France 2012: the official thread

[big bad fab]

IPSOS poll for Le Monde, Radio France, France Télévision, 16-22 August 2011, 404 potential voters in PS primaries out of a whole sample of 3677

At least, we know that the real sample is 404 (which isn't so bad), based on the "potential voters", who are supposed to be 9% of the total (well, 404 is more than 10% of 3677... don't ask me why...).

These potential voters include 71% of socialists, 8% of Front de Gauche people, 6% of Greens, 4% of far-leftists, 3% of MoDem people, 3% of UMPers, 2% of FN people, 3% of no-party.
Which seems to be logical.

31% have more than 60 years old, 32% between 45 and 59, 19% between 35 and 44, 18% between 18 and 34.
This is in line with other pollsters and this is good for Hollande.

31% are retirees, 14% are other inactive people, 17% are employees, 15% are workers, 10% have superior posts.
In line with other pollsters and good again for Hollande, the candidate of old and retired people .

Among potential voters / among the whole sample:
Hollande 42 / 45
Aubry 31 / 26
Royal 18 / 9
Montebourg 5 / 1
Valls 3 / 2
Baylet 1 / 0
don't know 10 / 17

Good numbers again for Hollande BUT some elements aren't so good for him: he seems as though he is plateauing....

Forecast by potential voters:
Hollande 42
Aubry 29
Royal 9
Montebourg 1
Valls 1
Baylet 0
don't know 18

This may seem good for Hollande, but it just says that, in fact, forecasts aren't better for him than votes: that may mean that, apart from his supporters, there is some sort of rejection of him. Royal supporters aren't all naive and clearly see that she is doomed. But they don't seem to acknowledge Hollande's advantage.

My pick is definitive:
Hollande 51
Aubry 49
Royal 50

Another result which is a bit disappointing for Hollande (in the sense that it's not better for him than for the others).

Now, the second round is not so good for him:
Hollande 53 - 47 Aubry
with 62% of Hollande voters sure of their vote and 58% of Aubry voters.

Royal voters support Aubry at 53% and Hollande at 31% (16% don't know).
This is a change in comparison to spring polls. But it can be explained just because those of Royal supporters who preferred Hollande have already joined him (hence his rise in the first round since mid-May and the long decline of Royal since the beginning of 2011).

Who is the best ?
Hollande is ahead or on par on 8 items out of a total of 9, but Aubry isn't completely crushed, far from it:
on proposals, he is at 31, she is at 32
on embodiment of left values, he is at 29, she is at 40
on managing an international economic crisis, he is at 37, she is at 30.

Of course, she has a problem on 2 items:
on character, he is at 36, she is at 27 (but that could have been worse)
on ability to win against Sarkozy, he is at 48 and she is at 28: this is of course the main point on which he can make a difference in the end, especially if it's razor-thin in the 2nd round.

Well, all this gives us a suspenseful campaign, after all, as debates may well push Montebourg or even Valls up, as they'll become more known, and may well kill Aubry or Hollande if they make mistakes.
By putting Royal even more down (no, she can't go up ), they may favour Montebourg or Aubry.
And the 1st round results may change all the landscape for the 2nd round, if Hollande isn't high enough.

In late June, medias were talking about Aubry rise (whereas she was just trying to erase the gap) and now they are talking about Hollande being the favourite without hesitation (whereas he is strong but not enough to be sure of his election at all).
If Hollande is still the favourite in late September, but do not have a big result, he may lose momentum and the medias will kill him for the 2nd round.

Last numbers from this poll, forecasts among the whole sample of French people:
Hollande 48
Sarkozy 38

Aubry 41
Sarkozy 47

Royal 25
Sarkozy 66

It's bad for Aubry, but I'm not sure primary voters will be aware of these obscure figures ....

[big bad fab]

Well, if I make another tracker, that would be for PS primaries
But polls are really too erratic: you've got 3 almost the same day and then nothing for some weeks. Too bad.
Everyone should do like YouGov in the UK

[big bad fab]

IFOP poll for le Journal du Dimanche, 23-26 August 2011, 854 self-declared leftists out of a whole sample of 1968

This is a finer sample.

Among leftists / among socialists
Hollande 41 / 47
Aubry 31 / 31
Royal 13 / 11
Valls 6 / 5
Montebourg 5 / 4
Baylet 1 / 0
none of them 3 / 2

Hollande 55 / 58
Aubry 45 / 42

It's a bit boring to repeat things: Hollande is stronger, especially among old people, men, retirees, these categories which are also those who are likelier to GOTV the 9th of October.

Still, in a second round, Hollande isn't above 60.
But Aubry has really a lot of work to do.
We'll see if La Rochelle is good for her, as she is doing quite well, I think. But only formally: as for the ideas, medias don't help her.

