, so I'll leave it to you to choose. Jasavina (talk) 21:04, 6 July 2024 (UTC)[reply]
I think the term "moderate" is relative, and in this quote Yang could be referring to Peltola as more moderate than Palin, who probably would have won in the traditional plurality system with partisan primaries. Whether the comment is notable is debatable but I don't think its inaccurate. Also, we've gone over extensively how this article is not meant to be a thrashing of or even an in depth look into RCV and yet it keeps ending up that way. The lede should be concise and only summarize the main points. Stormy160 (talk) 18:03, 7 July 2024 (UTC)[reply]

who probably would have won in the traditional plurality system with partisan primaries.

I don't see how that's mathematically or logically possible. No matter what happens, Peltola wins the Democratic primary and then faces either Begich or Palin. (Probably Palin, but we can't mathematically prove Begich would've lost a closed primary, and chances are it would've favored Palin.) Which is the main defense for IRV here, which is that even if it didn't help, at least it didn't make things any worse.
That said, I'd somewhat agree with your argument about "moderate" being relative, given Peltola won the next election. Peltola is very moderate at the national level, but because this was a low-turnout special election in Alaska, she was quite far left compared to the median voter. Going off the ballots alone, you can show they're a single-peaked profile along a left-right spectrum, with the median voter positioned closest to Begich; but when a politician is speaking off the cuff, I don't expect them to understand or talk about the median voter theorem. Yang could argue two wrongs made a right here, I suppose. Closed Limelike Curves (talk) 22:20, 7 July 2024 (UTC)[reply]

Final results

[edit]

I think it is important to use the official final results as they appear twice in this article, certified by the state: 51.5% for Peltola and 49.5%. Your references to 48.4% and 44%, for Peltola and Palin, respectively, may confuse readers. Parkwells (talk) 18:52, 10 June 2024 (UTC)[reply]

Guys, it's really pathetic the hatchet job this is to the basic facts

[edit]

This election is an important edge case in the same manner as the 2009 Burlington mayoral election. It is an example of an RCV election, using IRV rules, that had an anomalous outcome due to the tallying rules of IRV. Now, certain things are just factual and supported in citation and should be made clear and not swept under then rug. I think I and Limey agree on that. But these factual things are getting twisted into what are actually falsehoods or into possible hypotheticals that didn't actually happen. Why all these weird contortions?

We should just be clear about what the facts are. 146.115.164.202 (talk) 07:55, 9 July 2024 (UTC)[reply]

By contrast, other activists and social choice theorists criticized the instant-runoff procedure for demonstrated pathological behavior.[1][2] Analysis of the ballots showed a simple majority of voters preferred Nick Begich to either of his opponents, despite the instant-runoff procedure eliminating him first.[3][4][5] However, Palin spoiled the election by splitting the first-round vote, leading to the elimination of Begich, the only candidate that could defeat Peltola.[5] Had Palin not run and the same voters supporting her had voted their same preferences with the remaining candidates, Begich would have defeated Peltola with a margin exceeding 8000 votes.
Begich's elimination was an example of the center squeeze effect, which disproportionately harms the candidate closest to the center of public opinion.[6][7][8] This is because the instant runoff voting does not consider second-choice rankings until a voter's first choice is eliminated. The centrist candidate tends to receive more second-choice votes from voters supporting either candidate on his left or right than the two candidates on the left or right can expect to receive from voters at the opposite extreme. This uncommon anomalous outcome (center squeeze and spoiler effect) of IRV was also previously demonstrated in the U.S. in the 2009 Burlington (Vermont) mayoral election. 146.115.164.202 (talk) 07:55, 9 July 2024 (UTC)[reply]
What are the contortions? And I agree that the “Begich would’ve won by 8K votes” claim is a hypothetical. Also I don’t know why all your new language got taken out that’s a question for Lime. Stormy160 (talk) 11:20, 9 July 2024 (UTC)[reply]
Lime is being a bit goofy: Peltola also would have lost the election if 8,000 more Palin supporters had ranked her first. ... WTF??? That is semantically not at all the same as Had Palin not run and the same voters supporting her had voted their same preferences with the remaining candidates, Begich would have defeated Peltola with a margin exceeding 8000 votes. The latter is simply a fact supported by the numbers from the cast vote record. What Lime made it into is not. It's a contortion. I am going to revert it back to where you and I were, Stormy, before Lime came back and really fouled it up. It we cannot make these simple statements stand, then I'm outa here. I have been involved in Wikipedia revert battles before and I don't have the stomach for it anymore. 146.115.164.202 (talk) 18:11, 9 July 2024 (UTC)[reply]
I hope something closely approximating the current version gets to stand. Because otherwise this article is going to become a real joke, a revert war might start, and the grownups will come and shut this down and freeze the article in a bad state. Lime, your mods change the meaning from simple factual statements supported by evidence and citation into something else that simply isn't factual. Leave it alone. 146.115.164.202 (talk) 18:38, 9 July 2024 (UTC)[reply]
I added a second unusual election and a citation for both just for completeness's sake, but otherwise I'm fine with it as-is. There's some copy-edits I'd make (replacing that cumbersome "and the same voters supporting her had voted their same preferences with the remaining candidates," with "-all else being equal-") but otherwise the notable aspects of the election seem to be covered fairly. Jasavina (talk) 19:02, 9 July 2024 (UTC)[reply]
@146.115.164.202 I think the reference to 8,000 votes was probably confusing. Sorry for not explaining myself very well!
Basically, say we start with the current results, but subtract 8,000 votes from Palin and add them to Peltola's vote total. The 8,000-vote drop in Palin's vote total pushes Begich into round 2 of the election, where he then beats Peltola by 500 votes.
Any number between 5,200 and 8,500 works because 8,500 is Begich's margin of victory over Peltola. I just picked 8k as a random example and didn't realize this could easily cause confusion with Begich's margin over Peltola. Sorry about that! Closed Limelike Curves (talk) 20:08, 10 July 2024 (UTC)[reply]
@Closed Limelike Curves @Stormy160 Hey Stormy, can you find a better source for that Yang line? The citation only says "Supporters outside of Alaska, like former presidential candidate Andrew Yang, laud the system for forging compromise." in reference to Yang's statement. Jasavina (talk) 23:46, 12 July 2024 (UTC)[reply]

