2022 French legislatives

[jaichind]

Harris poll has

LREM   338-378
RN        65-95
LR        35-65
FI         25-45
PS+     20-40
EELV     1-5

I assume RN will get a lot of seats due to a second-round alliance with RI?

[jaichind]

Thinking about it, one thing I'm not sure that has been mentioned is the potential role for triangulaires (or even quadrangulaires). In so far as, any candidate receiving over 12,5% of registered voters in a given constituency will qualify for the second round (iirc there are some technicalities but that is the broad jist of it). Leading to the potential, especially if turnout is reasonable, of second rounds opposing three or even more candidates in cases where they all acheive this threshold.

Or, in an election where the electorate seems to be splitting into three large and very antagonistic blocks, it is entirely possible that there are quite a large number of these. The antagonism between them therefore reducing the number of anti-RN désistements républicains (ie a non-RN candidate stepping down in the second round so as to concentrate the non-RN vote around one candidate).

In which respect, it's not always a simple question of where do RN/NUPES/LREM voters go in a second round they are absent; because all three are going to be present. And likewise, in terms of who wins where, that doesn't depend on the strength of one particular block, but in the degree of forces all have - as theoretically a fairly evenly split (lots of the South West, maybe) area would mean a much lower share of the vote being needed in order to win than in say the Paris region where there will be lots of simple left v centre-right duals; or in much of the rural North East where it will be centre-right vs far right with little left presence.

Good point.  And with the NUPES alliance would not turnout increase acrosss all voting blocs on the premise that the election might not be a preordained blowout?

[jaichind]

I assume if Nupes holds the pro-Macron bloc to below 289 seats he will do a deal with LR to get to a majority?

[jaichind]

Turnout so far looks like 2017

[jaichind]

Nationwide results: 55% counted

24.1% Ensemble
21.8% NUPES
21.7% RN
11.3% LR
  3.9% DG
  3.8% Reconquête !
  3.1% DD
  2.3% Ecologists
  1.9% Regional parties
  1.5% DC
  1.2% Sovereign parties
  3.4% Others

47.3% Turnout

Are all Macron bloc parties running under Ensemble? Similar question for NUPES as well as LR allies.  Namely are any of the vote shares of DG and DD really allies of these larger blocs?

[jaichind]

RN total vote already exceeds FN first-round total vote in 2017

[jaichind]

Tell me how this is possible

and still you end up with Ensemble having likely absolute majority (perhaps with the center right).

The congressional district system in the US, senate and electoral college are all much fairer than France's political system.

I assume this is because there will be a good deal more ENS-RN runoffs than NUP-RN runoffs.

[jaichind]

Is it reasonable to assume that the ENS-NUP runoffs would tilt ENS given that LR+ votes will lean heavily in favor of ENS?

[jaichind]

SNIP

As for NUPES' results, they're good in the current context but not particularly amazing. Mélenchon, Jadot, Roussel and Hidalgo won 30.6% in April, and the NUPES result alone is comparable to the 2017 totals, which, obviously, weren't particularly good. I think a lot of people have just gotten so used to the left being hopelessly divided and losing badly in nearly every election in the past 5+ years that the result of the united left looks more amazing than it really is - although, well, at least in terms of seats, their result will be amazing compared to what the left has become accustomed to since 2015, and particularly compared to 2017 (when the left lost seats they had no business losing and stupidly missed out on many runoffs because they all insisted on running alone), and definitely a good base from which to rebuild the left into a winning force (I'm not betting on that though).

Obviously if we get some round 2 polling or projection data in the coming days then this take will change, but right now it's highly likely that NUPES round 1 success is a mirage - precisely because their overall percentages haven't moved much since 2017 and 2022 round 1. In a lot of their runoffs I have looked at almost all potential sources of transfers are exhausted because they consolidated in round 1 to get so many candidates through. That's not a bad thing, the total left-aligned parties should more than double their seat count from 2017, but that doesn't achieve any loftier ambitions besides rescuing each faction from obscurity. Comparatively, the outcome should end up resembling 2017 in a way, just with NUPES and LR+ trading places.

The only real way this changes, barring dramatic changes in the electorate, is if NUPES can win over a comparatively significantly greater share of RN voters in their runoffs with Ensemble. Realistically we have no data on this situation beyond the polls from the presidential round 1 which had multiple hypnotical runoffs, and these are mere apples that we can try to compare to current oranges. Usually Macron was defeating Melenchon by the same amount or slightly greater than he was defeating Le Pen. Which suggests RN voters choose Macron marginally, likely with a large abstention rate similar to what Melenchon voters actually did in Round 2. Obviously there will be geographical variance to the RN electorate, abstention, and some voters of all eliminated parties will just go for non-Ensemble to "stick it to Macron." Unless there is a secret reserve of NUPES voters they will just find themselves losing a lot of runoffs as the elderly and/or wealthy Round 1 LR+ voters (with nothing better to do) show up and cast yellow ballots against the left.

I think in a bunch of seats there are various minor Left candidates that ran and won some votes so NUPES have some room to grow in the second round for some seats.  Also, NUPES's great hope for the second round has to be the mirror hope Len Pen had for the Prez second-round election.  Namely, Le Pen had to hope that the  Mélenchon vote was not a mostly Left-wing vote but an anti-system vote.  The vote share she got in the second round proved her hope was partially correct but not enough for her to win.  Now NUPES has to hope that the RN vote is not just a Right win vote but also an anti-system vote.  I suspect NUPES's hope will be partially realized and potentially deprive ENS of a majority.

