France 2012: the official thread

[big bad fab]

2012 Big Bad Tracker #24 - 17 October 2011      
            
Each opinion poll is ponderated with its sample and loses a ponderation of 15% each week: it may change after the autumn.            

I haven't given a ponderation based on the quality of pollsters, as big changes have occurred in their teams, without any certainty that the 2007 rating is still OK.      
And this tracker is built more out of fun than out of scientific seriousness...       

      

Hollande   30,53
Sarkozy   22,25
Le Pen   16,27
Mélenchon   6,90
Joly   5,58
Bayrou   6,63
centre-right candidate   6,30
Villepin   3,43
Dupont-Aignan   0,41
Boutin   0,37
Nihous   0,04
Arthaud   0,90
Poutou   0,32
Chevènement   0,04
   
Hollande   58,33
Sarkozy   41,67

No new poll this week, so Sarkozy is just keeping heading down and Mélenchon heading up.

Hopefully, we'll soon have some polls, with 2 big changes:
- Borloo is out
- Hollande is the survivor.

Expect a big boost for Hollande, probably towards 35%.
It'll be interesting to see if Sarkozy is at least able to grasp 2 points from Borloo.
Bayrou should be slightly up. Morin should be tested.
Pay close attention to Joly and Mélenchon, in order to see if Aubry's defeat means a boost for them (I doubt it, though Mélenchon may be up again, as he is clearly on the rise).

[big bad fab]

The strength of Hollande with 6 groups of 16 departments.

Well, it's mostly the same map as in the first round.
Montebourg's, Royal's and Baylet's appeals have had effect only in their own local strongholds: Burgundy, Poitou-Charentes, Tarn-et-Garonne and few other departments with leftist radical tradition.

It gives us a clue on vote transfers from 1st to 2nd round (and of my big failure in predicting the 2nd round ): vote transfers were pretty homogeneous and didn't have, overall, an effect on the hierarchy of the 2 final candidates.

After all, maybe polls weren't bad at all: 56,5%, that's what they have predicted before round 1, including OpinionWay, which was then very accurate....

Maybe the 1st round results were the real.... outlier !
Maybe many "natural" Hollande voters didn't vote because they thought he would be ahead anyway. And maybe some Hollande voters had some "fun" in 1st round, exactly like in a presidential election, when Besancenot, Laguiller, Villiers,... have bigger results than anticipated.

[Californian Tony]

Fab, can you give us a last tracker for Aubry, so that we know exactly where she stood when she lost ?

[ZuWo]

I'm certainly not adding anything new to the discussion, but these numbers look awful for Sarkozy. It seems Sarkozy needs a DSK-esque scandal by Hollande two weeks before the election in spring to win ...

[Hashemite]

Strong Aubry departments which lack a major city:
Landes (Emmanuelli?)
Hautes-Alpes (good Montebourg vote transfers?)
Haute-Loire
Ardeche (though the north is basically Lyon exurbs, though that is favourable to Hollande)
Ain (see above)
Yonne (see above)
Drome (probably Green-effect)
Haute-Marne (imo, the weirdest)

You might say Eure and Somme don't really have a huge city, but they're more urbanized than the above and are close to Aubryist-Fabiusian bases.

Any ideas or explanations?

[Vasall des Midas]

So, you f**king got to beat Sarkozy in 2012. Or I swear God you'll pay.

And you bloddy well got to beat him by more than you beat Aubry!

[big bad fab]

Fab, can you give us a last tracker for Aubry, so that we know exactly where she stood when she lost ?

I'm sorry, but no: already erased her datas
Basically, it's the same thing as the week before...
Sorry again.


As for Hash's questions, well,
Eure is under Seine-Maritime influence and Parisian influence, like Yonne.

Hautes-Alpes has been good for Greens in the past years.
And maybe there is a personality factor: rough people in Queyras and Briançonnais who like a rough woman...
It may be the same in some parts of Somme, Meurthe-et-Moselle. But, sure, it doesn't work in Lozère, Doubs, Vosges or Ardennes

Ardèche has Lyon's influence, sure, plus a Green minorty vote along the Rhône, plus Dussopt, the young PS deputy who has been very active in her campaign (he is a "fan" of the northern lady, you know ? ).
Haute-Loire is maybe a mix of Ard-èche and Hautes-Alpes explanations.

Landes is of course the land of Emmanuelli (Hamon did big there in 2008, and in Pyrénées-Atlantiques too).

Drôme has a big Green effect (Michèle Rivasi, chemical industry, nuclear plants) and Lyon's influence too, plus an old Mermazian tradition (more than Isère, which had Dubedout and has now Destot or Vallini), which means quite a leftist and orthodox local PS.

