Talk:Opinion polling for the 2022 French presidential election

Requested move 10 May 2021

 * The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion. 

The result of the move request was: moved. (closed by non-admin page mover) Vaticidalprophet 01:18, 18 May 2021 (UTC)

Opinion polling for the next French presidential election → Opinion polling for the 2022 French presidential election – French presidential elections are dates set in stone like the US presidential elections as per the constitution, they are not affected by the same phenomena as Opinion polling for the next United Kingdom general election, which could be brought forward. Looking at the page history, it looks like this page grew out of one redirect, and the desired title is at another redirect Unknown Temptation (talk) 21:15, 10 May 2021 (UTC)
 * Support per nom. The only reason the election would be a different date is if the president dies or resigns. Not likely. Mottezen (talk) 03:17, 11 May 2021 (UTC)
 * Support per nominator. Recurring events such as elections should not be titled with "next" but instead with the actual date. J I P  &#124; Talk 12:16, 11 May 2021 (UTC)
 * Support per nom Chessrat  ( talk, contributions ) 16:08, 17 May 2021 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Recent revisions
User:Le Lis et le Lion Can you explain why:


 * you keep removing the Redfield & Wilton Strategie poll from the graph.
 * you label Bertrand as "independent", when in France, this designation is rarely used. The official labels of Divers Droite (DVD), Divers Gauche (DVG) and Divers Centre (DVC) are usually used to further define these candidates, and Bertrand is considered a right-wing candidate. Most media organisations consider him as part of the Miscellaneous Right.

I reverted your latter change because you did not have the consensus needed to change Bertrand's label in the first place. Mottezen (talk) 18:41, 26 May 2021 (UTC)

I also reverted the removal of the poll from the graph, which you also never had consensus to remove. Please expand on your reason for the removal, because "obsolete" might as well apply to every polls except the most recent ones. Mottezen (talk) 22:29, 26 May 2021 (UTC)
 * Good evening,
 * I have my reasons for not wanting to put the R&WS poll : on Wikipedia FR, the poll has not been accepted. Le Lis et le Lion (talk) 15:22, 7 June 2021 (UTC)
 * Frwiki is friki, but this is enwiki. Please spell out the reason for removing it. Mottezen (talk) 19:07, 7 June 2021 (UTC)
 * As said here, this institute is not mentionned by a secondary source (of quality). The only source we have for this poll is the poll institute himself. This is a primary source : not enough. When a secondary source mentions that poll, then, it can be reintegrated in the page. Le Lis et le Lion (talk) 16:46, 9 June 2021 (UTC)
 * The English wikipedia does not operate by the standards of the French Wikipedia! Here, we do not have such standards for the admissibility of polls in French election. However, for british elections, they add in every polls from members of the British Polling Council, which Redfield & Wilton Strategies is a part of.


 * But if we were working by frwiki's standards, the poll was covered in english-language secondary sources here and there. It's also been included in this poll aggregator. Mottezen (talk) 17:29, 9 June 2021 (UTC)

Data made up or removed from graph for cosmetic reasons
We can't remove or make up data on the graph for cosmetic reasons. Please stop making up or removing data from the graph.

However, your cosmetic concerns with the graph can be resolved by other means. Maybe we could merge back the right-wing line with each individual points being an average of the right-wing candidates' results. Or we could add multiple right-wing and left-wing candidates on different lines with the same colors, with notes clearly explaining that only one of these candidates is expected to be running in the election in the end. Let's discuss it here and establish a consensus before making the change in main-space. Mottezen (talk) 19:38, 17 June 2021 (UTC)

Why does the graph not match the data at all
Seriously Justknowthatyourenotinthisthingalong (talk) 16:27, 11 October 2021 (UTC)

Thank you english-language editors
As a french citizen, I wanted to thank the editors of this english version of the page, for refraining to (dis)qualify Éric Zemmour as (the english equivalent of) "extrême droite": "extreme right", or even the less dismissive "far right" -which is not used in France, for essentially political reasons. On the french page the editors' activism is unrefrained. One could believe that current events could incite them to control themselves, to protect the neutrality of wikipedia, or even democracy in their country, but one would be mistaken. Anyway thank you to those responsible for editing this page. — Preceding unsigned comment added by 88.142.239.170 (talk) 21:18, 22 October 2021 (UTC)