Aubry is even is bit weaker in the first round and Royal is a bit up again, with Valls around 5 rather than around 3.
That's very small, but it's not good for Aubry: she hasn't managed to convince during the summer.

Debates will really be important for all of them.
Especially if it's "anybody but Hollande", especially for Royal and Montebourg (I can't think Valls would harm him and Baylet is too clever to weaken his party towards any of the socialist bigwigs ).
We'll see, some primary debates in the US have been awful for some...

[big bad fab]

2012 Big Bad Tracker #17 - 29 August 2011               

My aggregated tracker with a generic socialist candidate is unchanged this week: 0.6 Hollande and 0.4 of Aubry, based on their current probability of being the PS candidate.
            
Each opinion poll is ponderated with its sample and loses a ponderation of 15% each week until October at least (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...       

      

29 August Aubry sub-tracker      

Arthaud   0,54
Poutou   0,50
Mélenchon   6,36
Chevènement   0,20
Aubry   25,73
Joly   5,40
Bayrou   7,03
Borloo   8,07
Villepin   3,46
Nihous   0,00
Boutin   0,54
Sarkozy   25,57
Dupont-Aignan   0,54
Le Pen   16,07
   

29 August Hollande sub-tracker
   
Arthaud   0,87
Poutou   0,40
Mélenchon   6,04
Chevènement   0,10
Hollande   27,13
Joly   5,40
Bayrou   6,83
Borloo   7,90
Villepin   2,83
Nihous   0,00
Boutin   0,50
Sarkozy   25,40
Dupont-Aignan   0,54
Le Pen   16,07


29 August generic PS candidate sub-tracker   
   
Arthaud   0,74
Poutou   0,44
Mélenchon   6,17
Chevènement   0,14
PS   26,57
Joly   5,40
Bayrou   6,91
Borloo   7,97
Villepin   3,08
Nihous   0,00
Boutin   0,51
Sarkozy   25,47
Dupont-Aignan   0,54
Le Pen   16,07

OK, here we are: after the very small numbers of polls since the beginning of July, we've got an overvaluation for the last CSA one.
But I don't want to change the tracker, because it would mean to take into account in an even larger way the old ones.
And since June, we've got some big changes: Aubry is officially candidate, DSK was in again and out again and maybe in again and definitely out, La Rochelle has occurred and, what is more, the Big Kriziz is back.

So, we have to leave with those poor French medias and those lazy pollsters: no data ? No smooth changes in trackers...

But it's the last big vacation before the election.
So, let's be patient for 2 or 3 weeks, and polls will be numerous again.
The Xmas recess won't be long enough to produc the same effect .

[big bad fab]

Hopefully the new bump in the Bettencourt affair will kill Sarko's momentum once for all.

I'm not so sure: it smells too much the Le Monde-Libération manipulation (I even suspect them to have given the info to Aubry this morning so that she can be the first to demand another inquiry... come on, guys, it's too clumsy... even Valls seem to have understood this ).
And they are less efficient than Rue89 or Mediapart in this kind of stuff
Sarkozy, friend of some ugly rich persons ? Not really something new...

And well, DSK isn't back yet: so, the "soap opera" will probably erase the effect of these words by an angry judge.

I think the Karachi affair (and all the dirt around it, I mean 1995 presidential campaign) has a far more killing potential for Sarkozy. This really frightens me.

In Karachi, you've got at once the moral aspect (people killed as a consequence of money), the classical money scandal and the political remnants of the latest great rightist divide (Chirac vs Balladur).
The perfect affaire d'Etat, à l'ancienne.


Aubry is really everywhere since this week-end. But is her message really clear ?
It's funny to see her act like a classical party chief, just at the moment when she is no longer PS first secretary... Again, it's not very good politically, I think.
And she must hate DSK now, as he will completely blur her attempt to come back.

As for the Senate, if it switches to the left on the 25th of September, I don't if the fact that the socialist leader there, Jean-Pierre Bel, is a "hollandais" will have any importance.
(positive, with some images of Hollande around the conquered High Chamber, or negative, as Hollande may appear as the "local barons" man)

[big bad fab]

Hopefully the new bump in the Bettencourt affair will kill Sarko's momentum once for all.

I'm not so sure: it smells too much the Le Monde-Libération manipulation (I even suspect them to have given the info to Aubry this morning so that she can be the first to demand another inquiry... come on, guys, it's too clumsy... even Valls seem to have understood this ).

ZOMGZ Conspiracy !

No, don't say that, that's not the way I'm reasoning. I know that, nowadays, you can't anything anymore, because there is always a suspicion of manipulation either of the facts themselves, or of the interpretation of these facts or alleged facts.