References

  1. ^ Maskin, Eric; Foley, Edward B. (2022-11-01). "Opinion: Alaska's ranked-choice voting is flawed. But there's an easy fix". Washington Post. ISSN 0190-8286. Retrieved 2024-02-09.
  2. ^ Graham-Squire, Adam; McCune, David (2022-09-11). "A Mathematical Analysis of the 2022 Alaska Special Election for US House". p. 2. arXiv:2209.04764v3 [econ.GN]. Since Begich wins both … he is the Condorcet winner of the election … AK election also contains a Condorcet loser: Sarah Palin. … she is also a spoiler candidate
  3. ^ Atkinson, Nathan; Ganz, Scott C. (2022-10-30). "The flaw in ranked-choice voting: rewarding extremists". The Hill. Retrieved 2023-05-14. However, ranked-choice voting makes it more difficult to elect moderate candidates when the electorate is polarized. For example, in a three-person race, the moderate candidate may be preferred to each of the more extreme candidates by a majority of voters. However, voters with far-left and far-right views will rank the candidate in second place rather than in first place. Since ranked-choice voting counts only the number of first-choice votes (among the remaining candidates), the moderate candidate would be eliminated in the first round, leaving one of the extreme candidates to be declared the winner.
  4. ^ Clelland, Jeanne N. (2023-02-28). "Ranked Choice Voting And the Center Squeeze in the Alaska 2022 Special Election: How Might Other Voting Methods Compare?". p. 6. arXiv:2303.00108v1 [cs.CY].
  5. ^ a b Graham-Squire, Adam; McCune, David (2022-09-11). "A Mathematical Analysis of the 2022 Alaska Special Election for US House". p. 2. arXiv:2209.04764v3 [econ.GN]. Since Begich wins both … he is the Condorcet winner of the election … AK election also contains a Condorcet loser: Sarah Palin. … she is also a spoiler candidate
  6. ^ Graham-Squire, Adam; McCune, David (2022-09-11). "A Mathematical Analysis of the 2022 Alaska Special Election for US House". p. 2. arXiv:2209.04764v3 [econ.GN]. Since Begich wins both … he is the Condorcet winner of the election … AK election also contains a Condorcet loser: Sarah Palin. … she is also a spoiler candidate
  7. ^ Clelland, Jeanne N. (2023-02-28). "Ranked Choice Voting And the Center Squeeze in the Alaska 2022 Special Election: How Might Other Voting Methods Compare?". p. 6. arXiv:2303.00108v1 [cs.CY].
  8. ^ Atkinson, Nathan; Ganz, Scott C. (2022-10-30). "The flaw in ranked-choice voting: rewarding extremists". The Hill. Retrieved 2023-05-14. However, ranked-choice voting makes it more difficult to elect moderate candidates when the electorate is polarized. For example, in a three-person race, the moderate candidate may be preferred by a majority of voters to each of the more extreme candidates. However, voters with far-left and far-right views will rank the candidate in second place rather than in first place. Since ranked-choice voting counts only the number of first-choice votes (among the remaining candidates), the moderate candidate would be eliminated in the first round, leaving one of the extreme candidates to be declared the winner.