I am taking care of everything
we can talk here if you want
 * Can you please explain your edits? Your behaviour is openly disruptive. You cannot just delete some polling figure you single-handedly choose to cherry-pick just because you feel like it. Please, elaborate on your motives or I'll have to assume this goes against Wikipedia policy outright. Cheers.  Impru 20 talk 10:42, 6 November 2021 (UTC)
 * could you be more precise. I have done a lot of modifications on this page. Which ones are problematic ?
 * I have summarized it very well in the edit summaries: you are removing polling data without any justification, then automatically collapsing polling tables (which goes against Wikipedia's MOS). I please ask you to undo this edit until this issue is resolved. Your edits have no consensus and you have made no effort at all to explain anything of it, other than saying "I am taking care of everything" (which sounds pretty much as ownership over the page). Please, seek a consensus before reinstating your edits and stop engaging in edit warring.  Impru 20 talk 10:45, 6 November 2021 (UTC)
 * which edit is problematic. Is it the edit about putting the whole names of candidates in vertical ? Is it the edit about putting only the HI data on the graphs because the HI polls are more regular (one per week) unlike the other polls ? What is it exactly ? We can solve the problem without undoing everything I have done, I hope.
 * Is this a joke? I've already told you what the issue is: removal of polling information without any justification against WP:REMOVAL advice (a number of polls for the first round, then the whole Bertrand vs. Le Pen set of polls for the second round, though you have since restored it) and automatically adding collapsed tables going against MOS:COLLAPSE (while I see the benefit of collapsing tables for older polls, why should all tables be collapsed by default in this case puzzles me, specially considering no effort at justifying it has been made). I don't see the benefit of splitting tables by year for those years that only had very few opinion polls conducted; that's an unnecessary waste of space and hampers the ability of comparing trends in the table. On a related note, you've made 92 edits to the article in the last 24 hours, so I don't know which exact edit (or combination of edits) resulted in this, nor am I required to waste my time checking it. It's obvious that yesterday these changes weren't there and that now they are, and that it was you the only one editing the article in that time, so you will surely know how and when did you perform such changes. I am not of the strict opinion that you have to perform all of your changes in a single edit, and see it as logical that you may be conducting your planned changes by parts. But 92 edits, really? That's counter-productive for Wikipedia, since it hampers the ability of properly reviewing the page history.  Impru 20 talk 11:04, 6 November 2021 (UTC)
 * I suggest that we do not put any graph for the 1st round until we know who is the LR candidate (because I have put Bertrand but it will perhaps not be Bertrand), and until Zemmour is officially candidate (because I have put only the HI polls since September that mention him). When there will be a graph for the 1st round, all polls (no matter which institute) will be included in the graph. What do you think about it ?
 * I fail to understand your reply. You have answered nothing of what I just told you, and I have never mentioned anything about the graph! And again, please, revert your edits until this discussion is over as per WP:BRD. I am asking you this politely.  Impru 20 talk 11:21, 6 November 2021 (UTC)
 * I see we have trouble understanding each other. I will not revert everything. I will only revert the problematic things. What are they ?
 * Yes, you should revert your edit until there is no consensus for those. WP:BRD is a Wikipedia guideline. Seriously, I've already told you what the issues were here, here, here and here, yet you keep going around in circles. I don't know whether you are just plainly trolling here or if this is a serious competence issue, but you cannot just act like this. You removed polling data, collapsed all tables against the MOS, and even as we speak you keep unilaterally splitting and dividing the tables without any consensus at all . You have not even cared to give any kind of justification for all of this mess except for this discussion's section title ("I am taking care of everything")! Either you revert your edits to the version of the article as it was yesterday and stop making such large unilateral changes until this discussion is over, or I'll have to ask for another kind of intervention here.  Impru 20 talk 12:28, 6 November 2021 (UTC)
 * I am now going to give you some justification :