No, when I say "manipulation", it's only on the calendar and on the way the info is given and used in the medias, not on the info itself.
(I of course don't know if it's true; it wouldn't be surprising at all to see Sarkozy having taken money from Liliane)

As for Aubry's reaction this morning, I've found it too precise and well-balanced to be a spontaneous answer...
or maybe she is really very, very clever and I should support her, after all

[big bad fab]

Yes, that might be as you say. It's quite possible that they waited the right moment to reveal the story. As we say, "c'est de bonne guerre".

And again, Hollande and Moscovici were better than Aubry : they attacked Sarkozy on a system of "pressures" as a whole, not on Bettencourt affair particularly.

All the more that the nurse who saw money given to Sarkozy according to the judge has just said that it's wrong...

Now that Aubry "has always something to say" (le professeur Rollin a toujours quelque chose à dire ! don't you remember ? ), she is prone to say also some BS.
At least, it's interesting as the PS internal campaign is now very active. Fine !

Royal is trying to tilt towards Hollande again, after having been kind to Aubry, after having first been closer to Hollande.
She is really out of control .

[big bad fab]

Let's wait for the primary debates... I'm sure Aubry will trounce The Wimp.

Hollande has strong polls, but, contrary to many medias, I think he appeared weak, on a personal level, during La Rochelle.
Royal seems to paralyze him...
And he has a real difficulty to be tough in front of other people, which could be quite bad during a debate...

[big bad fab]

OpinionWay-Fiducial poll for LCI and Le Figaro, 31 August-1 September 2011, sample 1311 leftist RVs out of a whole sample of 3202

Among those 1311 leftist RVs, 819 are socialist and 222 are the "likely voters" in the primary.

Among leftists / socialists / LVs
Hollande 44 / 48 / 42
Aubry 30 / 31 / 29
Royal 13 / 14 / 21
Montebourg 8 / 4 / 7
Valls 4 / 3 / 1
Baylet 1 / 0 / 0
don't know 11 / 4 / 0

Of course, LVs is a tiny sample, but Royal is at +7 in this sub-category ! Hollande at -4 and Aubry at -1.
Hollande has a good popularity, but it's a bit less clear in electability.
And what is surprising is that the first criterion for voting (ability to beat Sarkozy) is amazingly important for LVs especially. Interesting, as Hollande should be the best to fight against Sarkozy if you believe the other polls.

2nd round:
Hollande 57 / 57 / 53
Aubry 43 / 43 / 47

Among LVs, Aubry is at +9 ! Again, the sample is really tiny. But nothing is done.

Royal supporters are 25% behind Hollande and 51 % behind Aubry
Montebourg supporters are 42% behind Hollande and 39% behind Aubry.
Valls supporters 50% and 32%.
It may not mean anything, but it sounds pretty logical.

Same conclusion as before, Hollande still stronger, but not enough to be quiet, far from it.

[big bad fab]

IFOP poll for le Journal du Dimanche, 30 August-2 September 2011, 1918 RVs out of a whole sample of 2036

CSA should be ashamed when you see the big samples of the latest IFOP, IPSOS, OpinionWay, even Harris polls...

Hollande 29 / Aubry 25 / Royal 17
Sarkozy 23.5 / 24 / 25
Le Pen 18.5 / 19.5 / 19
Borloo 6 / 6.5 / 8
Bayrou 6 / 6.5 / 8
Joly 6 / 6 / 9
Mélenchon 6 / 6 / 7
Villepin 2.5 / 3 / 3
Boutin 0.5 / 0.5 / 0.5
Dupont-Aignan 0.5 / 0.5 / 0.5
Nihous 0.5 / 1 / 1
Chevènement 0.5 / 1 / 1.5
Poutou 0.5 / 0.5 / 0.5
Arthaud 0 / 0 / 0

Second round:
Hollande 59
Sarkozy 41

Aubry 54
Sarkozy 46

Hollande stronger, Sarkozy not really on the rise anymore, though he is usually low in IFOP polls and Le Pen high.
(Royal is again behind Le Pen: she has been weak for
so many months now...
Borloo and Bayrou are down and we have now 4 candidates around 6 !
Beware, under 5%, no public money !!!

At last, one pollster takes Nihous into account. Difficult to say from whom he has stolen his points. Probably Le Pen and Dupont-Aignan.

[big bad fab]

OK, the US have Zogby.
France had CSA in 2007.
We may have LH2 for 2012 !!!!
(or is there really a classical polarization around the socialists and Sarkozy ?)

