Sunday, 17 February 2019

338 Projection Update: February 17th 2019

The details of this week's 338 Electoral Projection are published on L'actualité magazine's website at this address. I invite you to visit the page and support this project. If you do not read French, online translators like Google Translate do a surprisingly decent job (I've tried it myself and it's not half bad!).


Here are the seat projections per party:



The complete map of this projection has been updated. Click here to visit the interactive map.



All 338 district projections are now available on 338Canada.com. To find your home district, use this list:



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.

Saturday, 16 February 2019

Campaign Research Poll: Conservatives Ahead Nationally, BQ surges in Quebec

The polling firm Campaign Research was last on the field in December and it measured a two point lead for the Conservatives over the Liberals (essentially a statistical tie if we consider the poll's uncertainty).

Last night, Campaign published fresh numbers that show the Conservatives taking an outright lead in voting intentions in Canada. Campaign was on the field February 7th to 11th, meaning it was polling Canadians as the Globe & Mail story about Justin Trudeau, Jody Wilson-Raybould and SNC-Lavalin was unfolding.

Here are the results:



The Conservative Party of Canada leads the pack with 37% of support among the poll's 1 590 voter sample. It is the highest score for the CPC in a non-Forum Research poll in months.

Also of note in this poll are the Quebec numbers. In the past weeks, the polling firms Léger, Innovative Reseach, Nanos and now Campaign Research all measured an increase of support for the Bloc québécois in Quebec. For the better part of 2018, the BQ was polling in the low teens and its seat projection hovered around 3-8 seats at best. The arrival of new Bloc leader Yves-François Blanchet may have revived the hopes of the pro-independence party.

Here are Campaign's latest Quebec numbers:


These new numbers will be added to the list of polls included in the 338 model. A complete new electoral projection will be published tomorrow, Sunday February 17th. Stay tuned.

Enjoy the weekend!


* * *

Here is the report from Campaign Research.



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, 10 February 2019

338 Projection Update: February 10th 2019

New polls were added to the 338 Projection this week: the Nanos weekly tracker and the Léger/Journal de Montréal Québec poll.

Here are all the national polls from the last twelve months. The complete list is available here.




Popular Vote Projection


Again this week, we have a statistical tie between the Liberals and Conservatives on top of voting intentions:


Here are the popular vote intentions with 95% confidence interval brackets:




Seat Projection


We still have a significant overlap between the Liberals and Conservatives for the seat projections. The threshold for a majority is 170 seats.



Here are the seat projections per party:




Odds of Winning the Most Seats


The Liberals win the most seats in two thirds of all 100k simulations. The Conservatives win the remaining third.



Regional distribution


The complete map of this projection has been updated. Click here to visit the interactive map.



All 338 district projections are now available on 338Canada.com. To find your home district, use this list:



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, 3 February 2019

338 Projection Update: Return to a Two-Party System in Canada?

New polls were published in the past seven days: the Nanos weekly update and the federal Mainstreet Research Ultrapoll. Both polls showed similar tendencies: 1) the Liberals still stand slighly ahead of the Conservatives, but are losing ground in the West, 2) the NDP are near a complete collapse, 3) The Bloc québécois has not (yet) enjoyed a boost from the nomination of its new leader Yves-François Blanchet.

All federal polls are listed on this page. Methodology for this projection follows the same basic rules as did the Qc125 model in Québec and Ontario*: the model calculates a weighted average of all polls publicly published by professional firms. The weighting depends on field date, sample size and historical performance of the polling firm on similar polling endeavours (i.e. provincial polls are not the same as federal polls or municipal polls, etc.). Moreover, the model takes into account the electoral history of all regions of Canada and the demographic data published by the Canadian census.

(*In 2018, the Qc125 model correctly predicted the PC and CAQ majorities in Ontario and Québec respectively, and correctly identified the winners in 111/124 ridings in Ontario and 112/125 ridings in Québec.)

As the last parliamentary session of the 42nd federal legislature gets under way next week, here are where the main parties stand less than 8 months away from the general election.