Haute-Marne had a special result in 2008: the ecologist wing of the PS was strong only here. I don't know what this "pôle écologique" is made of now, but that can be an explanation.

Sometimes, I think there are also some regional votings: I mean, Burgundy was loyal to Montebourg even towards Hollande.
Maybe Eure, Ardèche, Drôme, only vote like the rest of their region.
There are many counter-examples of course, but which we can explain: Ille-et-Vilaine is not really Britaany , Aisne has the very special Thiérache in it, Hautes-Alpes isn't entirely in "Provence"...

And, eventually, remember, there may be only 0.1% between the lowest department of a colour and the highest of the other colour .

[Tender Branson]

The next polls that will come out will probably show Hollande destroy Sarkozy by 65-35.

Well, not so bad - but almost:

New CSA poll: Hollande 62%, Sarkozy 38%

http://www.tagesschau.sf.tv/Nachrichten/Archiv/2011/10/19/International/Sarkozy-Herausforderer-Hollande-liegt-klar-vorn

[Californian Tony]

Fab, can you give us a last tracker for Aubry, so that we know exactly where she stood when she lost ?

I'm sorry, but no: already erased her datas
Basically, it's the same thing as the week before...
Sorry again.

Heck Fab, I didn't imagine you'd be so petty ! As soon as she lost you wanted to have the satisfaction to "erase" her, literally ! That's not a very "sportive" attitude...

So, anyways, here is her final, adjusted graph :




And now what's become the official presidential tracker's graph :




So far, the numbers look amazing for Hollande. He only has to maintain such a lead, and victory is assured. Still, I long to see what numbers will look like once we'll have the first post-Borloo and post-Primary polls.

Fab, I advice you to switch your ponderation differential to 20% now. With the PS having its candidate, pollsters will fully focus on the main race and we'll probably see enough polls to make the tracker viable that way.

[big bad fab]

Eh, if I've done what you've already said, we would have run out of polls...
Remember, French pollsters are lazy and French medias are very suspectful towards polls (hence, order less and hence weaken polls' reliability ). I'll wait a bit as I'm not so sure we'll have many polls until... Xmas.

In CSA (I'll publish it tomorrow), Hollande is already at 35%. Morin and Lepage are in but Borloo dissipated with giving only 1 or 2 points to Sarkozy...
Honestly, how can he win ?

[Hashemite]

CSA, September 19-20 2011
1010 people 18+, of which 859 RVs.

Hollande 35% (+7)
Sarkozy 25% (+1)
Le Pen 16% (-2)
Bayrou 9% (+2)
Melenchon 5% (-1)
Joly 3% (-1)
Villepin 2% (-3)
Boutin 1.5% (+1)
Morin 1%
Arthaud 1% (-0.5)
Poutou 0.5% (=)
Lepage 0.5%
NDA 0.5% (=)

Hollande 62%
Sarko 38%

Other bad stuff for the right:
61% of people say they certainly won't vote for Sarkozy, and his vote potential is 36% to Hollande's 57%
60% of UMP sympathizers want an open primary on the right
50% of righties say the PS comes out the primaries united and strengthened
13% of righties say Hollande is "really left-wing"

Not as awful stuff:
Sarkozy is the favourite of UMP and righties for "best candidate of the right" stuff. But only with 44% and 39% respectively.

https://docs.google.com/viewer?url=http%3A%2F%2Fwww.csa.eu%2Fmultimedia%2Fdata%2Fsondages%2Fdata2011%2Fopi20111017-le-bilan-des-primaires-socialistes-et-les-intentions-de-vote-pour-l-election-presidentielle-de-2012.pdf

[Math]

In CSA (I'll publish it tomorrow), Hollande is already at 35%.

And let me assure you he will be higher in tomorrow morning's poll, it gonna be really big

What I like in this CSA poll is the analysis: they dare to say Hollande is at a fair level among the working-class voters when he polls at 28%, 7 points under his polling total. Well, ok, he is first among this group, but he is tied with Marine... Clearly, Terra Nova was right earlier this year, saying the working class is lost for the left in France, too.

BTW, I was just noticing that many pollster now have a sample of 1000 respondents from which they extract about 850/900 registered voters. I'm not sure, but IIRC, in 2007, the standard was a 1000 RV's sample, right?

[Californian Tony]

Eh, if I've done what you've already said, we would have run out of polls...
Remember, French pollsters are lazy and French medias are very suspectful towards polls (hence, order less and hence weaken polls' reliability ). I'll wait a bit as I'm not so sure we'll have many polls until... Xmas.

