Wednesday, May 31, 2017

Liberal Majority in Nova Scotia

The Liberals have won another majority in Nova Scotia.

The Liberals took 27 seats and 39.51% of the vote, compared to 33 seats and 45.71% last time.
With 51 seats in the legislature, 27 means room for a speaker, and to still have a 26-24 balance in voting members. The Liberals needed another majority and they got one. This means another 3 years in power, if not a full 4.

The PC Party took 17 seats and 35.78% of the vote, compared to 11 seats and 26.31% last time.
The Tories ran on one of the most right-wing platforms any maritime PC Party has ever run on, and managed to increase their seat count and were competitive in the vote count with the Liberals. Jamie Baillie, the leader, can honestly call this a success.

The NDP took 7 seats and 21.41% of the vote, compared to 7 seats and 26.84% last time
The NDP achieved what it needed to. While it lost votes, it retained 7 seats. Beyond that it ran on a left platform, something that had never been endorsed to such a degree in Nova Scotia. The NDP stemmed off a collapse, and as such, were successful.

The Greens took 2.78% of the vote, with 32 candidates, compared to 0.85% last time, with 16 candidates
The Greens did well. Doubling their candidate count would normally come with a doubling of their vote, but the vote more than tripled.

The Atlantica Party took 0.41% of the vote, with 15 candidates, compared to 0.06% last time, counting the 2 candidates who ran for the party.
The Party is Libertarian in nature, and supports Maritime Union. Prior to the last election the party was deregistered due to financial reporting issues.

3 Independents took 0.11% of the vote, compared to 5 who took 0.26% of the vote last time.

So what is next?

Due to the size of the majority, there is a chance the Liberals won't last the term. Parties on the way out are obvious, and members don't like going down with the ship. MLAs would want to line up private sector jobs, and this likely means some resignations in the final year of government, leaving vacancies that could be won by the opposition.

NS is the only province without fixed election date legislation, but the next election can be expected between May of 2020 and May of 2022.

Monday, May 29, 2017

Nova Scotia

Tomorrow is the election in Nova Scotia. As usual, I do not make predictions for the smaller provinces as the local factor makes province-wide swings more inaccurate.

However, I did want to give readers a heads up about a new blog that just popped up that is covering it.

You can find Project and Elect's Nova Scotia projection here.

Sunday, May 28, 2017

UK Projection

Labour has actually been able to gain a large boost over the last week. Why? It has little to do with the terrorist attack, and, in fact, had the attack not happened, Labour might be leading in the polls.

The answer as to why is quite simple: The dementia tax. In short, it would require you to sell your house to receive care. That's quite a simplification and not exactly true, but that's most certainly how a large portion of the electorate understood it.

The reason this resonates is that it is similar in many ways to concerns under Cameron, in particular, suicides of disabled people.

In short: the Tories are seen as hurting the most vulnerable and just not plain caring about the damage they do. This does not go over well with moderate voters. May was seen as a possible change to that, but the whole dementia tax debacle has shown that this may not be the case.

As such, Labour is up.

Current prediction:

Saturday, May 27, 2017

Speakerships and Leaderships

Andrew Scheer has won the Conservative Party of Canada leadership.

Scheer is a former speaker.

The question came up: has a speaker ever become party leader before.

The answer; from my prior research, is no. However, I will begin looking once more to see if this is truly the case or if I missed something.

As well, it was asked if any leadership this close (Scheer won 51%-49%) has ever produced a leader that won the next election, so I will look at that as well.

Friday, May 26, 2017


The Republican Candidate in Montana, mentioned yesterday, was elected and will become the new house member, backed by a majority of voters.

This is a picture of something I've been working on for a while, off and on. Technically I've been working on this since 2001. I've simply yet to find a quality way to get it online. Hopefully by posting it here, it will spur me into some more action.

Thursday, May 25, 2017

25MAY2017, France and Montana

Not much to update on.

I've run a quick prediction for the upcoming French parliamentary elections:
300 Macron
200 Conservatives
30 Communists
25 Greens
20 Nationalists
2 Regionalists

The biggest news is probably out of Montana, where witnesses say the Republican candidate assaulted a journalist. If true, it is an interesting choice of timing, as today is the date of the by-election for Montana's single statewide house rep in Congress, and the GOP and Democrats were neck and neck (no pun intended) in the polls up to this incident. 

I will post tomorrow morning if he wins the race or not.

Wednesday, May 24, 2017

UK Projection update, and Terrorism and Elections

I spent some time yesterday digging into historic polls in relation to terrorist attacks and elections. I was able to look at polls in various countries including Spain, Australia, Denmark, Sweden, Germany, France, and even Turkey.

My conclusion is that terrorist attacks during an election do not amplify right-wing parties; they amplify the existing biases and opinions regarding terrorism.

Countries known for a more left approach to dealing with islamic extremism and other forms of terrorism actually saw the left parties have a boost in the polls following a terrorist attack. Terrorist attacks bring issues regarding terrorism to the fore, and thus boost parties seen as having the "right" stance to "solve" the problem; be that tolerance or security.

