Talk:2018 London local elections

Please comment on a RfC at Talk:Jeremy Corbyn#RfC on infobox image

Please share your input in an RfC relating to what image should be used in the infobox at the Wikipedia page for the Leader of the British Labour Party (and in election infoboxes, such as the one at this page) AusLondonder (talk) 09:28, 19 May 2016 (UTC)

Lucas and Bartley image

Anybody know how we might be able to crop that Lucas/Bartley image? It really doesn't fit very well in the InfoBox but I'm not sure of the process of cropping such pictures. FriendlyDataNerdV2 (talk) 13:52, 12 April 2017 (UTC)

Infobox: parties to include?

We haven't had an infobox discussion lately, so let's start one! Woo! This article was created with an infobox containing Con, Lab and LDem. It stayed like that for about 18 months until FriendlyDataNerd added UKIP and the Greens to the box. It then stayed like that for about 6 months until FriendlyDataNerd removed them again. Three months later, 86.130.188.95 put them back in arguing "added ukip to table as in the last election they kept their deposit and all parties that keep deposits in any election should be included in the infobox" and something similar for a second edit re-adding the Greens. FDN removed, arguing, "Local elections don't have deposits. The popular vote wasn't relevant in these elections - only winning seats. 12 and 4 seats not notable". The reduced infobox stayed until a few days ago when Helper201 re-added them, I then removed them (saying, "both have trivial Councillor numbers & control no councils, so do they merit infobox inclusion?"), and Helper201 has now re-added ("Yes, they do. Each of the two parties secured approximately 10% of the vote in the last London local elections (over 200,000 votes each) and both are major parties.)").

We've obviously been here before and I appreciate there is no clear answer. As the infobox says, the results last time were as below. There are also 4 directly-elected mayors: results for which are not currently noted in the infobox.

• Lab: 1,060 councillors, 20 councils, 3 mayors
• Con: 612 councillors, 9 councils, 0 mayors
• LDem: 116 councillors, 1 council, 0 mayors
• UKIP: 12 councillors, 0 councils, 0 mayors
• Green: 4 councillors, 0 councils, 0 mayors

Looking at that, I think the obvious answer is a 3-party infobox. The UKIP and Green numbers are trivial in comparison. Indeed, Tower Hamlets First did better than UKIP and Green put together (winning a mayor and 18 councillors, although some of those results were overturned). Residents Associations won 27 councillors. So why have an infobox with UKIP and Green and without two groups who did better? And why include any of them when all put together (UKIP+Green+TH1+independent+Residents) is way below the LibDem figure?

However, if you look at vote share, as per Helper201, the picture does look very different:

• Lab: 37.4%
• Con: 26.1%
• LDem: 10.2%
• UKIP: 10.0%
• Green: 9.9%
• Residents: 1.1%
• Tower Hamlets First: 1.1%

On vote share, a five-party infobox seems obvious.

My feeling, as I think FDN's, and as I think is consensus from prior infobox articles for other UK election articles, is that UK elections, under FPTP, are not about vote share. They are about winning seats, and thus also winning control, and it's on that basis that we should be judging inclusion. The infobox should reflect the results last time in terms of seats/councils won. Vote share doesn't matter. The Greens got way more votes than the DUP or Sinn Fein at the last general election, but the latter get into the infobox for United Kingdom general election, 2017 and the Greens don't. Likewise, here, 9.9% was a great result for the party, but 4 councillors isn't and it's the latter that matters.