 * 1) Splitting the tables like I do (Year 2017-2018, Year 2019, Year 2020, Year 2021 by months) is the best solution (unless you have a better one) : more clearer. What was the logic in From 4 June to 11 September ? Or in "since 11 September". It is, according to me, the most coherent solution.
 * 2) On lots of page that list the polling intentions for elections, polling intentions (by year mainly) are collapsed. Why not here ? I can understand that some can have objections but what are they ?
 * 3) Putting the whole names of candidates in a vertical size is a good idea. I can prevent the table to be bigger than the page (like it always happen without this system).
 * Of course, I am open to any discussion about it.
 * 1) The best solution for what? I find it as an incredible waste of space and even contrary to any sense of clarity. I can see the argument for clarity for more recent times, in which there are a lot of opinion polls and many candidates. For 2017 you had one poll. For 2018 you have two. For 2019 you have four. Another four for 2020. All of these can be shown in a single table without any issue. Then for 2021, you divide it by months, despite some months having literally one or two polls only! This is a move that you conducted unilaterally, that is ugly, a waste of space and effort and one for which I can't see any merit (PS. It is also disgusting that you don't revert the page to the status quo version, i.e. the one before your hundred edits, yet you keep editing the article as we speak, worsening the situation despite knowing you lack any consensus for all of this here). Revert-the-article-while-the-discussion-is-ongoing).
 * 2) Your edits here have nothing to do with what is done for other articles (not even the collapsing is conducted in the same way as other opinion polling articles. For example, you could take this version of the article (though it should be noted you had already unilaterally divided tables here, since the state of the tables before your intervention was this one), add a collapsible function to the tables and leave the article as it is without touching anything else. That would be comparable to other opinion polling election article, not what you have done to the page, which is horrendous). If you use comparability as an argument, then you have just killed off any reasoning for your edits. Now, as a second reminder, please revert until the discussion is over.
 * 3) Vertical orientation may be a good idea. I am not opposed to that nor mentioned it as a problem. Surely, we can work out from there upon this version of the article and wouldn't need any other of your edits to accomplish that, right?
 * If you were truly "open to any discussion", you would 1. Revert your edits as per WP:BRD, 2. Obtain the required consensus for your contested edits, or for those that do truly secure such consensus, 3. Not keep reverting others without any justification while concurrently working on the article, aggravating the situation with the discussion ongoing. Cheers.  Impru 20 talk 14:39, 6 November 2021 (UTC)
 * Hi, I'm not really involving myself in this conversation (although I agree with that your behavior has been rather WP:TENDENTIOUS). I'm just recommending that you remember to sign each of your comments with four tildes: ~ ~ ~ ~ but without the spaces separating them. Without signatures, it's very hard to keep track of who is saying what. Jacoby531 (talk) 15:04, 6 November 2021 (UTC)
 * I am just astonished. 14 edits to the page since mine's and Jacoby's replies here, with no further answer here, no attempt made at reverting the article to the status quo version until a consensus is reached and further changes conducted without any explanation. I had to revert several font sizes because 80% or lower fonts were used despite them being expictly prohibited under MOS:SMALLFONT. I'm afraid I'll have to end up bringing this to WP:ANI if this keeps going on like this, as you seem to show a clear lack of competence to work colaboratively in Wikipedia.  Impru 20 talk 22:42, 6 November 2021 (UTC)
 * I stand with the majority of users in this discussion, the changes enacted are completely tendentious, just unnecesary or have major unnecesary effect on the article. The changes to the graph in particular are completely bewildering, you don't blindsight other pollsters for lack of "regularity" or because they're perceived as better than others. Their is no precident nor neccesity for this change and frankly has a detrimental effect on the value of the graphs and the article as a whole. This whole cherade seems nothing more than a narcistic power trip as seen by your inability to accept reasonable critism of your behaviour (i mean this is in the most respectful way possible), all that we ask is that you revert the changes and seek consensus as the drastic effects of your changes should not be left to just one interpretation. EnglishPoliticalPerson (talk) 20:04, 10 November 2021 (UTC)