Popular Vote Projection


Although every poll published since mid-December 2018 shows the Liberals in the lead nation-wide, the gap between the Liberals and the Conservatives is still too narrow to be considered statistically significant. Indeed, as we will see below, we have a statistical tie on top of voting intentions.

The Liberal Party of Canada leads the way with an average of 35.9%. As we will see below, the Liberals are leading in the Maritimes, Québec and are neck and neck with the Conservatives in Ontario. The Liberals seem to have lost ground lately in BC, where it could cost them some crucial seats.

The Conservative Party of Canada currently stands at an average support of 34.0%. According to current numbers, the Conservatives could win as many as 40 more seats than in 2015. Those net gains could potentially come from New Brunswick, Ontario, and all of the western provinces.

For most of 2018, the New Democratic Party polled, at best, in the low 20s and in the high teens, but in the past few months, its score has fallen closer to 15% or below. This week, the NDP's average stands at 13.0%. On February 25th, all eyes will turn towards the Vancouver district of Burnaby South where NDP Jagmeet Singh hopes to get elected to the House of Commons. Should he failed to do so, it could have national consequences.

Here are the popular vote projections with 95% confidence intervals:






Seat Projection


The Liberal Party of Canada, which won 184 seats in 2015, wins a average of 174 seats over one hundred thousand simulations performed by the 338 model. The threshold for a majority is 170 seats, so a Liberal majority - or even a Liberal win for that matter - would be far from certain should those numbers hold until the fall.


The Conservative Party of Canada climbs again this week in the seat projection, this time with an average seat count of 143. As it stands right now, the CPC would be the only party in position to make net gains compared to the 2015 general election (with, to a much lesser extent, the Green Party of Canada).

Consider the following graphs. It depicts the probability densities of seat projections for all major parties with a comparison with their 2015 seat totals.

For the LPC and CPC, the probability functions overlap significantly, meaning that although the Liberals still hold a slight edge, a Conservative (minority) win would still be a plausible scenario:


Here is the NDP seat probability density. Obviously, it falls way short of its 44 seat total of 2015:


For the Bloc québécois, I would urge to use caution with these numbers. As mentioned above, the BQ has a new leader since mid-January, and it would be a reasonable assumption that his arrival has not yet made the needle move for the pro-independence party.


The Green Party of Canada is projection to make modest gains, especially in Vancouver Island in BC:







Projection of outcome


With these numbers, the Liberals win the most seats in about three quarters of all simulations. This graphs shows the probability density of the seat difference between the LPC and the CPC.





Regional Distribution


All 338 district projections are now available on 338Canada.com. To find your home district, use this list:


The complete map of this projection is available on this page.



New polls are expected later this week. 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.

Saturday, 26 January 2019

Alberta 87

Albertan voters are scheduled to head to the polls in the coming spring. Rachel Notley, Alberta Premier since 2015, has a steep hill to climb in order to win reelection: not only does she have a record to defend, but she now has to face a united conservative challenger.

Since the election is still months away, data on voting intentions of Albertans is rather scarce, but polls conducted in the last six months or so all showed the same basic tendencies: 1) the NDP has lost some ground since the 2015 election (it is polling somewhere between the low 30s and high 20s), and 2) the UCP has, at least so far, amassed most of the PC and Wildrose support under one banner.

Here are the polls:




By breaking down polls by regions and taking into account the electoral history and demographics of Alberta voters, the 338Canada electoral model calculates voting intentions and seat projections - with uncertainty* - province wide.

[*The term "margin of error" is most commonly used in colloquial speech, but it has a very specific definition in the field of statistics, especially with polls using random samples. The term "uncertainty" is generally more appropriate for these types of projections.]

Taking into account the data currently available, here is the 338Canada Alberta Popular Vote Projection for January 26th 2019:



Unsurprisingly, the United Conservative Party leads the way with an average projection of 51% of support. Considering that, back in 2015, the Wildrose and PC parties had combined for 52% of the vote (24.2% and 27.8%, respectively), we can safely conclude - at least for the time being - that Jason Kenney has successfully managed the post merger right of centre party.