Deletion of pairwise comparisons matrix

[edit]

@CRGreathouse I've tried to find a phrasing that works for everyone, but at this point I'm sick and tired of this dragging on for months. @Stormy160 has consistently deleted or reverted any attempts to include a display of the pairwise-comparison matrix, calling them "made-up results" despite the matrix being well-sourced with links to several scholarly journals. At this point, I think it's clear that they don't object to the presentation of the paired comparison matrix, they object to the inclusion of a well-cited dataset provided by the state and widely used by social choice theorists, which happens to show a result they don't like (that 52% of voters supported Begich compared to only 48% for Peltola). Closed Limelike Curves (talk) 04:10, 18 July 2024 (UTC)[reply]

I need everyone here to cool off. Edit warring isn't productive.
@Stormy160: You said that this claim was false: Peltola received no support on a majority of ballots. Could you provide evidence? The claims is sourced, so the burden of evidence is on you. Please do not remove this before providing at least one WP:RS here supporting that.
The claims of Palin being a spoiler (I think that's a more common term in social choice literature than spoiler candidate) and a Condorcet loser are readily verifiable and quite standard. For now, they stay in.
CRGreathouse (t | c) 04:26, 18 July 2024 (UTC)[reply]
I am so sick and tired of this user over and over again trying to use this article to trash ranked choice voting. It’s obvious, other editors have pointed it out, and they keep trying to sneak things in. I have tried over and over to incorporate their concerns just for them to come back with more. And every time I try and have a productive conversation on the talk page I get opinionated responses about how RCV or IRV sucks.
As for the “majority against Peltola” claim, that’s not true if you include third choice rankings, and it’s misleading to say that in the first place when she got a majority in the final round. Stormy160 (talk) 11:28, 18 July 2024 (UTC)[reply]
This is why I think excluding exhausted ballots was a mistake, and is likely to substantially confuse users. Peltola didn't win a majority in the final round, she got a plurality. The number of ballots indicating no support for Peltola or Palin (truncated ballots) is enough to push Peltola below a majority.
(In a 3-candidate race, *every* candidate has unanimous support if you include 3rd preferences. 3rd preference is last here.)
@CRGreathouse I'm getting tired of the constant accusations of bias, making up fake election results, or "just saying RCV sucks". I've done my best to be polite through this conversation. What Stormy is calling "opinionated responses about how RCV sucks" are me citing sources on pathologies present in IRV (e.g. nonmonotonicity) that are well in line with the scientific consensus on this issue, and are directly relevant to this article. When there are accurate points in defense of RCV (namely, that it did no worse than partisan primaries here, albeit no better) I'm happy to bring them up and include them.
I'm trying to accurately summarize what happened in this election, not play some kind of partisan game where I delete or obfuscate information about IRV. Closed Limelike Curves (talk) 14:20, 18 July 2024 (UTC)[reply]
@CRGreathouse unfortunately, @Stormy160 has continued edit-warring (this is basically the pattern of editing that forced me to call you in). Closed Limelike Curves (talk) 16:32, 18 July 2024 (UTC)[reply]
I don't want to play babysitter here. I've protected the page for a week; please find a way to discuss this productively either between yourselves or with a relevant WikiProject. (FWIW, Stormy, I'm a fan of ranked-choice voting -- or anything better than FPTP in general -- but we're not here to push agendas, we're trying to write an encyclopedia.) - CRGreathouse (t | c) 18:28, 18 July 2024 (UTC)[reply]
I don't want to play babysitter here.
Yeah, sorry about that... I'm not really enjoying this myself.
The issue so far is that even verbatim quotes taken with strong citations from academic papers get reverted, no matter how I phrase or display the results. So far, if it involves any analysis of pairwise-comparisons, despite these being included in other articles like 2009 Burlington mayoral election for a long time now.
(FWIW, Stormy, I'm a fan of ranked-choice voting -- or anything better than FPTP in general -- but we're not here to push agendas, we're trying to write an encyclopedia.)
For that matter, I agree! I'm a big fan of RCV methods, and I think IRV is still an improvement on FPP. But Wikipedia has to reflect the sources, and the commentary from academics on this election was universally negative. Closed Limelike Curves (talk) 21:27, 18 July 2024 (UTC)[reply]
Really? Lime, you are the one who repeatedly has gotten rid of everyone else's edits to put your own language exactly back as it is. I've consistently tried to work within whatever the current version is. And I have consistently made arguments based on what is encyclopedic, only to be ignored. Stormy160 (talk) 21:40, 18 July 2024 (UTC)[reply]
Stormy, I'd like to see some discussion here about how you'd like the article to change. You had a colorable claim about "Peltola received no support on a majority of ballots." being incorrect or misleading; would you please expand on that? Certainly we don't want bad info in the article. Maybe give evidence of what you believe to be the case and some wordings that would be better.
Also, I'd like to see both of you act like the other person is trying to do a good job. You don't need to fight, you need to collaborate. I genuinely believe both of you are trying to improve the article, I just think you've mostly been going about it in an unproductive fashion.
CRGreathouse (t | c) 13:50, 25 July 2024 (UTC)[reply]
I'm sure they are, and I don't think I've made any accusations of bad faith throughout this ordeal. I'm sure Stormy is trying to keep out material he thinks, from his perspective, is biased. However, the repeated personal attacks (particularly in the form of rude edit summaries) and continuing to edit-war despite warnings to stop make me think he's gotten a bit too emotionally invested in this to contribute effectively to the article. Despite multiple RfCs and attempts at letting the issue go (I avoided any edits throughout most of June IIRC), I've gotten little-to-no response from the broader Wikipedia community, probably because electoral systems are a pretty niche topic.
In several cases, Stormy has reverted without first attempting to comment or discuss the problem. I left an RfC 10 days ago in #Inclusion of pairwise-comparisons matrix, got no objections and some support on including paired comparisons tables. I explained how the paired comparisons table is just another way of displaying the results (much like an election map or a bar chart), and does not imply or require the use of any particular electoral system (in the same way that election maps can be displayed in articles for elections not using some kind of electoral college or single-member districts). Despite this, Stormy once again reverted the inclusion of the matrix. Stormy's most recent edit summary referenced my last RfC, so it seems he's aware of it but prefers to ignore it. At this point I'm genuinely not sure what more I can actually do. Closed Limelike Curves (talk) 05:00, 29 July 2024 (UTC)[reply]
Stormy has gone back to trying to remove the information on majoritarian failure in this election (sourced to Graham-Squire and McCune). See here and here; second reversion was after being warned. They've refused to provide sources or discuss this on talk. This is despite having been clearly told:

