Quebec Provincial election October 1, 2018

[mgop]

trudeau is done. liberals lost everything this year: ontario, new brunswick, quebec...

[Tintrlvr]

trudeau is done. liberals lost everything this year: ontario, new brunswick, quebec...

The PLQ and the federal Liberals are not really connected in a meaningful way (beyond both having bases among Anglophone and Allophone Quebeckers), and many PLQ politicians are federal Conservatives (with a handful being federal NDPers). This result doesn't impact Trudeau at all.

[EPG]

trudeau is done. liberals lost everything this year: ontario, new brunswick, quebec...

On the contrary, it's a normal pattern in most Canadian provinces (not B.C. which has its own unique dynamic). It is the typical second-order election phenomenon where the main government gets kicked, the kind that makes you wonder whether partisan sub-national elections should really be permitted.

[Oryxslayer]

Here is a question, is there any data or reports on if the NAFTA news changed voters? I'm wondering why the polls undershot to CAQ and overshot the PLQ by so much - I suspect it's a combination of PLQ voters having poor turnout rates, and perhaps an 11th hour rural voter movement to the CAQ.

[ON Progressive]

trudeau is done. liberals lost everything this year: ontario, new brunswick, quebec...

The PLQ and the federal Liberals are not really connected in a meaningful way (beyond both having bases among Anglophone and Allophone Quebeckers), and many PLQ politicians are federal Conservatives (with a handful being federal NDPers). This result doesn't impact Trudeau at all.

Not to mention provincial elections are very different in general. In 2015, Alberta elected an NDP majority then only sent 1 federal New Democrat to Ottawa (Edmonton Strathcona).

Also, the Ontario election result is actually helping Trudeau because Doug Ford already has really bad approval ratings.

[King of Kensington]

Any data on popular vote by region?

[adma]

Solitaire has won their first non-Montreal seat, in Taschereau

Haha this is a great autocorrect.

Hey; I warned you about Taschereau. ;-)

Maybe not in Quebec City, but I'm wondering whether a motivational factor (however token) behind the QS victories in Sherbrooke and Rouyn et al was their being represented by the NPD federally.

Oh, and Ungava wins the Sask-Humboldt '04/Inverness Skye & Lochaber '92 award for being won with barely a quarter of the vote...

[Lachi]

trudeau is done. liberals lost everything this year: ontario, new brunswick, quebec...

If brugh was here, he'd have already told you this, but provincial politics and federal politics, ESPECIALLY in Quebec are so far disconnected from each other.

[Hatman 🍁]

Solitaire has won their first non-Montreal seat, in Taschereau

Haha this is a great autocorrect.

Hey; I warned you about Taschereau. ;-)

Maybe not in Quebec City, but I'm wondering whether a motivational factor (however token) behind the QS victories in Sherbrooke and Rouyn et al was their being represented by the NPD federally.

Oh, and Ungava wins the Sask-Humboldt '04/Inverness Skye & Lochaber '92 award for being won with barely a quarter of the vote...

Well, it helps to know that it IS POSSIBLE for a left wing party to win. Sherbrooke being a college town helps. Rouyn-Noranda is a bit more interesting, considering it is a working class/mining city, much like a typical city in Northern Ontario. QS's brand of being bobo leftists is not a good fit on paper for a riding like that, but I suppose their economic arguments resonated with them.

[JG]

Solitaire has won their first non-Montreal seat, in Taschereau

Haha this is a great autocorrect.

Hey; I warned you about Taschereau. ;-)

Maybe not in Quebec City, but I'm wondering whether a motivational factor (however token) behind the QS victories in Sherbrooke and Rouyn et al was their being represented by the NPD federally.

Oh, and Ungava wins the Sask-Humboldt '04/Inverness Skye & Lochaber '92 award for being won with barely a quarter of the vote...

Well, it helps to know that it IS POSSIBLE for a left wing party to win. Sherbrooke being a college town helps. Rouyn-Noranda is a bit more interesting, considering it is a working class/mining city, much like a typical city in Northern Ontario. QS's brand of being bobo leftists is not a good fit on paper for a riding like that, but I suppose their economic arguments resonated with them.

Rouyn-Noranda also has an university (Université du Québec en Abitibi) and, according to Radio-Canada, QS had a huge get out the vote operation around that area.