Rachel Notley's New Democratic Party is currently projected at an average of 30.5% of support, down significantly from her 2015 result of 40.6%.

It remains to be seen whether the smaller parties can hold their poll numbers all the way to the ballot box, but so far the Alberta Party stands at 7.9%; the Alberta Liberals at 5.4%


With such a overwhelming lead in voting intentions, it is no surprise that the UCP is currently heavily favoured to win the next Alberta election. Here is the 338Canada Alberta Seat Projection:



Out of the 87 districts of the Legislative Assembly of Alberta, the UCP is currently projected to win between 60 and 80 of them (far beyond the threshold of 44 seats for a majority) with a seat average of 69.5 seats.

The governing NDP would be reduced to its strongholds in and around Edmonton, and not much else. This projection has the NDP at an average of 16.3 seats.

The model ran 50k simulations and the UCP won a majority of seats in every single one of them. There is no doubt, according to the data currently available, that the next Alberta election is the UCP's to lose. It would take both a dramatic turnaround of the NDP's fortune and several major stumbles by the UCP to change this projection.

Of course, opinion polls do not vote, and there are still several months to go before the campaign begins. I will follow the numbers closely over the coming months and, of course, will update the Alberta projection as we approach election day.

* * *

All 87 district projections are available here:





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, 20 January 2019

338Canada Electoral Projection Update: January 20th 2019

Here is the 338Canada popular vote projection for January 20th 2019:



Here is the seat projection:


The Liberals win three quarters out of simulations. With these numbers, a liberal majority is highly uncertain:



All 338 electoral district projections are available per region:


In total, six political parties are included in this projection. For regional projections of those parties, click on the following links:

To browse through the complete map of the projection, click on the following link.





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, 15 January 2019

Mainstreet Research: Jagmeet Singh Takes the Lead in Burnaby South

 A new poll of the Vancouver district of Burnaby South was conducted by Mainstreet Research last week and the results were published today. It looks like Jagmeet Singh may be in a better position to win his by election then was anticipated.

The poll was conducted through automated calls (IVR: Interactive Voice Response) from January 8th to 10th. The sample size is significant for a riding poll: 740 adult voters. Such a probabilistic poll has a margin of error of 3.6%. According Mainstreet's report: "Respondents were interviewed on both landlines and cellular phones, and were given the option of taking the survey in English, Cantonese, or Mandarin," which is understandable given the languages most spoken in Burnaby South.

Here are the results of decided and leaning voters:



The leader of the federal NDP is ahead of the pack with 39%. In a similar poll back in the fall, Singh was third in voting intentions. Perhaps Singh's ground game and visibility in the district have helped boost his appeal towards voters.

Karen Wang, the liberal candidate for the by election, is second with 26%.

Jay Shin, under the conservative banner, obtains 22%.

Finally, Laura-Lynn Tyler Thompson, Maxime Bernier's choice for his party's first by election, is a distant fourth with 9%.

The Green Party of Canada will not have a candidate for the Burnaby South by election.

I will add this riding poll to the 338 model. As usual, all 338 projections will be updated this Sunday. 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, 14 January 2019

338Canada Hockey Model Methodology

Table of contents

  1. The Basics;
  2. Goals per game per team - a Poisson distribution!
  3. Success and Expectancy Rates
(Last updated on January 14th 2019)



1. The Basics

Sometimes, a simple model which provides reasonable amounts of uncertainty is much wiser than a model that considers too many variables and too much information. Yes indeed, too much information can lead to more noise in the results and, even worse, can make us believe we are interpreting better data.

And so the 338 hockey model is quite simple as it stands now (January 14th 2019). It takes into account goals for, goals against, home & away teams, shot % and save %. That's it. Another important aspect: Away games and home games are treated independently, so the performance of a team at home doesn't affect its away stats and vice versa.

The model takes into account games that have been played so far this season and simulates the rest of the regular season 50k times based on stats accumulated so far. Based on these 50k simulations, the model then calculates the playoff qualification odds, final point projections, and so forth.