Well, I understand your concerns. It's a shame because having a lot of polls right now would be absolutely necessary, but if you are right it's indeed wiser to stay at 15% until december... And maybe to stay at 20% until the very end. That way, we're sure it remains a tracker and not a simple polling average of the moment.

In CSA (I'll publish it tomorrow), Hollande is already at 35%. Morin and Lepage are in but Borloo dissipated with giving only 1 or 2 points to Sarkozy...
Honestly, how can he win ?

His odds are pretty low right now and to win he would ned Hollande to make a (maybe several) major mistakes, but let's not exagerate things : nobody is "done" with 6 months to go to the election, and there is always the possibility of an upset.

So, you f**king got to beat Sarkozy in 2012. Or I swear God you'll pay.

And you bloddy well got to beat him by more than you beat Aubry!

No, sadly that won't happen, Lewis. Don't trust the current polls, because they will inevitably go down a bit. Presidential elections in France are always extremely polarized, and it's almost impossible for a candidate to win over 55% (even De Gaulle only barely went over this threshold).


Anyways, here's the famous map I was talking about Saturday :



Hollande-Aubry relative margin in the 1st round, ie the margin of victory minus the national margin, useful to see more clearly each candidate's strength.

I'll do the same map for the 2nd round as soon as we have the official data (BTW, when are they going to count these damn 18 polls left ?!?).

[Keystone Phil]

Chances Sarkozy just doesn't run?

[Californian Tony]

Chances Sarkozy just doesn't run?

Impossible. If there is one certainty in this campaign, it's that Sarkozy will be the UMP candidate.

[Tender Branson]

The next polls that will come out will probably show Hollande destroy Sarkozy by 65-35.

Well, not so bad - but almost:

New CSA poll: Hollande 62%, Sarkozy 38%

http://www.tagesschau.sf.tv/Nachrichten/Archiv/2011/10/19/International/Sarkozy-Herausforderer-Hollande-liegt-klar-vorn

It's getting closer ! New BVA poll:

Hollande 64%, Sarkozy 36%

http://www.expatica.com/fr/news/french-news/socialist-hollande-would-crush-france-s-sarkozy-poll_183275.html

[Californian Tony]

These numbers are marvellous... Too bad they won't last.

[I left.]

These numbers are marvellous... Too bad they won't last.

What would your actual prediction be if the election were held today?

[Californian Tony]

These numbers are marvellous... Too bad they won't last.

What would your actual prediction be if the election were held today?

Today ? My guts would say 58/42 at very best. French people have never given commanding margins to any presidential candidate, even when the polls indicated it was possible. Right now Hollande has a big mo' because of the universal hatred toward Sarkozy, the primaries and Borloo's retirement driving a few center-right voters who don't like Sarkozy toward him. But even then, I'd easily see 3-5% of voters saying they will vote Hollande an then vote Sarkozy.

[big bad fab]

I'm lazy, guys.
If you want details on CSA and BVA recent polls, go to my blog (link in my signature). In French but results are quite clear

Of course, I'll keep publishing my tracker here, as usual.
Hollande will bounce big !

[Hashemite]

If anybody thinks Hollande or anybody can win nearly 40% in the first round, I'd like to sell them my beachfront property in Saskatchewan.

[Californian Tony]

If anybody thinks Hollande or anybody can win nearly 40% in the first round, I'd like to sell them my beachfront property in Saskatchewan.

Yeah, he will end up in the very low 30s at best (and that will still be a great score).

[big bad fab]

If anybody thinks Hollande or anybody can win nearly 40% in the first round, I'd like to sell them my beachfront property in Saskatchewan.

Yeah, he will end up in the very low 30s at best (and that will still be a great score).

Indeed, there is the foreseeable effect of the primary, the fact that these are the first polls since Borloo dropped out and the fact that BVA didn't include any other rightist or centre-rightist candidate than Sarkozy (and didn't even add Lepage).

We are entering a weird period: Hollande will try to be more discrete (clever, I think), while Sarkozy will be burned out by the crisis
Hollande may well still be over 32 in January.
Of course, in April, he'd be glad to have 28: Bayrou, Mélenchon, maybe some small candidate will have better numbers. And, of course, there is Panzergirl, again the big question mark.

[big bad fab]

I'm lazy, guys.
If you want details on CSA and BVA recent polls, go to my blog (link in my signature). In French but results are quite clear

Of course, I'll keep publishing my tracker here, as usual.
Hollande will bounce big !

And, today, an IFOP poll, close to CSA numbers, with a 60-40 for the second round.
To be published later today.

[Senator MaxQue]

Does the new prostitution scandal which is perhaps related to DSK will hurt PS?