French presidential election, 2022

[parochial boy]

I have many questions about the expats in Paraguay and Monaco now.

Paraguay has become a popular destination for German anti-vaxxers and other assorted weirdos fleeing frome the "corona dictatorship". I haven't heard much about French people doing the same, but considering it's only a couple of hundred people, I expect the reason lies somewhere there.

[Sir John Johns]

Truly, Le roi des cons:

https://twitter.com/BFMTV/status/1519360886409994240

Presidential election: Jean Lassalle’s staging led to the nullification of the votes in his commune.

Runoff results in the commune of Lourdios-Ichère (Pyrénées-Atlantiques) where Jean Lassalle voted on last Sunday have been nullified. The reason: Lassalle, when in the polling station and in front of the ballot box, made before other voters and journalists, a political discourse in which he proclaimed he was refusing to participate to the vote and hence was abstaining. He immediately thereafter posted a video of his feat on social networks.

This is a blatant violation of electoral code you would think a deputy and a presidential candidate would be aware of as the diffusion of political propaganda messages on election day is prohibited (especially when coming from politicians) as well as the public manifestation of a political preference inside a polling station when vote is taking place (already in the first round, people wearing T-shirts flocked with a photo of Zemmour were told to leave a polling station). Such provision is aiming at preventing pressure on voters.

The Constitutional Council has ruled that Lassalle’s publicity stunt is of nature to alter the sincerity of the vote in Lourdios-Ichère as he used to be the mayor of the commune and is currently the outgoing deputy for the constituency it is part of.

I’m sure the 90 electors of Lourdios-Ichère who have cast a valid votes are very happy to see their ballots rejected because of the moronic clowning of Lassalle who just few weeks ago was complaining about democracy being no longer a thing in France and the country now living under a ‘soft dictatorship’ while signing before a notary a paper saying that if elected his assets would be seized and gave to a peasant association in case he wouldn’t have fulfill his key election promises (like issuing a decree calling for referendums before June 2022, which is probably unfeasible in such a short delay).

Since then, Lassalle, who could face judicial prosecution for his act, has asking citizens of Lourdios-Ichère to forgive him for ‘the greatest dishonor they are confronted with, because of me, since the commune is existing’ and said he would personally apologize to each inhabitant of the commune.

[Antonio the Sixth]

To be fair, the Constitutional Council's policy of throwing out all the votes in a polling station if anything remotely sketchy happens in it is absurd and potentially ripe for abuse. I know France is a somewhat more civilized country than the US for now, but what if there was a close local election and a bunch of RN activists deliberately went and sabotaged the elections in the places where they knew their opponent was strong?

[Anzeigenhauptmeister]

Map of the expat vote in the runoff is much more boring - it went 86% for Macron, on about 38% turnout. Le Pen only won in Russia (minus Yekaterinburg's 4 votes) and Almaty (Kazakhstan as a whole seems to have gone Macron by adding Nur-Sultan to Almaty), and came close in Belarus, Paraguay, Moldova and Monaco (46.2%!) and did relatively well in Thailand, Andorra and Djibouti.

Interesting fun fact about Andorra:

Did you know that the President of the French Republic is simultaneously the Co-Prince of the Principality of Andorra, along with the Bishop of the Roman Catholic Diocese of Urgell? 🇫🇷 🇦🇩
That partial personal union between France and Andorra dates back to 1607, when France's temporal head of state was still a monarch. ☝🏻🤓

[Hash]

As promised, some interesting maps:

Increase in abstention from round one:


Very interesting map: unlike in 2017, when turnout dropped everywhere (except overseas), there are very obvious regional patterns: turnout marginally *increased* in the northeast and north, except in Brittany and very obviously IDF. It dropped nearly everywhere in the south, particularly in the southwest and the former Rhône-Alpes region. In general, turnout also dropped in most urban areas, even in the north, which makes sense given Mélenchon's urban strength in the first round, and the turnout dropoff was particularly big in low-income urban areas like the 93, Roubaix, and northern Marseille. As expected, there is a strong correlation between Mélenchon's support and the increase in abstention at RSQ 0.68. But in addition to Mélenchon's vote, the other explanation is non-political: school vacations. There's a striking correlation with school vacation zones (which follow regional boundaries): IDF and Occitanie (zone C) fell on spring vacation just a day before the runoff for 2 weeks, while zone A (Nouvelle-Aquitaine, ARA, Bourgogne-Franche-Comté), which all saw drops in turnout as well, were in the middle of their 2 week spring vacations. Turnout increased or remained relatively stable in almost all zone B regions (the rest of the country), with some exceptions, which was going back to school the day after the runoff (and had been on vacation from the weekend of the first round). School vacations were also a turnout factor, perhaps an unexpectedly important one, back in 2002. But, of course, political factors like a lot of Mélenchon's voters not turning up, remain the most important factor - school vacations just added to the demobilization in certain regions...

