Talk:2018 Victorian state election

RfC at Wikipedia talk:WikiProject Australian politics#RfC addressing the inclusion of minor parties in Australian election article infoboxes

A RfC has been started at Wikipedia talk:WikiProject Australian politics#RfC addressing the inclusion of minor parties in Australian election article infoboxes which may affect the infobox of this article. ColonialGrid (talk) 11:22, 3 March 2015 (UTC)

Orphaned references in Victorian state election, 2018

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 Victorian state election, 2018'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 "Morgan":

• From Western Australian state election, 2017: "Baird Government drops behind for first time in NSW; Barnett in trouble in Western Australia while Andrews Government still riding high in Victoria despite CFA union dispute". Roy Morgan Research. 10 October 2016.
• From South Australian state election, 2018: [1]

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⚡ 14:49, 10 October 2016 (UTC)

A Commons file used on this page has been nominated for deletion

The following Wikimedia Commons file used on this page has been nominated for deletion:

• Daniel Andrews.jpg

Participate in the deletion discussion at the nomination page. —Community Tech bot (talk) 13:23, 6 August 2018 (UTC)

Registered parties

There are currently 21 parties listed in this article. The ref, (Victorian Electoral Commission), last updated 14 August, lists only 20. The anomaly is Rise Up Australia Party. Can editor/s on VEC mailing list pls confirm Rise Up should be removed? Thanks, JennyOz (talk) 03:53, 23 August 2018 (UTC)

I have removed Rise Up, still not appearing on VEC registered parties list, JennyOz (talk) 08:09, 12 September 2018 (UTC)

Turnout

Why was the 2018 turnout so low compared to 2010 and 2014? I know that there were lots of pre- and early and absent polls. And it may have been a cold and rainy day. --Bronwyn Gannan (talk) 02:28, 26 November 2018 (UTC)

At this stage, the number in the Turnout box means the count progress, it will keep going up until the count is finalised. --Canley (talk) 06:10, 26 November 2018 (UTC)

Thank you, Canley! That does clarify things. It's not been usual for me to check the state election articles so early - at least until the counting is done. --Bronwyn Gannan (talk) 06:54, 26 November 2018 (UTC).

Pre-emptive calls

We need to be extremely careful about calling races where there are elimination order issues due to it being unclear which candidates will finish in the final two. If the VEC didn't choose the correct top two candidates on election night (and in several races we don't even know who the correct candidates are now), it isn't based on actual votes (because there is no 2CP count) but on the ABC computer's estimates.

Prahran is another example of this - the ABC computer called it multiple times on election night but any human commentator was well aware that, like last election, no one would have the faintest idea who the winner was until it was finalised who had finished third.

Melton has also been pre-emptively called because of the same error, and there is another five seats where one might possibly call it with the understanding that it's still possible for independents to win on current figures.

We should absolutely not be making calls here in these races where actual human commentators say it's still up in the air purely because the ABC computer has called it on an estimate. This is why we've always relied on human experts, not the ABC computer, to determine which races are still in doubt at every election ever on Wikipedia. The Drover's Wife (talk) 04:50, 26 November 2018 (UTC)

Strongly agree. Benambra and Melton, at the very least, are very clearly in doubt. Probably we should also be listing Pascoe Vale and South-West Coast as well. We have learned from long experience that we should not treat seats as called while serious commentators (which for us means basically William Bowe and Kevin Bonham, as the only people who do serious post-count analysis) are treating them as doubtful. Frickeg (talk) 05:27, 26 November 2018 (UTC)

Now that you bring it up, I think it would be a good idea to list Pascoe Vale and South-West Coast given Bonham's commentary on those races. The Drover's Wife (talk) 06:12, 26 November 2018 (UTC)

I agree as well (about Pascoe Vale and South-West Coast). --Canley (talk) 06:16, 26 November 2018 (UTC)

Hi The Drover's Wife, and others. I should have been more detailed in my explanation regarding Benambra. It has indeed been called by the ABC and Antony Green. It's important to note that this is not simply a computer automated process. Seats which have unnecessary VEC preference counts are being overridden and placed in the "seats in doubt" category, such as Prahran. The ABC computer would treat Prahran as a clear Greens retain, but this has been overridden manually. This is not the case for Benambra, which has not been overridden, so it's not simply a case of an automated process marking something as retain when it shouldn't be. Apologies if this has caused any confusion.