LH2 poll for Yahoo!, 2-3 September 2011, sample 959

Hollande 35 / Aubry 30 / Royal 18
Sarkozy 27 / 27 / 29
Le Pen 11 / 12 / 12

You've already fallen from your chair and your back is hurting ?
Fortunately, for the small ones, it's a bit less mad:

Borloo 6 / 8.5 / 10 (Borloo is the 2012 version of Bayrou 2007 if you follow LH2 )
Bayrou 6 / 6 / 8
Joly 6 / 4.5 / 10
Mélenchon 4 / 5 / 5
Arthaud 0.5 / 0.5 / 1
Poutou 0 / 0 / 0.5
Villepin 2 / 3.5 / 4
Boutin 2 / 2 / 2 ( after all, they may be right )
Dupont-Aignan 0.5 / 1 / 0.5
Nihous not even tested

Hollande and even Aubry at incumbent-Mitterrandesque levels ?
Sarkozy back towards his 2007 levels ?
Panzergirl almost as bad as Big Daddy the last time ?
Around 60% for the UMP and the (serious) PS candidates

Come on....

And, what is more, they've published it while I've already updated my tracker
I'm forced to update it again before posting it, as it's better to minimize its impact already this week with the big IFOP one.

[big bad fab]

2012 Big Bad Tracker #18 - 5 September 2011

PROVISIONAL !                

My aggregated tracker with a generic socialist candidate is unchanged this week: 0.6 Hollande and 0.4 of Aubry, based on their current probability of being the PS candidate.
            
Each opinion poll is ponderated with its sample and loses a ponderation of 15% each week until October at least (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...       

      

5 September Aubry sub-tracker      

Arthaud   0,24
Poutou   0,37
Mélenchon   5,89
Chevènement   0,54
Aubry   26,49
Joly   5,44
Bayrou   6,50
Borloo   7,35
Villepin   3,30
Nihous   0,52
Boutin   0,89
Sarkozy   25,16
Dupont-Aignan   0,63
Le Pen   16,68
   
   
5 September Hollande sub-tracker

Arthaud   0,33
Poutou   0,36
Mélenchon   5,53
Chevènement   0,27
Hollande   30,13
Joly   5,83
Bayrou   6,21
Borloo   6,43
Villepin   2,47
Nihous   0,26
Boutin   0,89
Sarkozy   24,89
Dupont-Aignan   0,50
Le Pen   15,90


5 September generic PS candidate sub-tracker
   
Arthaud   0,29
Poutou   0,36
Mélenchon   5,67
Chevènement   0,38
PS   28,68
Joly   5,67
Bayrou   6,32
Borloo   6,80
Villepin   2,80
Nihous   0,36
Boutin   0,89
Sarkozy   25,00
Dupont-Aignan   0,55
Le Pen   16,21


Yes, I've included LH2.

This is a poll and I'm not able to ponderate French pollsters regarding to their accuracy in previous elections. We have too few data to do this.

Of course, Le Pen seems hugely undervalued hugely by LH2. PS seems hugely overvalued. Sarjozy is probably too high.

But, after all, when we all said that Harris Interactive was wrong about Panzergirl in January of February, then almost all the other pollsters put her very high.

Maybe LH2 has found a new trend, or some new trends.
Each pollster has its method, regarding this big problem: are people giving the same intentions for Girly than for Daddy ?

Maybe IFOP is overvaluing Panzergirl.
And CSA had Sarkozy at the same level, roughly.

We don't have enough polls to afford to drop one of them .
And it gives some suspense .
And a tracker is precisely here to give us an average of all this (LH2 counts "only" for a quarter of the total in the current results of our tracker).

[big bad fab]

Fab, the new French pbower???

(Just joking)

Why are you so aggressive with me, Julio ?


Hold on, guys ! You are going to be happy !
Tonight, I'll modify my yesterday's tracker, as IPSOS has published, this morning, a new poll made this last week-end.

IPSOS poll for Radio-France, France-Télévisions, Le Monde, 2-3 September 2011, sample 963

Hollande 30 / Aubry 27 / Royal 19
Sarkozy 22 / 23 / 23
Le Pen 17 / 18 / 18
Borloo 7 / 7 / 9
Bayrou 6 / 6 / 7
Joly 5 / 6 / 7
Mélenchon 5 / 5.5 / 8.5
Arthaud 2 / 2.5 / 2
Poutou 0.5 / 0.5 / 0.5
Villepin 4.5 / 3.5 / 5
Dupont-Aignan 1 / 1 / 1
Boutin and Nihous not included

IPSOS is more in line with IFOP, with socialists high but not above 30, with Sarkozy under 25 and Le Pen still threatening for him.

Borloo is down and Villepin slightly on the rise.
Arthaud is really "high" in IPSOS polls. Joly is down here too.

The internals are as usual, with Hollande having a wider base than Aubry and a centre ground deeply splitted but very fluid.

