France 2012: the official thread

[Insula Dei]

I see than Aubry launched the "machine à perdre".

Which helps the right and is a very bad idea from her.
Does she want Hollande to lose, so she can run against Sarkozy in 2017?

This is one quote people. If she starts a bitchfest this monday, I'll be more than ready to condemn her along with some of you, but right now, she's still only a Margin of Error away from being the PS candidate for the presidency and not playing it very dirty, as far as I can tell.

[big bad fab]

Believe me, the words she has pronounced for 2 weeks, but notably today, will stick in the memories of all the Hollande supporters inside the PS apparatus.
Especially if she wins.

If she wins, of course, she'll be able to keep all her very diverse team (former leftist strauss-kahnians, some leftist rocardians, the "strong" left of Hamon-Emmanuelli-Lienemann-Quilès, the traditional aubrysts, Delanoµë's supporters, former and remaining fabiusians, etc.) but she'll have a bitter internal opposition from Hollande, Moscovici, Peillon, Valls, Royal, Ayrault, Rebsamen, Sapin, etc.

If she loses, Fabius, Delanoë and Hamon have been careful enough not to appear as attack dogs and will drop her...
But, in this case, I'm not sure it will be easy for her to take back the premiership inside the PS: don't forget Harlem Désir is only here ad interim.... This is another battle in sight: she won't want to quit, but Désir is well viewed by many of Hollande's supporters and by many of Delanoë's and Fabius' supporters, so she will endure big pressures to retire.

This is double or quits and she's now taking a big risk for the PS, in a selfish way.
Just saying.

[big bad fab]

BREAKING !

Montebourg's "personal vote" will be for Hollande, though he don't want to amke any appeal to his voters.

Well, even if it smells like Chirac for VGE in 1981, it's really good news for Hollande.
He has received a support from all the other candidates and Aubry is still on attack mode...
Montebourg would vote for the "candidate of the system" ?
Her final line of attack is dead.

Logically, it should be enough to prevent Aubry's momentum to reach the 50% threshold.
Though we should remain careful.

[Tender Branson]

Yeah, look a bit bad for Aubry.

I'll probably stick with my prediction of a 56-44 win for Hollande. But anything between 53-60% should be possible for him.

[big bad fab]

Yeah, look a bit bad for Aubry.

I'll probably stick with my prediction of a 56-44 win for Hollande. But anything between 53-60% should be possible for him.

It will be narrower than that.
Probably not more than 52-48.
(I'll do some maths this evening, before being more precise than that... and so fail even more on Sunday )

[minionofmidas]

I'm guessing at 54-46, maybe 53-47.

[Antonio the Sixth]

Yeah, it's getting bad for Aubry... She's given her explanation on France Inter this morning (an attack against polls, not against Hollande) but of course that won't convince anybody. It was a major gaffe and she know it, all the more that the medias seem to give a lot of coverage to this sentence. I think the hopes for her to prevail are somewhat decreasing, but things remain uncertain enough. Right now, a Hollande 52% victory as Fab said seems the most likely scenario...

Anyways, whether it's Hollande or Aubry, the only thing that really matters is kicking Sarko's ass in 2012. And I trus both candidates and their supportes to gather after Sunday with this aim. I mean, socialist know how to be terribly stupid sometimes, but there are limits.


Anyways, here is the promised candidate strength map ! Sorry for the small resolution, but you can click on the pic to view it in real size.



As you can see, what is striking is really the map's uniformity : Hollande always either in the 30s or 40s, Aubry between the 20s and 30s, Montebourg mostly in the 10s (considering he got 17%, we could have expected him to break 20% in a lot more departements), and all the others almost never breaking 10% (Valls is especially striking : he barely got 11% in his native Essonne !). This is a huge difference, not only with America's polarized politics, but also with the interal PS maps (like the Reims ones) which are usually quite messy.

Fabien, if you have done your own map, could you show us them ? I'm curious to see which of ours are better (certainly yours ).

[Math]

Very nice maps!

If I can suggest something to Fab (only if his maps are not already done, of course), it would be to choose a discretization by quintiles or quartiles, in order to see better where are the strongholds of the little candidates (Royal, Valls).

[Antonio the Sixth]

Yeah, I also considered to use increments of 5 points for minor candidates. I definitely should have, but didn't imagine vote shares would be that uniform...

Another interesting thing is to notice how Hollande weakness doesn't necessarily corresponds to Aubry strength and reciprocally : there are a lot of departements where Hollande is in the 40s and Aubry in the 30s, as well as places where Hollande is in the 30s and Aubry in the 20s. That's quite weird.

