Tuesday, 14 May 2019

About the Newfoundland and Labrador Election...

Dear 338Canada readers,

I wanted to let readers of this blog know why I am not covering the Newfoundland and Labrador election this week. One thing I can tell you all is that it is not by lack of interest.

I am a college teacher and the winter semester ends next week, so I am in the midst of final projects, final exams and department meetings before the summer break. On top of this, my new gig at Maclean's and my work L'actualité are eating up whatever free time I have left. For the past three months, it felt as if I have been working two full time jobs.

And I'm exhausted. The Alberta election was great to follow, but it almost burned me out. I want to stay sane and keep loving this work, so I have to manage where I concentrate my efforts, especially with the federal election coming this fall.

Moreover, there have been very little data on voting intentions of Newfoundlanders - even less so than PEI. The 338 model doesn't work like a crystal ball or tea leaves; it needs real, scientific data to produce reliable projections. And the data is just not there, so rather than making up numbers and hoping to get lucky (which would be dishonest), I have decided to sit this one out.

Thank you for your support and understanding,

Philippe J. Fournier
338Canada.com




Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Sunday, 12 May 2019

If Electoral Reform Had Happened...

Proponents of electoral reform couldn't help but reiterate their displeasure with the current First-Past-the-Post (FPTP) system when commenting on last week's 338 seat projection, and the numbers certainly suggested they may have a point. The Conservative Party of Canada, then with an average support of 36.6% of Canadian voters according to the latest poll aggregate, was projected to win an average of 174 seats - just above the 170 seat threshold for a majority at the House of Commons.

* * *

Here's an interesting exercise: what would the seat projection look like with current numbers if Canada had moved to a proportional representation (PR) system?

Read this article on Maclean's magazine website.






Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

338 Federal Projection Update: May 12th 2019

We add three new polls to the 338 federal electoral model this week:
  • Abacus Data measures the Conservatives with a three points lead over the Liberals;
  • The Nanos weekly tracker also has the Conservatives ahead by 3 points nationally and the Greens are polling in double digits (with 15% in Atlantic provinces and 21% in B.C.); 
  • Campaign Research has a four point lead for the CPC four points ahead of the LPC, but with the Liberals slightly ahead in Ontario;

Here is the 338 Federal Update for May 12th 2019. The full list of polls can be found on this page.


Popular Vote Projection

The Conservatives lead the way again this week with an average lead of 5 about points ahead of the Liberals.






Seat Projection

The Conservatives are still projected ahead in the seat projection with 163 seats on average, but the confidence intervals overlap significantly with those of the Liberals. The LPC's average is at 124 seats.



The complete regional breakdown per party can be found on these pages:




Odds of Winning the Most Seats

According to this week's numbers, the Conservatives win the most seats in three quarters of all 2590k simulations.




Regional Distribution


The map of the 338 Projection has been updated and is available on this page. Use this list to find your federal electoral district:

Have a great week!




Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Tuesday, 7 May 2019

Greens Win Nanaimo-Ladysmith By-Election and Justin Trudeau is in Trouble

... and Jagmeet Singh too, but we already knew that.

Let's cut right to the chase: at some point we will have to realize that the Green Party of Canada may be a threat to its rivals in the general election coming next fall.

Last night's by-election win in Nanaimo-Ladysmith for the Greens did not necessarily come as a surprise, as the latest 338Canada projection had the Greens projected in first place, but Nanaimo-Ladysmith was forecasted as a toss-up - and last night's results for the Greens shows they were significantly underestimated in the model.

Consider the following graph. The coloured bars show the latest projection in Nanaimo-Ladysmith for the main parties and the black dots, the actual by-election results as of this morning.



First, let's start with the NDP and Conservatives: the results for both parties lie within the model's confidence intervals, but the NDP was projected higher than the CPC. The Conservatives were thus modestly underestimated and the NDP, overestimated.

This is awful news for the NDP, which not only loses the district it had won easily in 2016, but loses ground once again in a by-election. Four years ago, NDP MP Sheila Malcolmson had won Nanaimo-Ladysmith by a significant margin of seven thousand votes. Last night, the NDP was almost 6000 votes behind the Greens (and in third place behind the Conservatives).

How difficult has it been for the NDP since the 2015 general election? There have been in total 18 by-elections in the current 42nd legislature. The NDP has managed to lose ground, compared to its 2015 result, in 16 of them. [The exceptions were Ottawa-Vanier (which was easily won by the Liberals) and Burnaby-South, where NDP leader Jagmeet Singh was elected last February.]