IPSOS worries me a bit since the beginning of this 2012 season:
Boutin and Nihous still not included; Villepin higher in Hollande scenario than in Aubry one; Mélenchon higher in Aubry scenario than in Hollande one;... there are some weaknesses in what was the best pollster in 2007.
But, again, one pollster is right and we don't know which one and won't know until the end .

Anyway, with an addition of 963 people, the effect of LH2 will really be diluted.
Tonight, a new tracker .

[big bad fab]

2012 Big Bad Tracker #18 - 5 September 2011

THE REAL ONE !                

My aggregated tracker with a generic socialist candidate is unchanged this week: 0.6 Hollande and 0.4 of Aubry, based on their current probability of being the PS candidate.
            
Each opinion poll is ponderated with its sample and loses a ponderation of 15% each week until October at least (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...       

      

5 September Aubry sub-tracker      

Arthaud   0,71
Poutou   0,40
Mélenchon   5,81
Chevènement   0,43
Aubry   26,60
Joly   5,56
Bayrou   6,39
Borloo   7,28
Villepin   3,34
Nihous   0,41
Boutin   0,70
Sarkozy   24,71
Dupont-Aignan   0,71
Le Pen   16,95

   
   
5 September Hollande sub-tracker

Arthaud   0,67
Poutou   0,39
Mélenchon   5,42
Chevènement   0,22
Hollande   30,10
Joly   5,66
Bayrou   6,17
Borloo   6,55
Villepin   2,89
Nihous   0,21
Boutin   0,70
Sarkozy   24,29
Dupont-Aignan   0,60
Le Pen   16,13



5 September generic PS candidate sub-tracker
   
Arthaud   0,69
Poutou   0,39
Mélenchon   5,58
Chevènement   0,30
PS   28,70
Joly   5,62
Bayrou   6,26
Borloo   6,84
Villepin   3,07
Nihous   0,29
Boutin   0,70
Sarkozy   24,46
Dupont-Aignan   0,64
Le Pen   16,46



Now, you've got IPSOS in also.
So, this is the real, exhaustive, tracker. This one is fit to be included in your graph, Antonio.

LH2 is only 20% of it.
And as there are also some surprising things in the CSA and IPSOS polls, it's not so awful to have it too.

[big bad fab]

Harris Interactive poll for Le Parisien, 31 August-5 September 2011, sample 888

A smaller sample than usual for Harris.
And still no Boutin or Nihous
(I really don't understand them: Boutin isn't sure to have 500 signatures, but Dupont-Aignan even less as he failed in 2007 while she succeeded in 2002; as for Nihous, he is the surest, among small candidates, to have his signatures; at least, those who don't include Chevènement are a bit more coherent, as I think his noise about a possible candidacy is just a way to "be alive" on the political scene)

Hollande 28.5 / Aubry 24 / Royal 15
Sarkozy 23.5 / 24 / 24
Le Pen 18 / 20 / 20
Borloo 7 / 7 / 9
Bayrou 7 / 8 / 9
Joly 8 / 7 / 10
Mélenchon 3 / 5 / 7
Arthaud 1 / 1 / 0
Poutou 0 / 0 / 1
Villepin 3 / 3 / 4
Dupont-Aignan 1 / 1 / 1

When I've put 0, it's because Harris says "<1". The problem is that even with 0.5, we are above 100%... Why do they use .5 (which is, in itself, a good idea) but not for some Poutou's and Arthaud's results
Polls aren't a mathematical object, I guess

Anyway, Borloo down, Bayrou up / Joly up, Mélenchon down : this may be coherent, but the latter pair of results isn't in line with other pollsters.
And a difference of 13.5 between Hollande and Royal ? Yeah, she'd be a bad candidate, but still...

The good news is that pollsters are active again, though I hope the PS primaries won't lead to a halt in polls for the presidential election itself.

The bad news is that, apart from IFOP, all the others seem to have weakenesses, small ones for IPSOS, OpinionWay and CSA, bigger for LH2 and Harris (Sofres and BVA ? Well, we don't know for sure, their publications are so scarce...).

This is the exact situation where a tracker is needed .

[big bad fab]

In the second year, I had a course on UK politics with Jacques Leruez (pretty satisfactory: it was the year when Maggie was ousted ) and a course in English on US social sciences (not really my first choice...., though I tried to slip a bit, by picking a big study on TR ).

I was far luckier in the 3rd year, with a course on Central Asia with Olivier Roy (YESSSSSSSSSS !!!!!!! My best course ever !) and a course in English on Soviet politics with Marie Mendras.
My exact first choices .
So, let's stay hopeful, Antonio .

What is your section, Antonio ? Something like the old PES (politique économique et sociale) ?

[big bad fab]

There were 4 sections, long ago, or 4 tracks or streams or branches, as you wish:
- SP (service public), in order to prepare ENA and other civil service schools,
- EcoFi (économique et financière), not yet another business school at that time,
- PES (politique, économique et sociale), for future teachers, journalists, communication specialists
- RI (relations internationales).