[big bad fab]

Very nice maps!

If I can suggest something to Fab (only if his maps are not already done, of course), it would be to choose a discretization by quintiles or quartiles, in order to see better where are the strongholds of the little candidates (Royal, Valls).

These are the only maps I love (but we of course need the 2 types of maps: one without the other isn't enough !). I've done some with 6 groups of 16 departments (so, sextiles ?).
But they are a bit dirty
And I'm busy right now: please wait a bit .


As for Aubry, her attacks aren't gaffes: it's a clear strategy of escalation.

And she is taking a big risk: that's what Chirac and Séguin made in 1994-95 against Balladur. Of course, they won, but, until 2007 or even until now, this is the dividing line inside the right.
Sure, we can't compare entirely the situations and the risk is here a bit lower. But...

Montebourg's decision seems to be highly minimized on Rue89, Mediapart and Libération (and they talk a lot about education, retirement and all the "lies" of Hollande...).
Le Monde is more neutral, just because they are so glad to have the interview, while being probably more aubryst...
Only Le Nouvel Observateur is more pro-Hollande and gives more room to Montebourg's decision.

Those who are already convinced among Montebourg voters will say that he is a traitor... and they'll vote Aubry... and Mélenchon in April 2012

I don't think Montebourg's decision will change much to the real vote, but it weakens deeply this last critic of Aubry on the establishement, the system, as Montebourg can't be said to be part of it. And it gives some reality to the "rally" around Hollande: well, nobody has endorsed Aubry... remember Hillary in 2008 ?

Still, Aubry will strike this chord of "angriness", "anti establishment", "we don't want to do what they are talking us to do", etc.
And this is a strong theme in France.

So, the only advantage for Hollande is that, mediatically, Montebourg counter-balances personal attacks by Aubry. But it's just the advantage of making the field even again.

[Antonio the Sixth]

If it's a strategy, it's a losing one for both her and the left. And I can't believe someone who has done so much for her party in order to win 2012, who wasn't even going to run weren't for DSK, now has become power-hungry to the point of willingfully adopting a strategy which is objectively bad for the PS. Until now, there is some aggressivity on Aubry's side, but that doesn't necessarily means it's a strategy. We'll see, but I feel the medias are making somethin big of very little actual stuff.

[big bad fab]

Hollande 51.5 / 48.5 Aubry

Yeah, this is what I think:
1/3 - 2/3 among Montebourg voters, with 20% of them abstaining
1/3 - 2/3 among Royal voters, with 30% of them abstaining
2/3 - 1/3 among Valls and Baylet voters
2/5 - 3/5 among new voters, these being 7% of the previous total
hence a 51.58 / 48.42 result, though spontaneously, I'd have said 50.8 / 49.2

We'll see, guys.

And I'm too lazy to publish here the latest (and last) Harris poll, again 53/47 (see my blog).

[Antonio the Sixth]

I'd rather say 2/5-3/5 for Royal voters and 3/4-1/4 among Valls and Baylet voters, but you get the main line. Hollande will likely win. Aubry's efforts to reverse the momentum have been relatively successful, but not as much as Fab expected (retrospectively, you gave me false hopes and I shouldn't have trusted you as much ) and not enough to take over Hollande. Well, that's a pity. An upset is still possible, of course (after all, I went to bed on november 21, 2008 with the dire perspective of having Royal as First Secretary ).

[big bad fab]

Let's see if it works:

Baylet:


Valls:


Royal:


Montebourg:


Aubry:


Hollande:


EDIT: Hollande map modified (there was indeed a problem with Cantal, Aisne and Loir-et-Cher: thank you, Antonio )

Remember it's only the relative strength of the candidates: each map has 6 groups of 16 departments.
Of course, this method can't be used alone, especially here, when Hollande is high everywhere and Aubry in many departments, with very homogeneous results, notably for Montebourg.
And the method is worse and worse when you have smaller and smaller overall results.

Still, many trends appear clearly: regional strongholds, old radical bastions, the south-east for Montebourg, the "real" Brittany for Hollande, etc.
It's always troubling to have hierarchies even in the tenths of points

More comments later, but do not hesitate to "find" something in each of these maps, guys !

[Antonio the Sixth]

Very interesting maps, thank you Fab ! They are probably less pretty aesthetically speaking, but far more useful to analyze geogrpahically.