Regarding overall, we could rely on Bonham and Bowe as reliable sources, but they have both indicated that their updates are going to be irregular if at all, where the ABC is more instantaneous. Onetwothreeip (talk) 06:38, 26 November 2018 (UTC)

EDIT: In the last half an hour or so, Benambra has been overridden and placed into doubt by the ABC now. Onetwothreeip (talk) 06:41, 26 November 2018 (UTC)

However, this doesn't explain reverting my edits which had nothing to do with Benambra. Onetwothreeip (talk) 06:42, 26 November 2018 (UTC)

You're still not understanding the problem. We couldn't call Benambra on the computer projection regardless because it wasn't based on any actual votes: there had been no 2CP count between Tilley and Hawkins. The same applies in Melton, Pascoe Vale and South-West Coast. (It also applies in Geelong, Werribee and Ovens Valley, except that the primary votes are such in those three seats that they can more or less be called in the absence of a 2CP count because an independent comeback would be incredibly unlikely). Whether (and/or when) the ABC overrides the computer projection or not in those seats goes to the accuracy of their own website, but it has zero bearing over whether those seats are factually in doubt when analysts are telling you from the primary votes that the seat is in doubt.

Intentionally adding seats as called when we know for a fact that they are still in doubt because an incorrect computer estimate is quicker than a human call is deeply unhelpful, and Bonham (at least) usually updates his postcount pages multiple times a day. There is just no reason to do it - and good reason not to because, as you just illustrated for yourself, it definitively results in incorrect pre-emptive calls. The Drover's Wife (talk) 07:03, 26 November 2018 (UTC)

The projection was made based on actual votes, the votes of first preference. It was not an incorrect computer estimate, the projection was changed from the actual 2PP count to an estimate from first preferences, and the result of "retain" was not overridden. I would not have made such a decision if it was only a computer process making the call. Using the results from Bowe or Bonham is a separate issue, they are irregular by their own admission, but they are undoubtedly reliable sources. I did not make any edits from any of their absence. Onetwothreeip (talk) 07:15, 26 November 2018 (UTC)

First rule of election postcounts: the ABC computer is an unruly beast, and may react unpredictably to small changes. We have no actual evidence that the Benambra call was made manually, and I rather doubt it was. We have always found the ABC computer unreliable with non-standard contests, where at this stage the margins are entirely educated guesses. There's also the fact that even when Antony Green does make manual changes, he rarely explains them, unlike Bonham and Bowe. No seat considered in doubt should be "called" by us without firm evidence from a reliable source (preferably more than one - I have usually gone with two out of three) that that doubt is no longer reasonable. By "irregular", Bonham and Bowe mean "maybe not more than once a day" - and if it's a matter of waiting an extra day to avoid a wrong call, that's a worthwhile thing to do. (Note that Bowe is actually including Mildura and Geelong as in some level of doubt.) Frickeg (talk) 07:39, 26 November 2018 (UTC)

If we have to interpret when the ABC calculations are right and wrong, then we really shouldn't be using them. At the time, Benambra wasn't manually overruled while other seats were, and was listed neither as in doubt or as a changing seat by the ABC. I did not make the call about Benambra on the article, I only removed it from the in doubt table. I agree that we ought to wait if the position is unclear. Onetwothreeip (talk) 08:56, 26 November 2018 (UTC)

I think you're right about the first part - at least as far as non-traditional contests go. The same problem doesn't exist for traditional Labor-Coalition seats, but it's still better to be informed by expert humans as well. The Drover's Wife (talk) 09:47, 26 November 2018 (UTC)

I think most of us are now aware that the same problem does indeed exist for Labor-Coalition contests, whether it's uncertain if they will be a Labor-Coalition contest, or something that happens late in the count like Caulfield. We have Kevin Bonham and The Poll Bludger, but they themselves caution us about irregularity and unreliability. I honestly can't think of a single part of the ABC results that are completely reliable, and this would surely have big implications for Wikipedia reporting on results before the VEC full distribution of preferences. Onetwothreeip (talk) 22:41, 3 December 2018 (UTC)

I don't think we're at that point yet: immediately or very soon after an election is when there's the most interest, and there's so many pages that need updating in the aftermath of an election that it's a really useful chance to just get it all done when people are noticing. There is the occasional case these days (like Wentworth and now Caulfield) where drastically different things happen with different types of votes in contrast to previous elections, with the potential to upset what had earlier looked like clear calls, but basically all of the others this election were quite clearly still in doubt to anyone who was reading human expert commentary: the only other potentially in doubt seat that wasn't immediately on the radar was Mildura, which briefly popped in and back out of doubt again this week without the lead changing. The main lesson in this is that the ABC computer projections are not god, which is pretty obvious once you've followed a postcount process anywhere, and that humans will always be necessary to follow close or complicated postcounts - which is why we've always called races on that basis. The Drover's Wife (talk) 02:23, 4 December 2018 (UTC)