2. Goals per game per team - a Poisson distribution!

Goal scoring in the NHL follows a Poisson distribution almost perfectly. Consider the following graph:


The red curve is the theoretical Poisson distribution of goals for per game per team of the 2017-2018 season, where the league average was 2.97 goals per team per game.  The blue curve is the actual results of amount of goals per team per game during said season - all 1271 games considered.

Here is the distribution so far this season (as of 2019-01-14 and 704 games played) with an average of 3.05 goals per team per game:

It's eerily close to a perfect Poisson distribution.

Of course such distributions only seem to work when the sample size of games is large enough. However, even at this almost-halfway point of the season, we can compare goal per team per game for a single team and still obtain a result close to the theoretical distribution. Here is the Calgary Flames curve after 46 games:



Here is the Minnesota Wild after 43 games:
And sometimes, of course, the same size is too small to fit the curve. Go home, Philadelphia Flyers - you're drunk.


I will test the model for the remainder of the 2018-2019 season and adjust new parameters as we go along.





3. Success and Expectancy Rates


The model is not supposed to call every game right. In a team parity league like the NHL, even a team low in the standings can have a good night and beat the league leading team. However, odds that a 25th to 31st place team beats a top five team are low, so the model should obtain above average results in the long run.

The success rate is the number of games whose winner was calculated as the favourite the morning of the game. Every game is labeled either as a Toss-up, Leaning or Likely.

  • Toss-up: the team favoured to win has odds between 50% to 55% to win the game;
  • Leaning: the favourite team wins between 55% and 60% of simulations;
  • Likely: the favorite team wins more than 60% of simulations;

Therefore, if the model is accurate, it should call the correct winner in certain fraction of games depending on whether the favourite was likely, leaning, or if the game was a toss-up. This is what is called the expectancy rate. (See table on main page).

If the success rate is much higher than the expectancy rate, then either the favourite odds are underestimated or it's been lucky. In the long run however, the luck factor should fade.

If the success rate is much lower than the expectancy rate, then either the model picks the wrong favorites too often (so the favourite odds are overestimated) or it's been unlucky.

Early in January 2019, when the model was in its first few nights, it picked the wrong winner in twelve of the first fifteen toss-ups. It made me cringe and think that maybe the model was wrong, but it has picked up the pace since with toss-ups. It had just been unlucky on its first nights.

At the end of the season, if the success rates are close to the expectancy rates, I'll be satisfied. Then' I'll try to obsess on more telling data to help make the model better.




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, 13 January 2019

338Canada Electoral Projection Update: January 13th 2019

Here is the 338Canada popular vote projection for January 13th 2019:




Here is the seat projection:


All 338 electoral district projections are available per region:


In total, six political parties are included in this projection. For regional projections of those parties, click on the following links:

To browse through the complete map of the projection, click on the following link.





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, 6 January 2019

338Canada Electoral Projection Update: January 6th 2019

All 338 electoral district projections are available per region:


In total, six political parties are included in this projection. For regional projections of those parties, click on the following links:


Here is the 338Canada popular vote projection for January 6th 2019:

With an average gap of only two points between the Liberals and Conservatives, we have an almost statistical tie in the popular vote projection:



As for the seat projection, even though the Liberals still hold an edge on the average seat count, the seat distribution of the Liberals and Conservatives overlap significantly, as you can see on the graph below:


The New Democratic Party still stands far behind its main competitors at an average of 24 seats. Even a result close to 2015's 44 seats look all but impossible according to current data.


In Québec, numbers for the Bloc québécois remain highyl uncertain. The curve below indicate however that the needle has not moved much since Yves-François Blanchet announced his candidacy for the leadership. We'll see in the coming weeks whether that changes or not.


Lastly,  the Green Party of Canada looks to be in a fairly good position to make modest gains. In 2015, only Green leader Elizabeth May had won a riding under the GPC banner.




To browse through the complete map of the projection, click on the following link.







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.