Blank and invalid votes


Blank and invalid votes were 8.7% of votes cast (over 3 million), not as high as in 2017 (11.5%) but still the second highest in a presidential election and much higher than in the first round (2.2%). Blank and invalid votes in a runoff are always quite different from those in the first round, feeding off of eliminated candidates' voters. In general, blank and invalid votes are a rural phenomenon - it still kind of is to a large extent - as urban voters usually don't vote at all rather than bothering to show up to spoil their vote, whereas rural voters historically tended to see voting as more a civic duty and could show up even if it was just to spoil their vote. However, from the map, it's clear that some, though not all, urban Mélenchon voters discovered the pleasures of blank and invalid votes - something few of them had experimented with even in 2017 - it's quite obvious in IDF, outside of Paris, and central Marseille (highest in Mélenchon's own seat) though blank and invalid votes remain noticeably low in most of the Lyon, Lille, Marseille, Strasbourg metro areas. The correlation with Mélenchon's vote is inexistent... but the strongest correlation is with Lassalle's vote (RSQ 0.47) - and the second highest blank/invalid vote share was in Lassalle's constituency, at 14.7% (the highest, at 15.6%, was in Corrèze-1... the Chirac-Flanby heartland).

Marine Le Pen vote compared to first round far-right vote


This map compares Le Pen's runoff vote to the first round combined total of Le Pen, Zemmour and NDA, as a percentage of votes cast. Nationally, her 41.5% was about 9% better than the far-right's first round total. After making it, I realized I wasn't the biggest fan of this map: it's deceptive, because it's based on votes cast, not the electorate, so it gives false impressions about Panzergirl's strength in places like the 93 where there was a very significant turnout dropoff. Yet, it does show some things: most revealingly, she clearly, under any metric, underperformed even the modest baseline of the far-right first vote in very wealthy western Paris (where Zemmour did very well), and didn't perform much better than that baseline in other wealthy urban areas (visible in Marseille, Lyon, Lille area)...

Marine Le Pen actual result vs. 'hypothetical' runoff result model (% of electorate)


So in my quest to determine how vote transfers were for each candidate, I developed a very simple and crude 'hypothetical' result for them in every constituency based on national vote transfer data and assumptions, all based on registered voters to take into account abstention as well - in each constituency, I calculated Panzergirl's hypothetical runoff result from her own first round result plus a certain percentage of transfers from other candidates (ignoring just the 2 Trots), based either on actual vote transfer data from pollsters or my assumptions (I assumed Roussel had the same transfer patterns as Mélenchon, that 60% of NDA voters went to her, that Hidalgo had the same transfer patterns as Jadot and that 20% of Lassalle voters went to her and 20% to Macron). I also assumed that about a quarter of first round non-voters turned out in the runoff: I find this is a reasonable assumption (Ifop suggests about 35-40% of first round non-voters turned out, but I find this is way too high), and I assumed that 65% of first round non-voters went to Macron and 35% to Panzergirl (based on this chart). Of course, my assumptions are probably at least partially wrong and open to interpretation, but really the point of this exercise is not really to give precise numbers but to provide a general impression... and I think the map does this well...

As in 2017 we see Panzergirl systematically underperforming her hypothetical modelled result in basically all urban areas. Some of this has to do with turnout but it's not the only factor. On the other hand, she overperforms her modelled result in her strongholds (north) and a lot of rural/rurban/exurban areas. We can surmise from this that Mélenchon voters' transfers to her were particularly poor in urban areas (more abstained and more voted Macron) but pretty good in the north (her strongholds) and in less urban areas. To be noted that her overperformance of the modelled result in the north also has to do with the good turnout there - and perhaps my 35% of new voters going for her was a bit low...