It seems to be far more complicated now.
Will you specialize only in fourth and fifth years ?

[big bad fab]

ViaVoice poll for Libération, 8-9 September 2011, 275 self-declared socialists among 501 self-declared leftists among a whole sample of 1007

We don't have all the results yet.

Among the whole sample / among leftists / among socialists
Hollande 33 / 40 / 49
Aubry 15 / 22 / 24
Royal 10 / 12 / 9
Montebourg 3 / ? / ?
Valls 6 / ? / ?
Baylet 1 / ? / ?
none of them 18 / ? / ?
don't know 14 / ? / ?

Popularity (among the whole sample I guess):
Hollande 55 (=)
Aubry 45 (-4)
Royal 33 (-1)
Valls 30 (+4)
Montebourg 27 (+3)

No big change here, with Hollande still far ahead.

But the debates may well have an impact for Montebourg and Valls if they are efficient.
And Aubry will of course perform better: if she performs far better (but only in this case), then she can break Hollande's momentum.
I'm repeating myself...

[big bad fab]

2012 Big Bad Tracker #19 - 12 September 2011               

My aggregated tracker with a generic socialist candidate is unchanged this week: 0.6 Hollande and 0.4 of Aubry, based on their current probability of being the PS candidate.
            
Each opinion poll is ponderated with its sample and loses a ponderation of 15% each week until October at least (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...       

      

12 September Aubry sub-tracker      

Arthaud   0,76
Poutou   0,33
Mélenchon   5,70
Chevènement   0,35
Aubry   26,20
Joly   5,77
Bayrou   6,63
Borloo   7,22
Villepin   3,31
Nihous   0,35
Boutin   0,59
Sarkozy   24,61
Dupont-Aignan   0,76
Le Pen   17,41


12 September Hollande sub-tracker   
   
Arthaud   0,74
Poutou   0,33
Mélenchon   5,05
Chevènement   0,18
Hollande   29,89
Joly   6,02
Bayrou   6,29
Borloo   6,60
Villepin   2,91
Nihous   0,18
Boutin   0,59
Sarkozy   24,18
Dupont-Aignan   0,67
Le Pen   16,37
   

12 September generic PS candidate sub-tracker
   
Arthaud   0,75
Poutou   0,33
Mélenchon   5,31
Chevènement   0,25
PS   28,42
Joly   5,92
Bayrou   6,43
Borloo   6,85
Villepin   3,07
Nihous   0,25
Boutin   0,59
Sarkozy   24,35
Dupont-Aignan   0,71
Le Pen   16,79

More or less, we are back to the situation pre-August recess.

Sarkozy is a bit up, Hollande too and Borloo is down in every hypothesis.
Le Pen, Mélenchon and Joly are up or down, depending on the hypothesis.

But Sarkozy's gains are very small and, as Borloo has lost ground, the socialists, especially Hollande, are really in very good shape.

Sarkozy is a bit like McCain '08: whatever the news, whatever he does, every situation or event eventually fires back or simply fades away, while some should normally favour him (Libya, financial crisis, DSK, Guérini,...).

And with a new war inside the right maybe in the making with Robert Bourgi's declarations, the path is even darker for Sarkozy (though I think Bourgi is just trying a personal vendetta against Chirac, Villepin and... Juppé as the latter recently made all he could to sideline him).

Wait a bit and Karachi will re-surface...

President Hollande ? Bah...

[big bad fab]

Fab, I think by now you can switch to 0.7/0.3 for Hollande. Not that it matters (I've never followed the generic tracker), but I think that would be a fairer estimate of probabilities.

Anyways, will you switch the ponderation downgrade to 20% after the primaries ? I think that would be fair.

60-40 still seems a likely result for the 2nd round Hollande-Aubry, don't you think ?

I find Hollande a bit too over-confident these days.
I find him less clear and efficient in public meetings.
I think that, anyway, I have to wait for Thursday's debate.

The medias, even many of leftist medias, have tilted towards him in the 2 or 3 last weeks, that's true, but I'd tend to consider this as a sign that Aubry is, on the contrary, able to pull a surprise .

Though the fact she fled from BFM TV's interview was a very bad point for her.


As for the ponderation, I think I'll indeed switch to 0.2 after the PS primaries. Fair indeed: that'll mean 5 weeks plus the current week, i.e. 6 weeks taken into account in a way.

Of course, nearing the launch of the official campaign (in 2007, February was already very "rocky", but March saw the real stuff beginning), I'll even switch to 0.25.
Except if pollsters don't give us more polls than today .