We have seen in recent polling the Green Party of Canada and the NDP in a statistical tie in parts of the country. Last night's result was plausibly not just an isolated bump-on-the-road incident for the NDP.

One wonders how much patience NDPers will have with Mr. Singh after October 21st should bad results continue to pile up.


Secondly, let's take a look at the Greens and Liberals. Green candidate Paul Manly won last night's contest with a little over 37% of the popular vote, a result which lies just above the 338 projection's superior confidence interval of 35.6% (see graph above). In terms of actual votes, Mr. Manly received a little over fifteen thousand votes, about a thousand more than in the 2015 general election - which means the Green Party has actually increased its pool of voters in the region.

And for the Liberals? To be fair, the LPC was not expected to win Nanaimo-Ladysmith - the 338 projection had the Liberals in fourth place and winning the district in less than 1% of simulations. Nevertheless, this dismal result (only 4,700 votes and 11% compared to 16,700 votes and 23.5% in 2015) cannot be good news for Justin Trudeau.


What should we remember from last night's by-election?

Many observers (not excluding myself) tend to give way too much importance to by-elections, when in fact they generally are poor barometers of the mood of the national electorate. Hyperlocal effects, the most recent news cycle, and local candidates all tend to add more noise to the system, which impedes us to see it for what it truly is: a local election.

However, we already knew, from months of polling and repetitive poor by-election results, that the NDP had been losing ground among voters - even in what we perceived as highly favorable NDP districts. However, it is no doubt the Liberals who come out of last night's as the biggest losers, for it was not just a regular loss: the Liberals could not take advantage of a weak NDP, and let a small party - whose platform treads on the very ground the Liberals have tried to re-brand themselves with -  win and take what will be perceived as "momentum".

Indeed, it appears Liberals strategists have tried to use new and fresh environmental policies to contrast themselves from Conservatives - seeking, perhaps rightly so, that the environment was the CPC's Achilles' heel.

They surely had not expected the Green Party to, potentially at least, spoil their plan. The Liberals will probably have to find a new ballot box issue before next fall.





Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Monday, 6 May 2019

Discussion with Todd van der Heyden about the Green Democrats

Here is an interview with the great Todd van der Heyden of CTV about the hypothetical NDP-Green merger column I wrote last week. Interview starts at 19:35:








Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Let's Compare the 338 Projection Averages with the Latest Polls

The 338 federal electoral projection was published yesterday in Maclean's and it shows the Conservatives with a six-and-a-half point lead over the Liberals in the popular vote, but a substantial lead in the seat projection (averages of 174 for the CPC and 111 for the LPC). See all the numbers and full map of the projection here.

Several comments I received last night were suggesting the latest Angus Reid poll, which measured the CPC with a 13 point lead, may have skewed the projection too much in favour of the Conservatives.

It is a legitimate concern that one poll (possibly an outlier) could skew the 338 numbers and therefore wrongly influence the week's narrative, but one that rarely applies here because the model actually detects outliers when it sees one. A poll that strays too far from the current averages will see its weight automatically reduced in the calculations. However, if said poll is subsequently confirmed by another polling firm, it regains the weight it was supposed to have in the first place.

And not one, but two polling firms measured the Conservatives with a 13 point lead in recent days: the aformentioned Angus Reid Institute poll and the latest Leger/Canadian Press poll (Léger had the CPC at 40% nationally).

In fact, let's compare the five latest polls which were considered for yesterday's 338 update with the projection averages:


You may notice that all dots but three lie within the confidence intervals (the exceptions would be Angus Reid LPC at 25%, Leger NDP at 12% and Ipsos NDP at 19%).

I believe this graph above clearly shows the 338 numbers are close to the current averages. I hope this answers the concerns that were directed at me last night.





Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Sunday, 5 May 2019

338 Federal Projection Update: Tories in Majority Territory

Since Ottawa was hit by the SNC-Lavalin/Jody Wilson-Raybould storm in early February, there have been a staggering amount of opinion polls conducted and published in the media. Data nerds like myself are surely not going to complain, but so many polls taken when public opinion is in flux creates a fair amount of noise - which may lead to considerable confusion for voters.

Read this article on Maclean's website.





Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Friday, 3 May 2019

Angus Reid Institute: Liberals Fall, Greens Climb, CPC Holds Commanding Lead

A new poll from the Angus Reid Institute was published this morning, and it is surely going to move the needle on Sunday's 338 projection.