Macron actual result vs. 'hypothetical' runoff result model (% of electorate)


Same as above, basically, just different input numbers. We see Macron overperforming his hypothetical modelled result in his strongholds - Brittany, the inner west and affluent parts of metropolitan areas (in low-income urban areas, the turnout dropoff was so bad that none of the two matched their modelled result...). In Paris proper, despite the lower turnout, Macron clearly overperformed his hypothetical numbers - Mélenchon's Parisian voters, unsurprisingly, broke more heavily for Macron than other Mélenchon voters. A closer look at the 93 shows an interesting pattern: despite the turnout dropoff, Macron still beat his modelled result in the more gentrified parts of the department, which are closest to Paris (Montreuil, Pantin etc), while underperforming because of the bad turnout in the rougher, poorer parts (Bobigny, Aulnay-sous-Bois), something also clear in the Val-d'Oise (underperforming in the constituencies including Goussainville, Garges-lès-Gonesse). This pattern of low turnout (primarily) causing him to underperform his modelled result in low-income urban areas is also obvious in Lille, Marseille, metro Lyon, even Bordeaux, Rouen, Le Havre, Amiens, Dijon,

Not shown here but of course the most dramatic over/underperformances by Panzergirl and Macron respectively were overseas...

[Hash]

Assorted first round precinct-level maps for various cities:

Lille


Bordeaux


Toulouse


Nice


Grenoble


Mulhouse


Rouen


Toulon


Perpignan including runoff


Le Havre


Metz


Caen


Interactive precinct-level maps for Paris, Lyon, Marseille and Strasbourg are already available online on different websites, and Rennes and Nantes have user-friendly official websites which include clickable maps with results included.

[Sebastiansg7]

Hi, this is a bit off topic, but I am new in the forum, and I can't see a single map you guys post. Does anyone know how to fix this?

[Zinneke]

Hi, this is a bit off topic, but I am new in the forum, and I can't see a single map you guys post. Does anyone know how to fix this?

Hi welcome to the mad house

You have to post something like 15 posts before you can see our dickpics for copy right reasons

[Good Habit]

Hi, this is a bit off topic, but I am new in the forum, and I can't see a single map you guys post. Does anyone know how to fix this?

It might also depend on what browser you are using - for me, MS Edge ignores many pictures, while they are well displayed with Opera...

[Hash]

I took a quick look at polling place results in the (Sephardic) Jewish neighbourhood of Sarcelles (95), which primarily corresponds to precincts 21, 22, 24 and 25:

These precincts have voted quite heavily for the mainstream right since 1995, giving well over 50% to Sarkozy by the first rounds in 2007 and 2012, and a bit less but still gave huge margins to Fillon in 2017. In 2002, it was Alain Madelin (DL), the most pro-Israeli/Jewish candidate, who did very well there.

In the first round this year:

Precinct 21
Zemmour 167 votes (35.23%)
Macron 120 votes (25.32%)
Mélenchon 109 votes (23%)
Le Pen 38 votes (8.02%)
Pécresse 13 votes (2.74%)
Valid votes 474 (57.59% of electorate)

Precinct 22
Zemmour 196 votes (39.52%)
Macron 143 votes (28.83%)
Mélenchon 91 votes (18.35%)
Le Pen 32 votes (6.45%)
Pécresse 14 votes (2.82%)
Valid votes 496 (56.62% of electorate)

Precinct 24
Zemmour 207 votes (37.1%)
Macron 161 votes (28.85%)
Mélenchon 91 votes (16.31%)
Le Pen 29 votes (5.2%)
Pécresse 33 votes (5.91%)
Valid votes 558 (55.97% of electorate)

Precinct 25 (more mixed, I think)
Mélenchon 184 votes (35.05%)
Zemmour 152 votes (28.95%)
Macron 121 votes (23.05%)
Le Pen 32 votes (6.1%)
Pécresse 12 votes (2.29%)
Valid votes 525 (53.63% of electorate)

In the runoff:

Precinct 21
Macron 308 votes (76.05%)
Le Pen 97 votes (23.95%)
Valid votes 405 (49.21% of electorate)