Given polls in the UK that show most brits want a strong hand when dealing with terrorism, it is likely that this will boost the Conservative vote. This is especially true as Corbyn has been attacked for being "weak" on terrorism.

As such, I've updated my projection for the UK:

You may wonder why the Tories have gained so much in the popular vote in Scotland. The answer is they have not. There are a large number of seats that are tipped by a very small number of votes, and with the comparatively tiny increase in popular vote I expect from the Tories here, they would gain a number of seats, including that of Angus Robertson, whose seat is just too vulnerable to withstand gains of this strength.

Tuesday, May 23, 2017

Manchester Bomb

The Bomb in Manchester will have an impact on the UK election; but now is not the time to talk about it.

I wanted to simply present my latest projection, and leave the words for another time.

Monday, May 22, 2017

UK Projection update

I've added a small model house of commons to the display.

In short, Labour is trending up, while the LibDems are trending down.

Thursday, May 18, 2017

Alberta merger and BC Map

I've created a new map for BC elections

Currently it shows the election night results, but this map will be used in future BC elections for making predictions and projections.

Also big news out of Alberta.

The PC Party and Wildrose are merging. This has been in the works for quite some time and we do have a mathematical model of the merged parties ready to go and to make projections with.

Here is the current prediction

As you can see, there are a large number of ridings that are "battleground" ridings between the two parties. As we near the election, I will refine the map and the numbers.

Official party status in the Senate

A recent CBC news article has explained how the rules for official party status in the Senate have now changed.

This has been in the works for a while, with the creation of the ISG Crossbencher group, and the granting of additional power to Independent Senators.

The former rule stated that your party needed one of two things.

1 - To already be an official party in the Senate.
2 - To have 5 Senators in your party, and, have a party that ran in the last election.

This meant that only the Liberals and Tories were ever official caucuses in the Senate, as no other party qualified. Theoretically, had the PC Party maintained an unbroken line from 2003 with more than 5 Senators, they too would have qualified; but the party dipped below this number.

The new rules are simple, you need 9 Senators.

This is much more in line with Commons rules that state 12 members are needed for official party status, regardless of what that party is. 12 Independents could group up to form a caucus even if they have no official party.

9, however, is a fairly high number. As a share of seats, the same number in the Commons would be 29, not 12. The number seems to have been set as 1/12th of the seats in the Senate, which, makes sense, as, being an official party gets you a lot on committee, and many committees have 12 members. 1/12th of the Commons would be 29.

This also prohibits certain things from happening. Only 4 provinces have more than 6 Senators, Ontario, Quebec, New Brunswick, and Nova Scotia; so none of the remaining provinces, on their own, could create a provincial caucus, however regional caucuses could be formed.

We could also see caucuses created for other purposes. For example, an "Amendment Caucus" could be formed by those senators who feel the main purpose of the Senate is to amend Commons bills to be, in their minds, better. We could see an "Urban Caucus" formed to represent the cities, and, a "Rural Caucus" for the rural areas; these are cross-province issues.

In the short term, this will officially give full recognition to the ISG, but in the longer term we may see more political movements that cross party lines, such as the formation of a general-left "Progressive Caucus"

Wednesday, May 17, 2017

Live Podcasts

If you follow my youtube channel, you'll know that yesterday I tried something new, a livestream ramble.

The permalink to my livestreams is here:

What I decided to do was take a video game without voice-over, and record a "podcast" against it, where I discuss various political topics.

Episode 0 helped to teach me what I did wrong and what I did right, and, it was successful enough in my mind, that I will be recording an Episode 1 at some point soon; today, or perhaps tomorrow, that will cover many of the same issues. The key difference will be one of structure, in that I will be pre-planning what I say, rather than randomly rambling.

The topics to be discussed will mirror that of episode 0.

Overall comparisons between the 4 main countries

Presidential systems VS Westminster systems

Parliaments compared (including election methods for all offices)

Parties compared

And terminology compared

You don't need to watch live, and at this time, I don't plan on pre-scheduling episodes. The key reason I livestream is the reduction in steps needed. While it does mean I can't edit things, it makes a fully un-edited video much faster to post online. Otherwise I'd have to record audio, export it to my desktop, merge it with a video, export that, upload it, and wait for it to process. With a livestream, I hit the "go" button, and when I hit "stop" everything is done.

Tuesday, May 16, 2017

Social Media's impact on Resignations

There are three candidates who have been dropped by their parties in the Nova Scotia election. One Liberal, one NDP, and one PC. The story links (where available) are here:

The actual content of what was said, especially in the third case, is rather tame compared to one might see on the internet.

None of these candidates were in particularly winnable ridings, but it never looks good to drop a candidate.