[Hatman 🍁]

https://twitter.com/robroc/status/1047500391791906819

This might explain why the polls were off. Clearly a huge drop off of Anglo voters. I guess without the separatist boogeyman out there, they stayed home. This tells me they were only voting Liberal because they had to, and now since they figured they didn't have to, they decided to stay home, not wanting to endorse a party they may not really align with, but were voting for anyway.  

Polls didn't account for this in weighting, so Anglos (and therefore the Liberals) were being weighted up more than they should've been.

[DC Al Fine]

https://twitter.com/robroc/status/1047500391791906819

This might explain why the polls were off. Clearly a huge drop off of Anglo voters. I guess without the separatist boogeyman out there, they stayed home. This tells me they were only voting Liberal because they had to, and now since they figured they didn't have to, they decided to stay home, not wanting to endorse a party they may not really align with, but were voting for anyway.  

Polls didn't account for this in weighting, so Anglos (and therefore the Liberals) were being weighted up more than they should've been.

Ah, that would explain why West Montreal's results resembled Calgary instead of North Korea At first I thought Anglos were moving to QS, but that doesn't seem to be the case.

[Oryxslayer]

Turnout clearly wasn't in the PLQ's favor this election.

[mileslunn]

Any data on popular vote by region?

I'll do more calculations tomorrow, but I've done some calculations in a few regions

Lower Saint-Lawrence/Gaspé/Magdalen Islands: PQ 34.6%, CAQ 28.7%, PLQ 23.2%, QS 12.3%

Saguenay–Lac-Saint-Jean/Côte-Nord: CAQ 35.2%, PQ 31.7%, PLQ 20%, QS 11.6%

Capitale-Nationale: CAQ 44.6%, PLQ 22.6%, QS 17.3%, PQ 12%

Mauricie: CAQ 46.5%, PLQ 22.6%, QS 14.7%, PQ 14.3%

Estrie: CAQ 37.1%, PLQ 22.9%, QS 22.7%, PQ 15.3%

What was Quebec minus island of Montreal, that would be interesting to see the difference here as I am guessing PLQ would have been in the teens CAQ over 40 percent.

[ON Progressive]

Any data on popular vote by region?

I'll do more calculations tomorrow, but I've done some calculations in a few regions

Lower Saint-Lawrence/Gaspé/Magdalen Islands: PQ 34.6%, CAQ 28.7%, PLQ 23.2%, QS 12.3%

Saguenay–Lac-Saint-Jean/Côte-Nord: CAQ 35.2%, PQ 31.7%, PLQ 20%, QS 11.6%

Capitale-Nationale: CAQ 44.6%, PLQ 22.6%, QS 17.3%, PQ 12%

Mauricie: CAQ 46.5%, PLQ 22.6%, QS 14.7%, PQ 14.3%

Estrie: CAQ 37.1%, PLQ 22.9%, QS 22.7%, PQ 15.3%

What was Quebec minus island of Montreal, that would be interesting to see the difference here as I am guessing PLQ would have been in the teens CAQ over 40 percent.

I'm not quite done my calculations yet. I plan on finishing them by tomorrow at least, then I can do that.

[Poirot]

There will probably be a recount in Gaspé riding. PLQ won by 132 votes.

There might have been a transcript error. Ballot box 61 from the town of Gaspé has the result of 194 votes for the PLQ and none for the other three candidates.  

[ON Progressive]

It does seem like there are small differences between the CBC and Elections Quebec numbers in some ridings. They aren't big for the most part, but could mess up my calculations at the margins if this is happening throughout the province. There is one that was pretty big (PQ has 9238 votes in Lac-Saint-Jean on CBC but 9138 on Elections Quebec).

I think I'll just go with Elections Quebec numbers, given the CBC doesn't display some third party candidates for some odd reason.

[Polkergeist]

I was always interested how CBC or other Canadian TV networks make calls on the overall outcome of the election.

On Monday a CAQ govt was projected 15-20 mins after poll close, a majority projected after 35 mins. While it was not close, this still seems pretty quick.

No doubt they have complex models, but in Australia elections are not called unless one party leads (after adjusting for matched booths) in a majority of seats where the vote is reported.

In Canada they seem to project winners with very little of the vote counted. Exit polls also don't seem to be part of the set-up either.

Does anyone have any insight on this?