Opinion polling graph
I propose to use in the article this polling graph I have done with R. The graph currently used just includes Harris' data and this one includes all the polls since January 2021. Thoughts?--Basque mapping (talk) 18:43, 4 December 2021 (UTC)
 * much better especially since Bertrand isnt even candidate anymore Braganza (talk) 11:08, 5 December 2021 (UTC)


 * I don't think it's useful to have an external chart which would need someone to manually update.

I will update with Pecresse and I agree we should use all polls and will change that so as to reflect the French wiki.Tbretagne (talk) 12:51, 6 December 2021 (UTC)

Valérie Pécresse
Wouldn't the tables be clearer if only polls with Pécresse as the Republicain candidate were included in the main tables, with alternative hypothetical candidate scenarios that haven't come to fruition sequestered into another table? It's quite confusing to follow as it stands. Ralbegen (talk) 12:55, 6 December 2021 (UTC)


 * I think we should make a new table for upcoming polls given we now have most of potential candidates set.

Then we can create a final table once all the candidates have been validated by the commission. Tbretagne (talk) 13:36, 6 December 2021 (UTC)
 * Polls starting on December will probably show just Pécresse as the LR candidate and not any other hypothetical candidate, whereas before her selection Pécresse herself was also one out of many hypotheses. The best course of action would be to rename the "September-present" table as "September-November" and have another one for December depicting only Pécresse.  Impru 20 talk 13:42, 6 December 2021 (UTC)
 * I'm pretty sure I've seen some polling pages—maybe American ones?—that seperate out previous hypotheticals once candidates are known. It'd be so much easier to follow the table if the 2021 tables only included Pécresse polls, as most polls would only include one or two variants. Making a ruling on Zemmour too would mean that most polls would include only one row and it would be much easier to read and match what I think the inclusion criteria for the main graph on the page. I'm not a contributor to this page, so I don't expect my view to hold weight, but as a user of the page this is the change that would have the biggest impact on my ability to follow it! I'd also advocate shifting the fieldwork dates to the left, not colouring the backgrounds of pollster names and finding a way to not use vertical text and stop identifying parties only by colour. I've not been a contributor to this page so I'm not comfortable making bold edits, but I'm happy to implement any of these if others agree... Ralbegen (talk) 16:38, 6 December 2021 (UTC)
 * I'm also not a contributor to this page, but I think the graph at the top would be a lot more useful to readers if it just showed Pécresse and not Bertrand. As for the polling tables, I think it would be better to show all of the hypotheticals, but the graph itself should have Pécresse and not Bertrand. Jacoby531 (talk) 17:10, 6 December 2021 (UTC)
 * Ah, I just checked the page, and it looks like the change I am requesting was made since I last looked at the page! In that case, disregard my comment. :) Jacoby531 (talk) 17:12, 6 December 2021 (UTC)