I also note that London local elections, 2014 only has Con/Lab/LDem in its infobox, so if the infobox for 2018 is, for now, based on the 2014 results, then clearly we should be following that. However, what does everyone else feel? Bondegezou (talk) 13:34, 29 January 2018 (UTC)

Thank you for opening discussion on this. I think its worth noting that in the last few UK local election pages all have contained at least one of the smaller parties (usually Plaid or UKIP) in the info box. As far as I'm aware, Tower Hamlets First had not took part in any elections prior to 2014, as that may be why they weren't included. I also think only parties that are standing in multiple areas should be included in the info box, not parties that are specific to only one local area. In regards to your point about seats, I think these party's should still be included as they are major national parties, they gained significant percentages in the prior election (and currently retain their gained seats), along with the fact the Green's were included in the info box for the London mayoral election, 2016, despite falling far behind the two main parties. Despite what I said prior I could see reason why someone may wish to have THF in the info box if they retained their amount of councilors, as it was quite significant, but with the party now defunct the point is moot. As for residents associations I wouldn't regard them as qualifying as they aren't a party but a plural. Helper201 (talk) 14:15, 29 January 2018 (UTC)

For London mayoral election, 2016, the decision -- after lengthy debate -- was to go with everyone who'd saved their deposit, i.e. over 5% of the vote. That's what we do for Westminster by-elections too. That doesn't directly translate to elections covering multiple seats and a simple 5% vote share cut-off has never been used on UK general or local election articles' infoboxes AFAIK.

I am generally in favour of inclusive infoboxes. However, the view to date has tended to be that seats won matters for these elections and the Greens' 4 councillors is, as I said, trivial. They don't win seats under FPTP. I'm all for dumping FPTP personally, but it is the system currently used.

United Kingdom local elections, 2018 does include UKIP and the Greens in their infobox based on prior results 4 years before, I presume. I wondered about editing that, but didn't. The seat numbers are more even there (LD 409 + 1 mayor, UKIP 154, Green 33). United Kingdom local elections, 2017 has SNP (431 councillors, 0 councils) and PC (202 councillors, 1 council), but not UKIP (1 councillor) or Greens (40 between E&W and Scottish parties). If the Greens don't get in with 40 councillors (GB-wide), why should they with 4 in London? United Kingdom local elections, 2016 also excludes the Greens (45 councillors), but does include UKIP (58 councillors and about double the Green vote share). United Kingdom local elections, 2015 again includes UKIP (202 councillors, 1 council), but excludes Greens (87 councillors).

The prior London local election articles all just have Con, Lab and LDem (or predecessors), back to 1971 which just has Con and Lab (Liberals got 9 councillors that year). Liberals are included for the 1974 article with 27 councillors. 1968 again excludes Liberals (who got 10 seats). 1964 again excludes (13 seats). Those numbers support excluding UKIP and Greens for now. Bondegezou (talk) 15:11, 29 January 2018 (UTC)

London local elections, 1906 does not have an infobox. Bondegezou (talk) 15:14, 29 January 2018 (UTC)

Empty results sections

Can we remove the empty results tables from the article until there's something to put in them? We could dump them here in Talk for now. Bondegezou (talk) 18:46, 22 April 2018 (UTC)

Flag

I have removed the infobox flag. This was the flag of the City of London, the one part of London that doesn't have elections at this time. It is not the flag of all London. Bondegezou (talk) 22:38, 27 April 2018 (UTC)

Being stored...

The following is moved from the main article and being stored here ready for when the results actually come in... Bondegezou (talk) 16:17, 2 May 2018 (UTC)

Results

Party		Councillors		Councils		Votes
		Seats	Change	Councils	Change	Votes won	% Votes	Change
	Labour
	Conservative
	Liberal Democrats
	Green
	UKIP
	Independent
	TUSC
	BNP
	CPA
	Socialist (GB)
	Advance Together
	Aspire
	Democrats and Veterans
	For Britain
	PATH
	Renew
	Women's Equality
	Others
	No overall control	—				—

The London Borough Councils are elected using block voting, meaning that voters could cast as many votes as there were seats. As a result, the popular vote totals reflect the total number of votes cast for that party, and do not indicate the level of turnout.

Political composition

London local elections 2018 Councillor statistics
Party		Seats	Gain/loss
	Green
	Conservative
	Labour
	Liberal Democrats
	UKIP
	Residents' associations
	Independent

Which vote totals to use

So one issue that we will soon have to decide is what vote count we will use to represent 'popular vote'.