Let's see... The Baylet map doesn't seem to mean anything outside of the obvious strength in Midi. Here the sextile stuff doesn't work very well, considering 90% of departements gave him between 0 and 2%. Valls definitely looks like a bobo vote, with IdF, Alsace and Rhône-Alpes as main strongholds. The Royal map doesn't seem to mean much except than the South hates her a bit less than the rest of the country. Montebourg stregth is pretty strikingly concentrated in the Southeast (I'm extremely surprised to see he's actually lower that average in the northeastern "desidustrialization belt" from Lorraine to NPdC where his rhetoric should have played well). Aubry is "northern" in an almost caricatural way. Finally, the "empty France" is in love with Hollande...

Edit : Cantal should be black on Hollande map, as he got 58% there.

[Antonio the Sixth]

Also, last tracker graphs before we know the candidate !


Aubry :



Hollande :




It's too bad pollsters are unable to survey two races simultaneously... We still haven't had one damn poll since Borloo stepped down.


Fab, I had a quick look at your tracker numbers on your blog and it looks like you messed things up somewhere : your blog indicates in the Hollande scenario "Joly 6,57 / Bayrou 5,66" while it's the reverse of what you posted here. Also, in both scenarioes, you seem to have flipped NDA and Boutin's numbers. I assumed the correct version was the forum one (as it's the most coherent with previous numbers). Check this out.

[Beet]

What is the position of this Montebourg gentleman on the euro? If he were truly strong in his principles he would not have endorsed Hollande. The PS has been completely captured by the neo-liberals.

[Antonio the Sixth]

What is the position of this Montebourg gentleman on the euro? If he were truly strong in his principles he would not have endorsed Hollande. The PS has been completely captured by the neo-liberals.

Montebourg supported "European protectionism" so he didn't really campaign against the Euro. Anti-Euro folks probably voted for him anyways.

[Beet]

Is "European protectionism" kind of like the preservation of free trade within Europe but raising tariff barriers to countries outside it?

[Antonio the Sixth]

Is "European protectionism" kind of like the preservation of free trade within Europe but raising tariff barriers to countries outside it?

Basically, yes.

[big bad fab]

Let's go, with Baylet.

Of course, there are the usual strongholds of left radicals: Midi-Pyrénées, from Hautes-Pyrénées to Lot, Corsica, Eure-et-Loir and Eure, Charente-Maritime, inner East, Ain.
It's amazing to see these strengths on the map, whereas he had less than 1% in almost all the departments !

But Hollande has crushed him in some places, either by making him too small for the map, or (more likely) by stealing him some votes: Haute-Saône, Dordogne, Indre are very good examples.

When you see this map and Hollande's map (especially in Tarn-et-Garonne, maybe more than 2/3 of Baylet voters will choose Hollande in the second round...

[big bad fab]

At first sight, I found this map weirdly close to an old FN map

But it's a bit more than that.
The weakness in a great Centre or Centre-West is due to Hollande's strength among "reasonable" people.

Strongholds are around big cities and especially "aware" and modern cities: Paris and close Parisian Basin, Strasbourg, Toulouse, Nice, Lyon, Montpellier, Tours, Orléans, Reims,...
He is really low in rural deserts and in the old workers North.

[Hash]

Valls definitely looks like a bobo vote, with IdF, Alsace and Rhône-Alpes as main strongholds.

Valls' map might look bobo, but his electorate was mostly wealthy peoples. The bobo-electorate voted heavily for Martine. He might have gotten some bobo-ish young professionals who are well off, for example.

[big bad fab]

As I've already said, Aubry's map make me think a bit of some LO-LCR maps, with this crescent (err... no... it's not a bad joke ) from Brittany to Alsace. Of course, it's not perfect, because Lorraine is quite low for Aubry.

Of course, regional influence is a greater criterion, with Aubry herself, with Fabius, with Delanoë and with Emmanuelli.
I doubt Destot alone may have won a good score in Isère for Aubry (after all, Vallini is "hollandais"). Even more for Delaveau in Ille-et-Vilaine not really a Great Leader, eh ?

Hence why I think the Green vote is readable in this map: Isère, Ille-et-Vilaine, Maine-et-Loire, Loire-Atlantique, Savoies, Alsace.

And really a "bobo" vote: Paris, Rennes, Grenoble, Lyon, Nantes.

It's not a very strong map, as it's related to some local bigwigs and as you can see she is low where you've got Royal OR Montebourg OR Hollande high.
But, of course, "weak" maps aren't what counts in urbanized countries...

[Keystone Phil]

What time will the polls close today (CET)? What's the best link for results?