You said that this claim was false: Peltola received no support on a majority of ballots. Could you provide evidence? The claims is sourced, so the burden of evidence is on you. Please do not remove this before providing at least one WP:RS here supporting that.

Not sure what to do at this point. – Closed Limelike Curves (talk) 19:12, 30 October 2024 (UTC)[reply]

Inclusion of pairwise-comparisons matrix

[edit]

Should we include a pairwise-comparisons table similar to the one at 2009 Burlington mayoral election and Draft:Center squeeze#Examples? Such a table would show, for each pair of candidates, how many voters preferred (ranked higher) one candidate or the other. Closed Limelike Curves (talk) 23:19, 18 July 2024 (UTC)[reply]

Support. The pairwise comparison table is a basic summary of the outcome that dramatically improves the clarity of the article and the interpretation of the results. I see no reason not to include it. Closed Limelike Curves (talk) 02:40, 22 July 2024 (UTC)[reply]
@Closed Limelike Curves Support. The pairwise match-ups are useful in the context of understanding why this election drew commentary and dissection from academic analysts. Without them, understanding the pathologies is more difficult. Jasavina (talk) 04:37, 22 July 2024 (UTC)[reply]
I'd like to see more commentary on this from outside opinions, how else can we bring people in? I still think these pathologies is what makes this election so interesting, but with me being essentially the only neutral party voicing an opinion in the talk section, the include/remove fight is always going to be emotional. Jasavina (talk) 04:29, 29 July 2024 (UTC)[reply]
@CRGreathouse So far I haven't gotten much in the way of responses to the RfC, so I'd like a resolution on this if possible. Closed Limelike Curves (talk) 14:54, 13 August 2024 (UTC)[reply]
Support this is the normal way to display this data. McYeee (talk) 02:08, 15 August 2024 (UTC)[reply]
Comment I have reverted an edit by @Stormy160: which would correspond to an oppose !vote. A summary of the argument against inclusion would be helpful here Stormy seems like the editor to provide it. McYeee (talk) 05:51, 26 August 2024 (UTC)[reply]