Some of its numbers are slightly outside the current margins, but they all converge towards the same trends: the left of centre coalition is splintering. And we know what scenario it usually leads to.

The poll was on the field from April 26th to 30th 2019, and was conducted from an internet panel of 1,535 Canadian voters.

Here are the results of leaning and decided voters:



The Conservatives lead the way with 38%, which is not unsurprising for those who have been following the numbers in the past weeks. However, it is the Liberal support that strikes the eye: 25%. This is a significant drop for Justin Trudeau. Earlier this week, a Léger poll for the Canadian Press had the Liberals at 27% and it looked at first as an outlier. It looks like it wasn't.

According to this poll, the LPC has dropped significantly in the Atlantic Provinces (although we should be wary of the small subsample size), Quebec and British Columbia. Under no scenario, with these numbers, could the Liberals hold on to power. This indeed would translate into a Conservative majority.

Notice also the Green Party of Canada at 11% nationally. The regional breakdown of the poll shows the Greens in double digits in every region of the country, except Alberta. Obviously, this is not yet a "Green wave", but the support for the GPC has been steadily climbing in the projection in the past months. We will have to keep an eye out for those numbers.

A full 338 Federal Projection update will be out Sunday afternoon. Stay tuned.




Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Monday, 29 April 2019

The Green Democrats?

Here is an interesting exercise of politics-fiction. What if the NDP and Green Party decided that they have more in common then have differences? With the recent success of the PC and Wildrose merger Alberta, would it be so unreasonable to imagine what a Green Party-NDP merger could look like?

Let's call them the Green Democrats.

Read this article on Maclean's magazine website.





Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Sunday, 28 April 2019

338 Federal Projection Update, April 28th 2019

First, please allow me some personal news: I have received an offer to join Maclean's magazine for the coverage of the federal election. I am over the moon.

What does it mean for this blog? Well, nothing changes. I will continue to update the federal projection every Sunday. The 338Canada website remains independent, but in addition to my columns in French in L'actualité magazine (which you can find here), I will also provide occasional content in English for Maclean's.

And I have you to thank, dear readers of this blog. Your support is very much appreciated.

* * *

We add three new polls to the 338 federal electoral model this week:
  • Innovative Research (Ontario only) shows the Liberals and Conservatives in a statistical tie, with the Grits slightly ahead in Ontario;
  • The Nanos weekly tracker has the Conservatives ahead by 3 points nationally (with the Liberals and CPC tied in Ontario); 
  • Ipsos/Global News measures the CPC four points ahead of the LPC, but that lead was 10 points a month ago, so the race has tightened up significantly according to Ipsos;


Here is the 338 Federal Update for April 28th 2019. The full list of polls can be found on this page.


Popular Vote Projection

The polls published this week showed numbers that were already in line with last week's projection, so this is very little movement in this week's numbers:






Seat Projection

The Conservatives are still projected ahead in the seat projection, but the confidence intervals overlap significantly with those of the Liberals:



Here are the seat projection probability densities for the main parties:










The complete regional breakdown per party can be found on these pages:




Odds of Winning the Most Seats

The tightening of the race is most apparent in the odds to win the most seats. According to this week's numbers, the Conservatives are merely 2-to-1 favourites to win the election.




Regional Distribution


The map of the 338 Projection has been updated. Click on the image to visit the map's page.




Use this list to find your federal electoral district:

Have a great week.




Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Tuesday, 23 April 2019

PEI 2019: Comparison of Seat Projection With Election Results

As of 22h00 eastern time, here is a quick comparison of the PEI election results with the vote and seat projections from earlier this evening (when the Leafs were still in the playoffs).

Popular vote wise, the pollsters underestimated the PC vote and overestimated the Greens, but the results are still within the model's confidence intervals:





Seat wise, with the Green overestimated in the polls, they lost a few close races that they could have won with a few more points on the Island. Assuming the Greens with the by-election in Charlottetown-Hillsborough Park (where the Greens are favoured), here is the seat projection compared to the 9 seat result:




Since the PC won a greater share of the vote than expected, so the PC seat total is just over its seat projection:



Finally, the Liberals won 6 seats. Its seat projection average was 4.




Since I strongly doubt the Liberals will want to prop up the Greens, we are in for a PC led minority government in PEI. On average, minority governments in Canada last about 18 months. We'll see whether the 66th PEI Legislative Assembly can manage to last that long.




Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Last Minute P.E.I. Update

There was a last minute PEI poll night published last night by Forum Research, which naturally had not been included in the projection I wrote for Maclean's.