Because London councils are elected using the bloc vote (with voters casting as many votes as there are seats) there are several different methods for determining 'popular vote' from the total numbers that are used by different organisations:

• Rallings and Thrasher use the 'top vote' method - they only count the result for the most successful candidate of that party within that ward. So if the result is Con 1,239, Con 1,093, Lab 1091, Con 998, Lab 901 and Lab 877, they consider the result to be Con 1,239, Lab 1,091 for the calculation. I believe Britain Elects uses this count method too.

• Votes standardised by number of candidates. The London Datastore uses this method, which they describe as follows:

"The correction chosen is the use of average votes per candidate for any givenparty or group. Percentages are based on the sum of average votes for a ward or borough, which is thus used as an estimate of the total number of valid votes cast in a single seat election."

• Total votes won. This is the simplest calculation, essentially taking all votes cast for the parties without making adjustment.

As you may have seen, I have standardized the current London election articles so that they all show the total votes won. I have done this for 3 main reasons:

• 1. It is the only vote count method that shows comparable results going right back to 1968 (1964 is unavailable by any method).
• 2. It is the only vote count method that provides comparable results over the years for minor parties - particularly relevant in 2006 when dozens and dozens of minor party councillors were elected.
• 3. It is the only vote count method that has not been altered or changed over the decades, and so provides consistent and comparable results.

All of the total votes won results are sourced from the London Datastore.

The question is, which count do we use? Bear in mind that someone will have to go back and change all of the previous articles if we choose a different count method this time.

I would make the case that we continue to use total votes won, as it is a factual statement (without 'corrections' or changes) of how votes were cast. I would also advocate it for the reasons above - that is the only method I have found that has comparable results across the decades, whilst including minor parties and thus giving *all* results. FriendlyDataNerdV2 (talk) 18:38, 2 May 2018 (UTC)

Update: I've read some arguments from Britain Elects that have led me to conclude that the 'top vote' method might well be the best way of presenting the data. We do already use it for Welsh elections and it would be good to have consistency. Reliable sources (Andrew Teale, Britain Elects, Rallings and Thrasher) all use the 'top vote' method, after all, and we should follow what reliable sources do. FriendlyDataNerdV2 (talk) 18:27, 15 June 2018 (UTC)

The 'top vote' method is the one I've always seen in reliable sources. I'd argue it better represents results for minor parties — if a major party stands three candidates in a ward and a minor party stands one, then the total vote suggests three times as many people voted for the major party compared to the minor party as they actually did. Getting consistent aggregate values sounds like a pain, though. Ralbegen (talk) 20:58, 15 June 2018 (UTC)

Aye - as I've said above I initially warm towards simply aggregating them, but after looking through the Rallings and Thrasher data and thinking about it more, I think you're right. Moving the articles over to the top vote method shouldn't take too long - Rallings and Thrasher have compiled them for the whole of London going back decades and Britain Elects has them for 2018. FriendlyDataNerdV2 (talk) 23:39, 15 June 2018 (UTC)

• Oppose: Whatever the benefits in terms of small party representation, the top vote method is notoriously difficult to calculate for historic election results, and thus compare results over time. Very disappointed I have only seen this now! I would strongly advise sticking to aggregate vote method. My pet project for the last few days has been to go back and compile historic results of Merton local elections (the entries were shockingly scarce); moving to top vote method would lose the comparability across election years. At the very best, it is better, then, to keep aggregate vote method for results tables and merely explain in a note the difference. MB190417 (talk) 09:26, 23 June 2018 (UTC)

I would have nothing in principle against using the top vote method for London-wide election results, for example, where they are available and, as aforementioned, go back years. I would be concerned in applying it to borough election results, where they may be more inconsistent/only retrievable for this election. Where differences arise, these can be explained in a table note. MB190417 (talk) 09:30, 23 June 2018 (UTC)