[big bad fab]

Re-posting this

ViaVoice poll for Libération, 8-9 September 2011, 275 self-declared socialists among 501 self-declared leftists among a whole sample of 1007

but with all the numbers in.

Among the whole sample / among leftists / among socialists
Hollande 33 / 40 / 49
Aubry 15 / 22 / 24
Royal 10 / 12 / 9
Montebourg 3 / 2 / 1
Valls 6 / 4 / 5
Baylet 1 / 0 / 0
none of them 18 / 10 / 5
don't know 14 / 10 / 7

Montebourg is surprisingly low here, but with so small samples, it may not mean a lot.
Valls is very slightly up.

Hollande might foresee a victory in the first round, but he'd be really too confident.

I'm so eager to see the first "debate" !!!

[big bad fab]

LH2 and IFOP had 2 recent polls about probability to vote in the PS open primaries.

In a nutshell,
the older you are, the likelier you know the primaries procedures and the likelier you are to GOTV,
the higher diplomas you have, the likelier...
the higher social and professional category you belong to, the likelier...
the more Parisian or southerner you are, the likelier...
the closer to the PS you are, the likelier...

As for sex, it's less clear: roughly equal for IFOP, the likelier for men for LH2.

Of course, almost all these numbers are good for Hollande, as he is doing better among old people, Parisians and southerners, men, socialists.
Aubry should do better among Greens and among CSP++ (Hollande is better among middle classes), but these are small advantages.

Antonio may be right, after all, on the ponderation of my generic PS candidate sub-tracker.
We'll see what happens tomorrow on TV (remember it's on France 2, so it will be more watched than the next ones on i-Télé and parliamentarian channels - what is more, the second debate is at 6 PM, I think, so not a very good time).

Hollande seems to make too much meetings around France: he may well be tired for his first debate.
He seems a bit messier these days.
And he is on top of polls, so he'll be under everybody's fire, like... uh... Rick Perry

[big bad fab]

BVA poll for RTL, Orange and regional press, 7-14 September 2011, sample 2645

Out of this whole sample, you have 402 persons "certain" to vote and 676 persons who may vote (the 402 are included in the 676 from what I've understood).
Globally, out of the whole sample, you've got 1265 leftists.

So...

Among the 402 sample / among leftists in this 402 sample / among socialists in this 402 sample:
Hollande 47 / 48 / 53
Aubry 29 / 31 / 31
Royal 13 / 12 / 10
Montebourg 6 / 6 / 4
Valls 4 / 3 / 2
Baylet 1 / 0 / 0

Among the 676 sample / among leftists in this 676 sample / among socialists in this 676 sample:
Hollande 48 / 49 / 55
Aubry 28 / 29 / 29
Royal 12 / 12 / 9
Montebourg 6 / 6 / 4
Valls 5 / 4 / 3
Baylet 1 / 0 / 0

Among the 1265 leftists:
Hollande 44
Aubry 31
Royal 12
Montebourg 7
Valls 5
Baylet 1

The results are the same as for other pollsters.

Please note that BVA has erased all the "NOTA", "don't know" answers, hence the fact that Hollande is close to or above 50%.

And the internals (no tables, just comments from BVA) are very coherent with the other pollsters:
Hollande is strong among old people, inactive people, employees, socialists and has a slightly better result among those "certain" or more likely to vote.

At this stage, he may be able to pull a result "à la Cristina Kirchner" .

But, again, maybe he is too high too early. And in front of cameras, he may well be weaker.
Of course, he may not be the only one under fire (after all, Royal has to kill Aubry in order to make it to the 2nd round: Hollande is too far away for her; and Montebourg has to steal the same people that Aubry gathers for the moment); but he is the favourite and so, any other candidate will try to fight with him, and every media will first look at him.

Conversely, the "debate" of tonight is so tightly choreographed that there may be no real change (that's probably Hollande's goal). And the real debate will begin only after more than an hour, when people begin to zap on other channels...

[big bad fab]

Royal is clearly a loser here.
She was incoherent, as usual, seemed under pressure and was unable to be really different.

Valls was probably the best in the debate, but has had a bad conclusion.

Montebourg, the other one (with Royal) who should have put some flames in the debate, was too theoretical and frankly not convincing.

As for Hollande and Aubry, well, they had their good and bad moments. She was better in the middle of the debate, but he was better at the end.
I think this is a draw for both of them.

And, yes, she attacked, but I don't know if those who will GOTV in the primary want to see attacks: Hollande has done well not to counter-attack so aggressively.
Conversely, he is clearly weak when attacked personally: maybe she'll find a cleverer way to attack him in the next debates and, especially, if there is a second round.
I think that, on this point, she did bad in this debate (I mean in a primary context), but that she also proved she'd be better against Sarkozy in a TV debate.