Precinct 22
Macron 352 votes (82.05%)
Le Pen 77 votes (17.95%)
Valid votes 429 (48.97% of electorate)

Precinct 24
Macron 354 votes (78.32%)
Le Pen 98 votes (21.68%)
Valid votes 452 (48.97% of electorate)

Precinct 25 (more mixed, I think)
Macron 342 votes (77.73%)
Le Pen 98 votes (22.27%)
Valid votes 440 (45.38% of electorate)

However, the Sarcelles Jewish vote (and the French expat Jewish vote in Israel) is the most politically distinctive - looking at precincts in Paris-19, Créteil and Villeurbanne, I don't find anywhere near this level of support for Zemmour (at best he's slightly over 10%), although those precincts may be more socially/ethnically mixed and given shifts from 2012, I suspect the Jewish population has decreased. From that, I wouldn't conclude that Zemmour won the Jewish vote - he probably won the most politically distinctive, right-wing segment of the Jewish electorate which lives in turbulent places like Sarcelles (or Israel) which have reinforced their sense of identity/community.

One thing that's quite clear though: even in Sarcelles and Israel, Panzergirl remains very weak with the Jewish electorate, even if they're pretty far-right themselves.

P.S.: For images, I suspect that it's because new members are restricted from seeking linked/hosted images. If you want to see them, you could probably quote the post and copy the imgur/whatever site url to view them there. 

[Zinneke]

Hi, this is a bit off topic, but I am new in the forum, and I can't see a single map you guys post. Does anyone know how to fix this?

It might also depend on what browser you are using - for me, MS Edge ignores many pictures, while they are well displayed with Opera...

Especially to some Chinese operative!

Firefox or Chromium ftw

[Antonio the Sixth]

I guess we've moved on, but I made a map of the evolution of the left vote on the first round from 2017 to 2022:



Top left: Mélenchon 2017 -> Mélenchon+Roussel 2022
Bottom left: Hamon 2017 -> Jadot+Hidalgo 2022
Right: Total Left 2017 -> Total Left 2022 (including the Trotskyist candidates, which I didn't bother mapping directly)

Some interesting patterns here. The Mélenchon+Roussel map isn't too different from the Mélenchon map I already made, except with a higher baseline. Thankfully that means their combined total grew almost everywhere, but the bulk of that growth remains concentrated in IdF, the DTOMs, and Rhône-Alpes. Even combined, they still manage to underperform or just barely tie Mélenchon's 2017 in some parts of the Pyrenees and in much of the postindustrial Northeast (Pas-de-Calais being the worst underperformance of all, with Mélenchon+Roussel over 0.8 points under Mélenchon's 2017 result. Overall, a map that shows further concentration of the left-wing vote in France's metropoles.

Unlike with Mélenchon and Roussel, Hidalgo and Jadot managed to somehow underperform Hamon's total in 2017 as a percentage of registered voters, winning a combined 4.6% against Hamon's 4.82%. This is illustrated by the map showing them losing ground around large swathes of France. Unlike Mélenchon, they don't have Ile-de-France or the DTOMs to bring them up - in fact, their underperformance is even worse in those areas (-0.88 and -1.74, respectively). They also lost significantly in old left-wing strongholds like Pas-de-Calais and Ardennes in the North, much of the Pyrenees in the South, and Nièvre and Limousin in the center, as well as in Finistère which I think was Hamon's home turf. On the other hand, their overperformance was even more notable than Mélenchon in almost all of France's East, stretching all the way from Bas-Rhin to Alpes-Maritimes (and, a personal satisfaction for me, with an epicenter right around the Savoies). This broad swatch traditionally included some of the weakest areas for the French left, but at least some parts of it appear to be increasingly amenable to it. A lot of it has to do with trendy emerging metropoles there (Lyon, Grenoble, Strasbourg, even perhaps Nice) but that can't be all there is to it. Also, Pays de la Loire sticks out really weirdly in that map too, which is weird given that it's one of the regions with the least distinct identity.