Hi User:MB190417 - actually they are readily available back to 1973 for all London boroughs, courtesy of Rallings and Thrasher. They're all here: http://www.electionscentre.co.uk/?page_id=3755 It is the aggregate results that in my experience are often absent. Andrew Teale's website also has them all back until 2002, e.g. http://www.andrewteale.me.uk/leap/results/2014/1/ FriendlyDataNerdV2 (talk) 12:36, 26 June 2018 (UTC)

FriendlyDataNerdV2 Are there any differences in the two sets? MB190417 (talk) 12:39, 26 June 2018 (UTC)

User:MB190417 Not from what I can see - Teale and Rallings/Thrasher appear to use the same method. Barking & Dagenham is a good example - in 2014 Rallings/Thrasher have it as 28,871 votes (52%) for Lab, Teale has it as 28,871 votes (52%) for Lab as well. FriendlyDataNerdV2 (talk) 12:40, 26 June 2018 (UTC)

In that case, I'd be open to using them. My sincerest apologies; I didn't realise there was such an extensive record! MB190417 (talk) 12:41, 26 June 2018 (UTC)

No worries at all, neither was I until I stumbled across them a few months ago. I totally understand your point of view above, indeed I shared it for a while, but I think the benefits of top vote method outweight potential downsides. FriendlyDataNerdV2 (talk) 12:42, 26 June 2018 (UTC)

Conservative Net Loss/Gain (Barnet)

Hi there,

I was wondering whether we should say that the Conservatives made a net loss of 2 rather than 1 as Barnet was won by the Conservatives in 2014 and on pages for general elections we don't include the change from a by-election to a general election but from the previous general election to that one, perhaps we should make the change from 2014 figures rather than from when it went into No Overall Control earlier this year.

86.158.223.89 (talk) 11:57, 5 May 2018 (UTC)

Agreed. Barnet counts as a Con hold. Bondegezou (talk) 19:55, 7 May 2018 (UTC)

Swing

The article infobox has a line saying "swing", but these are straightforward vote share changes. In a UK context, swing generally means a different calculation, as per Swing (United Kingdom). This could be confusing. Can we change how the line is labelled? Bondegezou (talk) 19:55, 7 May 2018 (UTC)

That's a fair point, but 'swing' is widely used on other UK election pages e.g. United Kingdom general election, 2017. It's describing simply vote share changes there as well, surely? FriendlyDataNerdV2 (talk) 16:34, 9 May 2018 (UTC)

It is... but then in results boxes for individual constituencies, it's generally showing Butlerian swing. This seems like a bad situation to me. Bondegezou (talk) 16:16, 18 May 2018 (UTC)

The BBC here refer to 'vote share change' rather than single-party swing, as they do for individual constituencies. Parliament's research briefing uses 'change' though less explicitly; ditto the FT. On an admittedly brief look, I can't find any sources that refer to single party swing as 'swing' when talking about the 2017 general election (as a proxy, because there's a lot more coverage of that than the 2018 locals!). Ralbegen (talk) 20:26, 18 May 2018 (UTC)

Proposed Removal of Aspire, Women's Equality, Duma Polska, Christian People's. TUSC, Plumstead, Renew, Advance Together and BNP parties from results table

Shouldn't these parties be removed from the table and factored into the other section as they received less than 1% of the vote and resulted in no councils/councillors like Ukip? 86.158.223.89 (talk) 16:03, 18 May 2018 (UTC)

I don't see the problem with a full results table. It's not huge. It's comprehensive. Bondegezou (talk) 16:14, 18 May 2018 (UTC)

Agreeing with Bondegezou. Full results table is comprehensive. MB190417 (talk) 09:31, 23 June 2018 (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:

• Theresa May Official.jpg

Participate in the deletion discussion at the nomination page. —Community Tech bot (talk) 19:11, 14 December 2019 (UTC)