The problem is calling seats as in doubt without evidence that they are in doubt, but rather the lack of evidence that they have been sufficiently called a certain way. Bowe and Bonham are both careful not to use words like called so it seems like we either trust the ABC results with all their flaws or we don't have a clear way of determining if a seat is in doubt or not. Should we start with every seat in doubt, for example? Onetwothreeip (talk) 02:35, 4 December 2018 (UTC)

We've never "trusted" the ABC projections because it's an obviously unreliable way of dealing with races that are close and/or difficult in post-election night counting - if we did use it that way, it would guarantee incorrect calls each and every election. No one uses them in that way - not even Antony Green uses them unequivocally without also using his own knowledge and experience, especially not in postcounts. The cautious approach we've learned over the years is to treat a seat as in doubt if any one of the main experts (Green, Bowe, Bonham, potentially a couple of others who don't do every election) views a seat as still in doubt, and that works just fine in the overwhelming majority of cases. This particular question has never arisen before because we've never had an editor insistently try to rely on the ABC projections over expert opinion before. The Drover's Wife (talk) 04:02, 4 December 2018 (UTC)

I don't think we will get editors who will insist we rely only on the ABC but it would go without saying that I agree it would be catastrophic if we relied literally on the ABC results, and that is a point I'm making. There have been of concern several seats that we've considered to have changed hands that were then moved to be doubtful, and those that weren't initially considered in doubt but then became so. Otherwise I'm not sure what question you're referring to. Onetwothreeip (talk) 04:48, 4 December 2018 (UTC)

I think we're all in agreement here, folks, aren't we? We can only go by the sources, so if they get it wrong, so do we - for a while, until someone catches it. I think, in all the years we've been doing this, we have only had two instances where we indicated a seat had been won and then the other candidate won instead (Frome 2009, Lismore 2015). Frickeg (talk) 05:03, 4 December 2018 (UTC)

Name Change to 2018 Victorian state election

Hey everyone.

Was just pinging across a message concerning the name of this article. As far as I know, its the only one related to Australian elections thats titled "2018 Victorian state election" whereas the 2014 article is still "Victorian state election, 2014" as well as the Federal Election pages. Is this going to be the norm across all electoral articles or is this page an anomaly?

Cheers, DestinationAlan (talk) 05:23, 29 November 2018 (UTC)

There was a Request for Comment (RfC) recently which proposed altering the naming convention for election articles to put the year first. The RfC was closed as supported by the community, and the naming convention was changed. There is a bot (User:TheSandBot) currently renaming election articles to the new format (and fixing up links and redirects), and some editors have manually renamed articles like this one. The bot will get there eventually, but you are welcome to do a manual page move if you see one not done. --Canley (talk) 06:00, 29 November 2018 (UTC)

Thanks for the clarification Canley. I'll be sure to help where I can. DestinationAlan (talk) 06:11, 29 November 2018 (UTC)

Results tables output

Results tables test output from the VEC media feed can be found in my user space at User:Canley/Results of the Victorian state election, 2018 (Legislative Assembly). If you spot any errors, correct them on the page (or leave a note) and I'll pick them up and make any fixes/changes needed. Don't worry about updating the vote figures or the win status, I'll refresh the feed each evening. --Canley (talk) 01:03, 2 December 2018 (UTC)

Looks very nice, Keep doing a good job and eventually adding them in anytime if you want. Chad The Goatman (talk) 20:16, 4 December 2018 (UTC)

I think technically Greens does only regain but new seat from different district, regardless by the result from by-election was subverted in State election

Because their someone objected the idea that thinking new seat gain for Brunswick district for Victorian Green Party in table list it doesn't matter because last year of by-election happened in Northcote district with Green gain it despite however lose district in state election with formerly project based on election day results of their party of keep two seats for days after official election day was over with still counting working on. Which until Brunswick district was counted for Green gain, days after the election was over days ago, So technically it would make sense they should still be qualified of a gain seat anyway in the rally table list. Chad The Goatman (talk) 20:09, 4 December 2018 (UTC)

There was no seat gain from the Greens' position pre-election, which is the way these tables have always been calculated. The Drover's Wife (talk) 20:15, 4 December 2018 (UTC)