I just wanted to update this projection before the night was out, but also wanted to wait - out of respect for PEI voters - until the polls closed to publish it.


Popular Vote Projection







Seat projection















Thank you and have a great evening!




Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.

Sunday, 21 April 2019

338 Federal Projection Update: CPC 161, LPC 136, NDP 24, BQ 12

The Alberta general election is now behind us. The final results (including all advance ballots) were published Friday. A full post-mortem will be published in the next few weeks, but at first glance the 338 electoral model performed generally well, even tough many polls missed the mark somewhat by underestimating Jason Kenney's United Conservatives. Nevertheless, the model correctly identified the winner in 82 of 87 districts, and three more were within the confidence intervals.

Many readers have asked whether there will be a 338 P.E.I model, the answer is this: a P.E.I projection from 338 can be found here. The PEI election will be held this Tuesday, April 23rd. Unfortunately, the timing of the PEI election made it impossible to have a regular tracker as was the case for Alberta, Quebec and Ontario. Moreover, data on the PEI election is scarce, so a full model would certainly be riddled with uncertainty.


As per every Sunday, we update the federal projections for all 338 districts. Three new polls were published in the past seven days: Forum Research, Innovative Research and Nanos.



The full list of polls can be found on this page.

Here is the 338 Federal Update for April 14th 2019.



Popular Vote Projection


Two of three polls published last week show diverging numbers to say the least. Forum Research measured the Conservatives with a 13 point lead. Innovative Research had the Liberals leading by 4 points. Which ones is closer to the mark? Innovative's numbers have leaned towards the LPC since the start of 2019, but by a much smaller gap than Forum's tilt towards the CPC (Forum has had the CPC leading in all of 2018, sometimes by double digits. Between February and March, in the middle ofthe SNC-Lavallin affair, Forum showed the Liberals gaining ground on the Conservatives, again going against the tide compared to other pollsters).

Bias? Incompetence? Propaganda?

No. Sometimes polls disagree, and that's why you need sites like this one.

Here are this week's averages. The Conservative Party of Canada are still leading the Liberals by an average of three points. Its average stands at 35.7%.



Here are the popular vote projections with 95% confidence intervals. The tightening has been incremental, but the LPC and CPC confidence intervals now overlap significantly.





Seat Projection


The Conservative Party of Canada wins an average of 161 seats per simulation, short of the 170 seat threshold in about two thirds of simulations.


The Conservatives trail the Liberals in the Atlantic provinces and Quebec. There is a statistical tie between the CPC and LPC in Ontario. West of Ontario, Conservatives lead voting intentions in every province, although British Columbia is still a very close race.

The Liberal Party of Canada wins an average of 136 seats. Most of those seats are located in the Atlantic provinces, Quebec and Ontario. The Liberals are also projected ahead in Winnipeg and parts of the Lower Mainland in BC.

Here are the seat projection probability densities for the CPC and LPC:



The New Democratic Party is currently projected at an average of 24 seats. It is virtually shut out of the Atlantic provinces, has minimal support in Quebec (where it won 16 seats in 2015), in the Prairies and in Alberta. Most projected NDP seats are found in Ontario (notably in Hamilton, Windsor and Northern Ontario) and BC (mostly Vancouver and Vancouver Island).



In Quebec, the Bloc québécois is holding steady with projected seats mostly in the eastern tip of Montreal and in the 450 area (suburbs of Montreal).





The complete regional breakdown per party can be found on these pages:




Odds of Winning the Most Seats


With this level of support, the Conservatives win the most seats in about seven of every ten simulations (or 69.1%). The odds of a Conservative majority currently stands at 35.8%.



The Liberals win the most seats in the remaining three out of ten simulations (30.1%). The CPC and LCP are tied for the most seats in 0.8% of simulations.

Regional Distribution


The map of the 338 Projection has been updated. Click on the image to visit the map's page.




Use this list to find your federal electoral district:

Have a great week, and thank you for supporting this page.




Philippe J. Fournier is the creator of Qc125 and 338Canada. He teaches physics and astronomy at Cégep de Saint-Laurent in Montreal. For information or media request, please write to info@Qc125.com.


Philippe J. Fournier est le créateur de Qc125 et 338Canada. Il est professeur de physique et d'astronomie au Cégep de Saint-Laurent à Montréal. Pour toute information ou pour une demande d'entrevue médiatique, écrivez à info@Qc125.com.