Removing the Bertrand sections?
Hi, I'm just taking my daily stroll around Wiki and found this page, which looks good 👍. Now that Xavier Bertrand has arrived at a measly 4th position in the Republicains primary, would it not make sense to remove the § Macron vs. Bertrand and § Bertrand vs. Le Pen sections? They have pretty much lost their relevance. JBchrch  talk  03:44, 11 December 2021 (UTC)
 * Actioned. JBchrch   talk  18:08, 12 December 2021 (UTC)
 * I reverted your unilateral edit because 1) it didn't have consensus (the removal of such content in the past was reverted several times already) and 2) that it involves an hypothesis that didn't come into fruition (Bertrand as candidate) is not reason enough to have it removed. It still has relevance from an historical and encyclopedic perspective, and I should remind you that Wikipedia is an encyclopedia, not a newspaper. Defending that Bertrand hypotheses should be removed would be akin to removing all opinion polling previous to December 2021 because it does not involve all candidates currently confirmed, which is absurd.  Impru 20 talk 18:22, 12 December 2021 (UTC)
 * As you know consensus occurs implicitly, so the fact that you had to revert similar removals multiple times seems to indicate that you are acting against consensus. As for the NOTNEWS argument, I don't really understand it. Precisely because of NOTNEWS, we do not need to keep data that's going to be of very little relevance in 10 years, when the Bertrand thing will be a footnote of history. JBchrch   talk  18:33, 12 December 2021 (UTC)
 * Implicit consensus occurs when no one opposes an edit and it becomes implicitly accepted by everyone editing the article, not when there is an explicity opposition to such an edit. You should first get aware of what WP:IMPLICIT is, before suggesting that reverting such edits go against consensus: it was those edits removing such data that went against consensus in the first place, since no discussion or agreement took place favouring such removal. On the NOTNEWS issue, what I mean is that we do not remove information just because events overcome them. Yes, opinion polls (including Bertrand) may have an encyclopedic interest within 10 years when people studying the 2022 election analyze how one of the frontrunners in the election ended up, I don't get what's exactly your point here.  Impru 20 talk 18:40, 12 December 2021 (UTC)
 * You should first get aware of what WP:IMPLICIT is... Yeah, very useful sentence, that will move the discussion forward for sure. JBchrch   talk  18:59, 12 December 2021 (UTC)
 * Maybe I haven't been clear enough, but let me clarify things for you. As per IMPLICIT: An edit has presumed consensus unless it is disputed or reverted. Which means that the edits removing the Bertrand polls, which by themselves overwrote a version of the article which had attained implicit consensus, cannot be regarded as having attained such implicit consensus themselves since those were disputed and reverted. No attempt at discussing the removal of the Bertrand polls was ever done until now, btw, so no explicit consensus attempt has even been attempted (because it would probably not be reached, not under the premise that Bertrand is no longer candidate at least). See opinion polling articles for the 2017, 2012, 2007, 2002 and so on listing all alternative hypotheses shown in polling (even those not materializing in the end). In fact, this article currently misses some other hypotheses (such as Macron vs. Mélenchon, Hidalgo vs. Le Pen, Pécresse vs. Le Pen, Mélenchon vs. Le Pen and some others), the reason for which I cannot currently understand. Hope this has moved the discussion forward for sure. Cheers.  Impru 20 talk 19:13, 12 December 2021 (UTC)
 * Thanks, and it has. I guess the evidence from articles on prior elections is convincing to me for the moment. JBchrch   talk  20:04, 12 December 2021 (UTC)
 * Not sure if that was intentional, but you have reverted this edit as well: . What gives? JBchrch   talk  20:15, 12 December 2021 (UTC)
 * Oh, that seems an unintended error. The reversion was aimed at the removal of the Bertrand polls, not that reference. I guess it got caught in the middle, so thanks for pointing it out.  Impru 20 talk 20:19, 12 December 2021 (UTC)

Graph
If we compare the graph at #Graphical summary with the one on the French Wikipedia, there are some significant differences. Does anyone know why? Tagging. JBchrch  talk  20:24, 12 December 2021 (UTC)
 * Hi! I do not know what has led to the users that edit the french Wikipedia to use that graph but the one I created was with local regressions with R software using all the polls since 2021. Is it that you see something to better to the graph?--Basque mapping (talk) 22:32, 12 December 2021 (UTC)
 * Turns out I'm an idiot. It's just that the x axis has a different timeframe in both graphs. Thanks a lot for this graph though, it's very nice! JBchrch   talk  22:52, 12 December 2021 (UTC)

One graph or two
Given the rollings that have been made since the start of 2022, the number of polls this year is much more higher than the previous year, that is why I think that it would be preferable to separate the graph between one of the 2021 and the other of the 2022. Which are your opinions?--Basque mapping (talk) 21:50, 6 February 2022 (UTC)

