French Elections 1848-2010

[Vasall des Midas]

Well, yes. Then again, the problem may be right here:

Considered as "left" all parties wich admitt themselve this qualification.

The pattern of MODEM voters for second rounds the last few years has not exactly been to support UMP en bloc.

[Colbert]

Well, yes. Then again, the problem may be right here:

Considered as "left" all parties wich admitt themselve this qualification.

The pattern of MODEM voters for second rounds the last few years has not exactly been to support UMP en bloc.

yes, i admitt this point, but I don't considere than all non-left block belong to a right-block.

The only (from my point of view) way to calculate force of the lefts is to objectivically aggregate all parties wich themsleves considere belong to left.


If I would had made a graph of aggregate of the rights, I would not had include MODEM

[Vasall des Midas]

Yeah, not including them in such a historical graph makes sense... drawing conclusions from the graph for this year's presidential election without taking the factor into account is what doesn't.

[Hashemite]

Something big will be up tomorrow (hopefully). Stay tuned or take guesses.

[Californian Tony]

Looking forward !

[Hashemite]

Here it is!



For 2007 comparisons: http://i257.photobucket.com/albums/hh225/hashemite/Cantons2007.png

Errors are possible, given that in a lot of places the Ministry has decided to troll us around by giving us 'inverted' results; this is the case in Lyon, Garches, Suresnes, La Garenne-Colombe, Romilly-sur-Seine, Chapelle-Saint-Luc and so on. I've self-corrected the obvious errors myself, some others might not be so obvious and appear wrong on this map.

[Californian Tony]

This is great. It's interesting to compare the two maps.

[Hashemite]

Wow well that's disappointing. Nobody seems to care

[Vasall des Midas]

Oh, I do care.

[Hashemite]

TEH FNS STOLESZ TEH VOTES FRUM TEH COMMISZ!

[Vasall des Midas]

What exactly is that graph supposed to prove?

Now, if "selected working-class areas" was instead "PCF's best cantons of 1985"...

[Hashemite]

What exactly is that graph supposed to prove?

Now, if "selected working-class areas" was instead "PCF's best cantons of 1985"...

That there is a strong negative correlation between strong PCF performances and strong FN performances, which holds true not only for 2010 but also for 1995. And still holds true even if comparing the FN's 2010 performance with Hue's 1995 performance (or Hue 1995 with JMLP 2002). Everybody with half a brain has proven that the FN has tended to perform below average in most historic PCF strongholds, and that the FN's best performances in working-class areas are/were in those which were historically right-wing. Apparently the [lack of] correlations could be even starker at a precinct level.

'selected working-class areas' are a sample of 130ish communes including historic PCF strongholds, strongly PS areas and right-wing areas.

[Vasall des Midas]

That there is a strong negative correlation between strong PCF performances and strong FN performances, which holds true not only for 2010 but also for 1995.

Strongish. It's clearly not that great.

that the FN's best performances in working-class areas are/were in those which were historically right-wing.

Yes, that sounds reasonable to me.

[Hashemite]

FTR, the correlation numbers are:
1995: -0.5557 (strong)
Jospin R2-Le Pen R1 1995: -0.5018 (strong)
2002: -0.484
2007: -0.0848 (the PCF numbers are likely way too tiny to be significant)
2010: -0.3164 (medium)
PCF 95-LP 2002: -0.4761
PCF 95-FN 2010: -0.36328

Strongish. It's clearly not that great.

Well, obviously. Nothing is ever black and white, but pretty clearly the media's favourite theory that the PCF's decline is caused in large part by their voters going FN is fabricated bullsh**t. The correlation tends to be weaker in inner suburban working-class hinterland and far stronger in old, more ruralish, PCF working-class villages/mining towns and so forth.

[Vasall des Midas]

FTR, the correlation numbers are:
1995: -0.5557 (strong)
Jospin R2-Le Pen R1 1995: -0.5018 (strong)
2002: -0.484

Yes, that's definitely "strong" rather than "strongish" as far as political correlations go.

Nothing is ever black and white, but pretty clearly the media's favourite theory that the PCF's decline is caused in large part by their voters going FN is fabricated bullsh**t.

Journalists. Voting patterns. Need I say more? Next thing we'll be discussing how journalists have no clue about football support culture. Or how politicians have no clue about voting patterns.

(The issue is ancient. I recently read a book about where the nazi voters came from.)

[big bad fab]

Oh, I do care.

Me too.
But it's rather depressing for me: it was the good old times, when the right was still pretty strong in inner West, in Ile-de-France or in all the Massif Central.
Corrèze is almost readable immediately in this map !

[Californian Tony]

Oh, I do care.

Me too.
But it's rather depressing for me: it was the good old times, when the right was still pretty strong in inner West, in Ile-de-France or in all the Massif Central.
Corrèze is almost readable immediately in this map !

The 1995-2012 swing in Corrèze will be hilarious to see.

[Hashemite]

Even though this election year is boring as sh**tte, it certainly has led to some very welcome developments in French electoral geography. I've found a website which gives some fantastic result maps within major cities at a poll-by-poll level. To wet your appetite:

[Hashemite]

Nobody seems to care, but here's an interesting map:

[homelycooking]

Good idea. Does this correlate at all with the decline in the vote for Laguiller and the PCF?

[Californian Tony]

Nobody seems to care, but here's an interesting map:

This is very interesting ! Could you explain the geographic patterns ?

[Sibboleth]

Commies for Besancenot, I think.

(oh, and we all care a great deal. This is one of the best threads in the forum. It's just that it's not always easy to think up of an intelligent response to such extended prettiness).

[Hashemite]

Nobody seems to care, but here's an interesting map:

This is very interesting ! Could you explain the geographic patterns ?

The main trends are a weakening of the Besancenot vote in urban areas, especially those which are bobo/well-off/middle class and a pretty important strengthening in rural working-class and urban working poor/suburban hinterland. Notice stuff like the 93, Lyon, Grenoble, Marseille, Lille or the Seine-Maritime. If exit polls are to be believed, he held only 40% or so of his 2002 vote in 2007, losing a significant amount of voters to Royal as part of the 'vote utile' but gaining a much more working-class base through people who had voted Arlette or Hue in 2002 or those Hue95>Arlette02 voters. The LCR, up until 2002, had attracted a much more bourgeois/revolutionary youth/bobo clientele than LO which has always had a fairly proletarian base. They totally lost it in 2007, might have started regaining it with the NPA in 2008, lost it to EELV in 2009.

btw, I picked up Goguel's atlas of sorts of elections between 1871 and 1956. Anybody up for some old maps?

[Californian Tony]

I see, Besancenot saw the erosion of his traditional bobo/trendy electoral base but benefitted from the regrouping of the "left of the left" under his name as a result of the utter failure of Laguiller/Buffet/Bové.

It's fun when you think about how pathetic the total "left of the left" vote was pathetic in 2007. The "useful vote" mechanic was indeed very strong at this time. So, quite ironically, there are probably a sizable share of Royal voters who won't vote Hollande this time (otherwise Mélenchon's standing is hard to explain). Which also tells us how much Hollande has taken away from centrists or center-leftist (Bayrou supporters in 2007).

[Sibboleth]

btw, I picked up Goguel's atlas of sorts of elections between 1871 and 1956. Anybody up for some old maps?

Always.