[mileslunn]

I was always interested how CBC or other Canadian TV networks make calls on the overall outcome of the election.

On Monday a CAQ govt was projected 15-20 mins after poll close, a majority projected after 35 mins. While it was not close, this still seems pretty quick.

No doubt they have complex models, but in Australia elections are not called unless one party leads (after adjusting for matched booths) in a majority of seats where the vote is reported.

In Canada they seem to project winners with very little of the vote counted. Exit polls also don't seem to be part of the set-up either.

Does anyone have any insight on this?

Australia uses ranked ballots so winning the first round doesn't mean one will win thus why it takes longer whereas Canada uses FTFP. They can call them quickly since as soon as all ridings are reporting a few polls you can see trends. Also based on past results and pre-election polling, they look at key ridings and if the party is winning in the ridings they need to in order to win they can call it.

[RogueBeaver]

https://twitter.com/CLecavalierJDQ/status/1047865464498655232

[Tintrlvr]

I was always interested how CBC or other Canadian TV networks make calls on the overall outcome of the election.

On Monday a CAQ govt was projected 15-20 mins after poll close, a majority projected after 35 mins. While it was not close, this still seems pretty quick.

No doubt they have complex models, but in Australia elections are not called unless one party leads (after adjusting for matched booths) in a majority of seats where the vote is reported.

In Canada they seem to project winners with very little of the vote counted. Exit polls also don't seem to be part of the set-up either.

Does anyone have any insight on this?

Australia uses ranked ballots so winning the first round doesn't mean one will win thus why it takes longer whereas Canada uses FTFP. They can call them quickly since as soon as all ridings are reporting a few polls you can see trends. Also based on past results and pre-election polling, they look at key ridings and if the party is winning in the ridings they need to in order to win they can call it.

Also, if it was a close election, it would take much longer to call. But this election was not close. I imagine elections like NSW 2011 were called quickly after polls closed, too.

[RogueBeaver]

Grits have chosen Pierre Arcand as interim leader. Couillard and Legault held their first transition meeting today, but no date yet. Previous reports said it would be after Legault returns from leading our Francophonie summit next week.

[Polkergeist]

Thanks for that.

In Australia, while the distribution of preferences in a ranked ballot are a source of uncertainty, they are not too big of a source, given preference flows are pretty stable.

I don't think they would hold up an election call unless it was deemed a tight race on the basis of only primary votes and preference assumptions.

I was always interested how CBC or other Canadian TV networks make calls on the overall outcome of the election.

On Monday a CAQ govt was projected 15-20 mins after poll close, a majority projected after 35 mins. While it was not close, this still seems pretty quick.

No doubt they have complex models, but in Australia elections are not called unless one party leads (after adjusting for matched booths) in a majority of seats where the vote is reported.

In Canada they seem to project winners with very little of the vote counted. Exit polls also don't seem to be part of the set-up either.

Does anyone have any insight on this?

Australia uses ranked ballots so winning the first round doesn't mean one will win thus why it takes longer whereas Canada uses FTFP. They can call them quickly since as soon as all ridings are reporting a few polls you can see trends. Also based on past results and pre-election polling, they look at key ridings and if the party is winning in the ridings they need to in order to win they can call it.

Also, if it was a close election, it would take much longer to call. But this election was not close. I imagine elections like NSW 2011 were called quickly after polls closed, too.

[RogueBeaver]

Leger's postmortem: 47% of non-Francophones didn't vote, only 12% of Francophones voted Grit (both probably worse than '48 IMO, but in a 2-party system with much more rigid partisanship), and CAQ had a 5% surge over the final weekend.

[MaxQue]

Finally (?), the PQ will have 10 seats, as they gained Gaspé in a recount (went from Liberal +132 to PQ +41). Seems like a big shift, but there was various irregularities, like the infamous Box 61, who reported all its votes as Liberal, which was improbable.

Recount in Îles-de-la-Madeleine went from PQ+18 to PQ+15.

PQ asked a recount in Ungava (CAQ+44), the judge didn't rule yet on whether it will happen.

Standings are now:
CAQ 74
PLQ 29 (-3)
PQ 10 (+1)
QS 10
Ind 1 (+1, Guy Ouellette (Chomedey) having been expelled from Liberals for leaking information to PQ and CAQ)
Vacant 1 (+1, Couillard)