Is it the way they've always been calculated (from pre-election status not previous election) though? Looking at some "recent" examples: in Victoria, the Coalition lost Benalla and Burwood in by-elections in 1999–2000, but the summary table and infobox for Victorian state election, 2002 say 42 + 20 = 62. In New South Wales, the Liberals lost three seats in by-elections in 2013–14, but the table and infobox in New South Wales state election, 2015 say 69 − 15 = 54. Federally, last by-election loss before Wentworth was Lyne to Rob Oakeshott in 2008, but the table and infobox for Australian federal election, 2010 say 65 + 7 = 72. All these seem to show the change from the previous election, and don't incorporate by-elections or change in party status. --Canley (talk) 21:09, 4 December 2018 (UTC)

I couldn't remember, but was assuming Frickeg was right when he said as much here. No objections if you want to change it. The Drover's Wife (talk) 04:07, 5 December 2018 (UTC)

I admit I was assuming too - perhaps because that seems like the logical course of action to me. There are certainly other elections where by-elections are taken into account - like Queensland state election, 1998 where Mundingburra is included. In the election tables we record Northcote, for example, as a "Labor gain from Greens", as opposed to MP defections like Morwell which would be an "Independent gain from National". This does get tricky when redistributions come into play, of course. Frickeg (talk) 05:36, 5 December 2018 (UTC)

Well prior to my most recent edit, the box showed an 8 seat gain for ALP (which is based on last election figures not pre-this election), as opposed to Coalition and Greens which were pre-this election figures. So I've reverted it to the last election calculation. But with examples to draw on using either approach, it would be nice if we could have a standard and agree to it for all election articles. FWIW I prefer the pre-election figures (ie: Greens had 3 going into the election and won 3 so they're steady). Global-Cityzen (talk) 05:15, 11 December 2018 (UTC)

I agree - with the proviso that this applies only to by-elections and not to defections (i.e. Melton was a Labor hold, Morwell an Independent gain). Frickeg (talk) 05:17, 11 December 2018 (UTC)

On the election results figures the Greens have made one gain since that's from 2014, but changing hands Labor has made a gain there. Onetwothreeip (talk) 05:31, 11 December 2018 (UTC)

Respectfully I disagree @Frickeg:. A party's share of seats in any legislature drops the moment a member is no longer part of that party. That happens when a by-election is called and the seat is vacant (and an alternate party wins that seat) or when a party MP defects to become an independent or join another party. I don't know why we'd treat by-elections and defections given the end result is the same; the party in question loses its share of seats. Global-Cityzen (talk) 06:22, 11 December 2018 (UTC)

Because in a defection, the party never wins the seat, whereas in a by-election they do. The last time Melton voted before this election, they voted for a Labor member; they did so this time too, so there was no change. The last time Morwell voted, it was for a National, but this time they voted for an independent, so there was a change. But the last time Northcote voted, that was for a Green; this time they voted Labor, so that was a change. It's all about whether we're talking about parties or individuals, and this table is clearly talking about parties. Russell Northe may have held Morwell, but the Nationals lost it. (Obviously the other changes need to be noted; I would suggest this be through a supplementary table under the seats changing hands column, somewhat like what we have done in the past with redistribution/notional holds/gains.) Frickeg (talk) 06:26, 11 December 2018 (UTC)

I feel like that's a fairly sound logic, although I don't have any strong opinions about this stuff. It would be nice to standardise this so we've got all articles on the same page. The Drover's Wife (talk) 06:44, 11 December 2018 (UTC)

I don't feel too strongly about it either, but agree it should be consistent. My feeling is the results summary table should be comparing to the last general election both in swings and seat change—I know a party can increase their vote percentage but still lose seats, but to throw by-elections into the presentation here adds another complication where the vote swing and seat swing are from different points in time (explaining it with notes or another table is also a good idea though). The infobox I think could show the change from the pre-election status because there is a field for both numbers, so it's pretty clear which one the change applies to and the full context is apparent without needing a note. --Canley (talk) 09:20, 11 December 2018 (UTC)

Final updates

South-West Coast is done, with Britnell re-elected and the independent Purcell not getting into second. Can someone update the tables? Only Melton still to go now. The Drover's Wife (talk) 04:07, 5 December 2018 (UTC)

All sections collapse under “Results” in mobile

Twiddling with the space in the section name eg == text == to ==text== doesn’t do anything.

However, clicking on the Legislative Council seats in the visual editor on mobile selects everything below it, as it f it were the same thing? Iamthinking2202 (talk) 07:07, 20 December 2018 (UTC)

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

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

• Matthew Guy.jpg

You can see the reason for deletion at the file description page linked above. —Community Tech bot (talk) 12:53, 7 September 2021 (UTC)