Remove line breaks

[edit]

Remove the line breaks in the quote for the reference "RCV Fools Palin Voters into Electing a Progressive Democrat". These are causing the page to be placed into Category:CS1 errors: invisible characters. Sammi Brie (she/her • t • c) 03:10, 22 July 2024 (UTC)[reply]

 Done * Pppery * it has begun... 18:18, 22 July 2024 (UTC)[reply]

Talk:Ranked-choice voting in the United States has an RfC

[edit]

Talk:Ranked-choice voting in the United States has an RfC for possible consensus. A discussion is taking place. If you would like to participate in the discussion, you are invited to add your comments on the discussion page. Thank you. – Closed Limelike Curves (talk) 18:25, 18 October 2024 (UTC)[reply]

Majority Winner Edit Squabble

[edit]

Yo, @Closed Limelike Curves and @Stormy160 see my recent edit and the summary. If you want to change that line or anything related to it, we all have to agree on the change. You two will go back and forth forever if we don't come to a consensus over here first, and it makes the edit history full of noise. Jasavina (talk) 19:32, 30 October 2024 (UTC)[reply]

More precise & correct phrasing of monotonicity failures

[edit]

@McYeee

for reference the quote in question:

""" The election was also a negative voting weight event, where a voter's ballot has the opposite of its intended effect (e.g. a candidate being disqualified for having "too many votes"). In this race, Begich lost as a result of 5,200 ballots ranking him ahead of Peltola; Peltola also would have lost if she had received more support from Palin voters. """

First off, just so we're all clear, I am aware that there is no disputing the objective facts that this election exhibited monotonicity failures both in participation failures and in preference orders. Those are unimpeachable truths and I am not trying to censor / whitewash that from this article.

But I take issue with the phrasing and find it not particularly precise.

  • "negative voting weight" this is not a well-defined term in this context. sure, you could point to some subset of Palin voters who could have defected to Begich and gotten a more-preferable outcome, or some subset of Begich bullet voters who (presumably) prefer Palin to Peltola and could have defected to get a more-preferable outcome, but this doesn't have much to do with "weight" and I find that to be a pretty meaningless term here.
  • No candidate was disqualified for having "too many votes." If Begich had gotten more "votes" (in this sense presumably using first-preferences?) he would have won the election. If Palin had gotten more she would have won the election. Yes IRV is not perfectly monotonic, but it remains true that as a general rule if you get more votes you're way more likely to win.
  • Begich did not "lose as a result of 5200 ballots ranking him ahead of Peltola." He lost because the IRV algorithm did not elect him. I'm sure it's factually true that there existed a subset of 5200 ballots ranking him ahead of Peltola, where if those preferences had reversed he would have won. But it is not appropriate to assign "blame" to these ballots --- like how should that subset of 5200 be selected out from all the thousands of ballots overall ranking Begich > Peltola ? surely if all of them defect then Peltola wins in a landslide

In general I think the wording here needs to be much more clinical and less editorial. Simply say "exhibited failures of monotonicity and participation." Mentioning that it would have been possible for 5200 Begich ballots to change the outcome is fine, but in my opinion to remain NPOV I think this needs to be paired with the mention that this coordination would likely have been extremely difficult to predict in advance. Affinepplan (talk) 22:47, 7 November 2024 (UTC)[reply]

Unless a source can be found, I agree that the phrase "negative voting weight" should be removed. I think the informal criticism wouldn't really make quite as much sense in the absence of a Condorcet Winner. Therefore, I would suggest the paragraph begin with a sentence ending "failures of the monotonicity, participation and Condorcet criteria." After that, I would suggest as much of the definition of each of those three criteria as is necessary for understanding what a single election demonstrating failure of all three of them is. I don't think the reader can be presumed to know the meaning of "monotonicity" or "Condorcet" in any context, or of "participation" in this context.
I don't think I can write clinically enough that you'll like what I'd write, so I'm not even going to try to. Instead I'll just suggest that you make another bold edit. Thank you for your patience — McYeee (talk) 07:35, 8 November 2024 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Talk:2022_Alaska%27s_at-large_congressional_district_special_election&oldid=1256110689"