[jaichind]

Interesting results. Both Ensembles and NUPES underperformed slightly, but are well within the margin of error. Particularly pitiful result for Ensemble, as they're barely ahead (according to the Interior Ministry's results - Le Monde, which uses more sensible categories, actually has NUPES slightly ahead) but their result is historically weak for a presidential party. RN didn't really overperform its polling, but the fact that it didn't underperform them is notable, and might have major implications for the runoff. The traditional right is the only force that overperformed (although even then, just by a couple points when all is said and done) and they will remain a presence in the coming legislature, which was far from a given.

The runoffs will be a mess. I've been spending the evening compiling a map of the runoff situations in each constituency, and I've had to use so many different colors that even I can't keep track. I'll have to refine it and use cruder categories to have any chance of making sense of it, but either way, that means this is still wide open. It seems there's a real chance of depriving Ensemble of an absolute majority, though, which would make the next 5 years a lot more interesting than we might have expected (in addition to being, frankly, better for democracy, since it's patently obvious today that Ensemble has no national mandate to govern). See you next Sunday, I guess.

Are you able to ahare what your seat by seat analysis concluded?

[jaichind]

The Latest Harris poll/project seems to indicate an increase in NUPES seats and a decrease in ENS seats relative to its projection election night
https://twitter.com/EuropeElects/status/1536774521453260800

[jaichind]

I know close to nothing about the candidates or seats, but I took a stab at going seat-by-seat to figure out the result by making very dumb eyeballing of the vote share of the first round.  I got

ENS          283
Ind-ENS        1
NUP          161
Ind-NUP        3
Other Left    15
LR               57
UDI              3
Ind-LR          6
RN              31
Others         17

[jaichind]

Ifop's model. Their MOEs are doing a lot of work here since it's all but guaranteed the others hit the top end of Ifop's estimate given the first round in the overseas departments.

This poll did have some breakdowns

https://www.ifop.com/wp-content/uploads/2022/06/119275-Rapport.pdf

It has

               NUP          ENS            LR             RN
NUP          96              3                               1
ENS            3             95               1             1
LR-UDI        6            46              43             6
REC            4             13               9            74
RN              5              4               2             89

But critically it does now show how RN voters would vote when faced with a NUP vs ENS runoff.  Likewise, it does not show how ENS voters would vote in a NUP vs LR runoff.  Going by their seat count for NUP which I find high, I suspect these votes would break better for NUP than one would expect.

[jaichind]

From a different pollster:
https://twitter.com/2022Elections/status/1537475956172103682?t=wRUVdkfkLEEoCYq3f...

Wow.  If this is true then I can definitely see how NUP can outperform in terms of seat count and drive ENS out of majority territory.  If true this also means that a good part of the RN vote is the anti-system vote and not a right-wing vote.

[jaichind]

At long, long last, here is is!

 

Thanks.  Did you get to do a seat-by-seat projection?  I wanted to compare what you came up with to my very dumb eyeball approach.

[jaichind]

Elabe projection seems to have LR and RN doing reasonably well

[jaichind]

And Cluster 17. Perhaps the highest numbers for the non-major pillars, I don't think the parliamentary right even have the candidates qualified and in position to win their projected seats.

I think they do if you include UDI and various pro-LR independents.  But yes, at the upper end of their projection for LR+ of 75 would mean an almost clean sweep of LR+ candidates.  Amazing projection.

[jaichind]

Gonna go against the grain of predictions here- I think it's possible that Ensemble and NUPES end up a similar amount of votes each in round 2 and therefore a similar number of seats each, given that poll suggesting RN transfers leaning towards NUPES.

Of course that poll could have been wrong, but something along the lines of ENS 240, NUP 230, RN 50, LR 40 would not surprise me.

The main problem with this reasoning is that there are a lot more ENS-RN runoffs than NUP-RN runoffs.  So even if the level of the "support" between ENS and NUP is similar ENS will have a lot more seats since RN will clearly have a much lower strike rate than all to other party blocs.  By the reverse logic, LR will most likely have the best strike rate given the fact that they will benefit from the most number of vote transfers (ENS and RN will vote for LR over each other and NUP).

[jaichind]

My final quick seat-by-seat eyeball guesses knowing nothing about the history of the candidates or seats.

ENS          269
pro-ENS        4
NUP          166
pro-NUP      12
Other Left     5
LR              60
UDI              3
pro-LR          6
RN              36
pro-RN          2
Others         13

[jaichind]

The winner in Polynésie française 1 is only 21 year old.

It seems he overcame a 40.9 vs 20.1 first-round deficit to win.  Amazing.

[jaichind]

It seems like in places like Guadeloupe and Martinique NUP ran multiple candidates and then just accept the winner into the NUP.  Is this like Japan where the LDP sometimes back several LDP candidates for a seat and retroactively nominates the winner under the rule of "if you win you are LDP"?  Is this where it is 'if you win you are NUP"?

[jaichind]

No results from Mayotte?

[jaichind]

RN 60-102
ENS 200-260
NUS 149-200

[jaichind]

Wow ... want a blowout number for RN.  It seems NUS voters must have voted RN to defeat ENS

[jaichind]

Putin 230, Non-Putin 330

Putin as PM ?