What has shocked me is the way Pujadas was utterly biased towards Aubry:
she hasn't had to reply on DSK because it came too late and had not enough time,
Pujadas ahas attacked Hollande on the 2005 referendum,
but he didn't even tried ti have Aubry answer on the retirement at 60 years old, on which she lied and lied gain one year ago to people who were in the streets (now, she says: "back to 60 but... not for everybody" ! come on !).

Anyway, probably no great effect for this debate.
Except indirectly if Royal goes even more down (though I think the 10-15% that are still supporting her would switch to Montebourg more than Aubry).

Hollande hasn't lost ground, which was his priority.
But Aubry may find a way to hurt him, which lets her some ground to come back.

The next debate will be at 6 PM on i-Télé.
But I don't know for the last one: time and media will of course be very important.
As many people may not bother to watch the other 2, as this one wasn't completely original and fascinating...

[big bad fab]

(the last debate will be on BFM-TV, more and more a "trash" TV, at 8.30 PM: so, let's expect blood and maybe a not-so-tiny influence on the final result, though it's on a TV less watched than the big old ones)

So, after having reviewed the debate:

as for the introduction:

1. Valls
2. Hollande
3. Baylet
4. Montebourg
(the last 2 did really badly)

5. Royal
6. Aubry (yeah, she started awfully, a lot under pressure)

as for the interview part:

1. Montebourg (brilliant, clear, energetic)
2. Valls (clear, convincing)
3. Hollande (not so bad: he appeared stronger than usual and with emotion and anger)
(these 3 were clearly better)

4. Baylet
5. Aubry (well, it was better in the end, but it was messy, unclear and utterly un-smiling)
6. Royal (come on... everybody says she has "made progress" since 2007, but where on what she's really our Palin... not managing the language, completely messy, unable to manage her own sentences, very weak on financial questions, forced to look at her paper,...)

as for the debate:

1. Valls (at ease, clear)
2. Aubry (she was better as an attack dog, clearly, and Hollande vascillated too easily)
(these 2 were clearly ahead of the others)

3. Hollande (it was pretty mixed for him: good on drugs and DSK, good by seeming a uniter, not an attacker; but very quickly a lil' boy not daring to answer Big Mamma Aubry; he has clearly a worrying weakness here)
4. Baylet
5. Montebourg (disappointing, too theoretical, unable to be a focus point, he was clearly underperforming after a very good interview)
6. Royal (does she have any credibility left ?)

as for the conclusion:

1. Hollande (at ease, very consensual, very good towards old socialists and leftist voters who want unity and efficiency)
2. Aubry (good on a personal level, at last, clearly relieved it was over, a phrase "j'en ai" very finely put in place)
3. Valls (brilliant though he has hesitated, unfortunately for him, as it was personally good)
4. Montebourg (better than in the debate, but less good than Valls in the personal speech)
(the last 2 missed an opportunity)

5. Baylet
6. Royal (a conclusion that was a good resume of the rest of her TV session: she is so passé, now...)

Overall, it's a draw between Hollande and Aubry.

I was surprised Aubry seemed to try to do what the medias wanted her to do and what many French people who won't GOTV are interested in: fighting.
But, of course, she has revealed Hollande's weakness in personal fighting.
So, that's mixed.

Hollande knew clearly what to say to old voters of the PS (those who will GOTV), but he also was surprisingly weak when Aubry attacked him on budget aims and on education (paradoxically, he was better on nuclear subjects, because she wasn't as clear as she said).
So, that's mixed for him too.

Both Hollande and Aubry weren't excellent, far from it, and had Sarkozy changed his track earlier, he would have had a real chance, because he is a greater campaigner and a breater debater.... but that's too late now.

It was globally very good for Valls, but he is too on the right to be able to gain grounds.
It was globally good for Montebourg, though he was weaker in the debate.
Baylet was happy to be there, but now the PRG will only be known for one thing: legalization of cannabls What Herriot and Daladier would think !!!

And, of course, Royal has completely lost the night. But, as she is now supported only by her hardocre fans, maybe she won't lose ground in polls...

So, that's a bad thing for the right:
- Royal down,
- Aubry and Hollande training for the big rendez-vous of 2012,
- many ideas repeated all the night on subjects which many French people don't understand (a "cultural" brainwashing in favour of the left, if you prefer ),
- questions by Fabien Namias of quite good quality (Pujadas is a crap and Fressoz too, as usual) and then answers of generally good quality (except from hesitating Royal and loony Baylet)
- two promising guys (Valls and Montebourg) already trained for big media events,
- globally speaking, a unity that still stands,
- no noise from Mélenchon, Joly, Besancenot or Laguiller.

Quite a good night for the PS, really, whatever the medias are now saying on a "boring" TV session.