The Total Left map is largely a superposition of the first two, as the trots had little meaningful trend. Generally speaking, the left improved everywhere, but by far disproportionally so in IdF, the DTOMs, and Rhône-Alpes, and to a lesser extent also Alsace, PACA and Franche-Comté. In less friendly areas, the most "metropolitan" departments still stand out from their surroundings as areas where the left gained most (Haute-Garonne again in particular). Meanwhile, the Pyrenees area once again stands out as an area of particular decline for the left (probably in large part a product of Lassalle's gains, although I suspect there's more to it than that). There also seems to be a particular weakness to the North of the Massif Central. This stretch around Creuse, Allier, Nièvre, Cher and Indre is a historical patch of left-wing strength (whether communist like in Allier or socialist like in Nièvre), representing the old tip of the old C-shaped arc of left-wing strength to the South. It's also one of the most rural parts of France, smack into the middle of the "Empty Diagonal", so the left's decline here can definitely be seen as part of Muh Trends. Finally, we can also see a decline of the left, once again, in its old Northern strongholds, once again Pas-de-Calais but also Seine-Maritime and Ardennes. The left did improve slightly in Nord, but I'm willing to bet that's mostly because of Lille.

[President Johnson]

Macron has been inaugurated for his second term today.

https://twitter.com/nytimes/status/1522968273607868422

[Хahar 🤔]

Today was the day Emmanuel macron finally became president

[Antonio the Sixth]

Last map because even I'm getting bored of this, but I've been working on this one for a while (before I got distracted by the legislative playing field taking shape. Here's a map comparing last month's runoff results to 2017's - Macron, Le Pen, and assorted non-votes.



Macron lost ground everywhere, and Le Pen gained ground everywhere except in New Caledonia, where turnout just plummeted across the board. The overall pattern is something we've seen already in previous map, but it bears repeating: the swing was most significant around the Massif Central and the Pyrenees, overlapping somewhat (but far from perfectly) with the traditional area of left-wing strength in the Southwest. The DTOMs also stands out as big Le Pen gains, as do to a lesser extent Brittany and the rural Alpine parts of PACA. By contrast, the swing was more muted in a lot of the North and East, places where Le Pen had already done very well in 2017 (but also Rhône, which was already an FBM stronghold and now stands out even more). Ile-de-France appears to have seen significant Macron losses, but little Le Pen gains, which leads us straight to the third map.

Probably the most interesting of the three, the map shows which places voted less (red) or more (green) than they did in 2017. Going in, it was really hard to say which way it would go: on the one hand, the closer race should motivate more people to turn out; on the other hand, Macron being the incumbent and having fallen short of the (already very low) expectations of left-wing voters meant that some of the people who voted holding their noses in 2017 might not bother this time. In the end, it seems like these two effects canceled each other out: the share of valid votes cast as % of registered voters dropped by just 0.22 points (interestingly, the share of blank and spoiled votes went down while abstention went up). That movement was not uniform across the map, however. Once again, IdF stands out as an area where significantly fewer people bothered to vote this time (and within IdF, the 93 in particular, a surefire tell of Mélenchonist rejection for Macron). Turnout in the big DTOMs also dropped massively, signaling that Le Pen's landslides there aren't quite as impressive as they look (although it went up in Mayotte, indicating some genuine groundswell of support there). Aside from those, the areas that turned out left are mostly found in that old Pyrénées-Massif Central region, although even there the drops are usually fairly small. In most of non-IdF metropolitan France, there was actually quite a bit of mobilization compared to 2017, especially in the Northeastern belt that makes up Le Pen's old stronghold (but also in some big-city areas like Haute-Garonne). The uptick in the non-IdF North might have something to do with school vacation pattersn, as Gaël's map a while ago showed. However, the trend is far more generalized across France to be reduced to that. A lot of people did seem to feel the need to get involved this time even if they'd skipped 2017.

[Antonio the Sixth]

Way late to the party, but I've been working on this map for ages and it's finally done:



This uses the same "ideological blocs" that I used in previous maps (though I've also decided to fill in the blanks by giving 50% of the Lassalle vote to the left and 25% to each of the two rights, as that happens to produce an almost perfect 33-33-33 split). Left is red, liberal right is green, sovereignist right is blue, and the various combinations of colors reflect the split of this three blocs in any given constituency. For example, a yellow-ish constituency would have about an even split of left and liberal, but few nationalist votes. White constituencies are almost perfect national bellwethers.

Basically, I've always wanted to make a map using this kind of "triangular" color scheme and realized this was the golden opportunity. I hope the result was worth it!