March 2022
I would like to suggest, based on the impending deadline for presentations, that the first few days of March 2022 be merged with February 2022. It looks like several candidates who were regularly polled for the last few months (namely Poutou, Asselineau, Taubira, and Thouy) may not make the ballot and will likely be dropped from polling. One or more might still make it, but it would seem to make more sense to break polling at the point where presentations close than at the end of February (and thus have 2-4 "empty" columns for most of the month).70.174.126.206 (talk) 07:56, 2 March 2022 (UTC)

Cluster 17
The new institute Cluster 17 seems to be dealing with a questionable methodology which, like Sentio in Sweden or Civey in Germany, cannot achieve any representativeness. While Sentio and Civey have repeatedly attracted attention with disastrous surveys on elections, nothing is known about Cluster 17, as there are only 3 elections in France within 5 years. The listing in the article seems questionable. --[[User:Petruz|Petruz] (talk) 23:01, 4 March 2022 (UTC)

Official campaign
The OpinionWay poll for 4-7 March was published at precisely the same time as the official list of candidates, noon on Monday, 7 March, and was obviously completed before (not after) publication of the list (this was, incidentally, just a coincidence; the OpinionWay rolling poll is posted every weekday at noon). The Ifop poll for 3-7 March was published five hours after the official list and little of its fieldwork, if any, would have been completed after the list appeared. The OpinionWay poll is clearly in the wrong place. The implication that the Ifop poll is informed by knowledge of the list is misleading. The first poll conducted entirely after publication of the list is the Elabe for 7-8 March, with the pollster saying excplicitly that fieldwork began later in the day on 7 March. Utilisateur19911 (talk) 15:04, 9 March 2022 (UTC)

Violation of WP:3RR
Someone keep reverting my edits. Can you please stop him, the graph is correct. —113.252.117.217 (talk) 08:57, 15 March 2022 (UTC)

Adding Melenchon - Macron Polling
Given Melenchon recent poll rise and the fact he polls frequently in 3rd or 4th place above pecrease and sometimes Zemmour. I think it would be reasonable to include head to head polling for him. The French Wiki already includes this so it shouldn't be too difficult and as I'm aware all the data is included on that table so it's just a task of transferring it over. EnglishPoliticalPerson (talk) 19:41, 16 March 2022 (UTC)

Macron aggregate polling needs updated URGENTLY
Can we please update the polling aggregate graph?! Macron's poll numbers have slipped to an average of 29% over the past week and the graph should show a noticeable downward curve were it maintained and updated regularly. Yet it still falsely shows Macron's polling as shooting up and up away from the other candidates, which might influence voters. This close to an important election, can we please show some professionalism and non-bias and ensure that this graph is updated daily and in a truthful manner? Tetsumonchi (talk) 06:46, 24 March 2022 (UTC)


 * The graph has been updated by its creator once a week (on fridays) which I think is good enough. There ought to be other real-time polling average services for your needs. I don't think French voters rely on this article, though it is a nice one. TFerenczy (talk) 11:24, 24 March 2022 (UTC)

A Commons file used on this page or its Wikidata item has been nominated for deletion
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. —Community Tech bot (talk) 03:52, 29 March 2022 (UTC)
 * Éric Zemmour Toulon 03-2022 (cropped).jpg

A Commons file used on this page or its Wikidata item has been nominated for deletion
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. —Community Tech bot (talk) 00:37, 3 April 2022 (UTC)
 * Eric Zemmour La seule solution (cropped).jpg

Should the Macron v Mélenchon 2nd round polling be placed just after the Macron v Le Pen table?
I think the 2nd round polling should be sorted by probability (Zemmour and Pécresse seem unlikely to access the 2nd round at the moment) 2A01:E0A:25A:74A0:142C:CEEF:EF7F:AE42 (talk) 09:05, 5 April 2022 (UTC)

OpinionWay 10-11 Apr 2022
The link is to OpinionWay's previous poll. Utilisateur19911 (talk) 10:52, 12 April 2022 (UTC)

A Commons file used on this page or its Wikidata item has been nominated for deletion
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. —Community Tech bot (talk) 15:23, 10 December 2022 (UTC)
 * Jean Luc MELENCHON in the European Parliament in Strasbourg, 2016 (cropped).jpg