Regardless, while we do see more and more of this, I am also seeing more and more backlash against it. People who use Twitter as a personal platform will have said many things similar (perhaps, though, not so racist) and will not think that such comments are so extreme. With the stuff Trump says and his use of Twitter, I also think that we will begin to see a growing unwillingness to toss candidates based on statements that are not extreme, and as such, this kind of thing is actually on the way out. Within 20-30 years, it will simply be expected that every candidate running has said something stupid at some point in the past on social media, and no one will care.

Monday, May 15, 2017

BC Electoral Reform

I wanted to look at some of the different options for BC should it decide to try to implement electoral reform.

First I want to look at MMP PR using a standard or "normal" fill-up system.

BC has 42 federal ridings, balanced to be roughly equal, ready to go. I've mapped my estimation of the transposition of votes from Provincial results to Federal ridings. It is not a perfect mathematical transposition, and some estimations were made (like, that Weaver could take the few extra votes he needs to beat Carole James to represent Victoria)

This would give the NDP 20 seats, the Liberals 20 seats, and the Greens 2

To this we would add 45 seats from the proportional list, to bring us up to 87, the number of seats currently in the BC Legislature.

No matter which of the three counting methods you use (D'Hondt, St.Laguem or Hare-Niemeyer) you end up with a grand total of 36 Liberals, 36 New Democrats, and 15 Greens.

A parallel system would have differing numbers.

Using D'Hondt, we get 19 additional seats for the Liberals, 19 for the NDP, and 7 for the Greens.

I've also calculated the results using STV. First multi-member:

(Note some transfer assumptions had to be made)

And lastly, single member riding STV (also known as AV) was also estimated using transfer assumptions.

The results are below for comparison purposes:

36 - Liberal
36 - NDP
15 - Green

39 - Liberal
39 - NDP
9 - Green

36 - Liberal
37 - NDP
13 - Green

41 - Liberal
43 - NDP
3 - Green

43 - Liberal
41 - NDP
3 - Green

Friday, May 12, 2017

Thursday, May 11, 2017

British Columbia 2017 with PR

Feel free to try the numbers yourself here:

Regardless, a short post today; if British Columbia had pure Proportional Representation, the results would have been:

36 - Liberal
36 - New Democrat
15 - Green

Wednesday, May 10, 2017

Advance Vote and BC

There has been a lot of talk about the 175K advance votes that have yet been counted. Let me address some math.

There are about 2,000 votes in each riding that have yet to be counted. The question is, can they tip the results of the election?

Included below is the full list of ridings, with the NDP vote subtracted from the Liberal vote. There are only three ridings within the range mentioned.

-170 Coquitlam-Burke Mountain
9 Courtenay-Comox
120 Maple Ridge-Mission

The first was a Liberal win by 170 votes.
The Liberals took 44.4% of the vote in this riding, compared to 43.6% for the NDP. In order to flip the riding, there needs to be 171 more NDP votes than Liberal votes. Assuming the Green vote (11.9%) stays the same, we get the following.

805 Liberal votes, based on the current trend within the riding
791 NDP votes, based on the trend
218 Green votes, and
186 NDP votes that go against the trend

This would mean, of these ballots, only 40.2% are Liberal, 10.9% Green, and 48.9% are NDP. This is a large swing. First, of 4.2% because that is the reduction in the Liberal vote, then, 5.3%, because that's the increase in the NDP vote. That's a 9.5% swing. Not impossible, but a very large swing

This, while possible, is unlikely.

Maple Ridge-Mission would need a swing to the Liberals of 121 votes, which, again, is a lot of ground to make up from only 2000 ballots when, in general, the voters will break similar to how non-absentee voters did.

It's certainly possible for either of these two to flip, but not very likely.

Courtenay-Comox is where the largest chance of a flip is to occur.

-9081 Kelowna-Lake Country
-8930 Kamloops-South Thompson
-8883 Kelowna-Mission
-8504 Kelowna West
-7706 Peace River North
-7420 Shuswap
-7341 West Vancouver-Capilano
-7190 Langley East
-6521 Penticton
-5949 Vancouver-Quilchena
-5673 Delta South
-5326 Abbotsford-Mission
-5307 Surrey-White Rock
-5129 Parksville-Qualicum
-5069 Prince George-Valemount
-5035 Abbotsford South
-5020 Abbotsford West
-4885 Vernon-Monashee
-4600 Surrey South
-4579 Chilliwack-Kent
-4481 Cariboo-Chilcotin
-4355 Prince George-Mackenzie
-4335 Kootenay East
-4255 Kamloops-North Thompson
-4189 Peace River South
-3581 West Vancouver-Sea to Sky
-3253 North Vancouver-Seymour
-2968 Chilliwack
-2705 Richmond North Centre
-2324 Nechako Lakes
-2298 Langley
-2233 Boundary-Similkameen
-2195 Surrey-Cloverdale
-2020 Vancouver-Langara
-1852 Cariboo North
-1836 Richmond-Steveston
-1359 Columbia River-Revelstoke
-1281 Skeena
-1266 Richmond South Centre
-706 Fraser-Nicola
-560 Vancouver-False Creek
-263 Richmond-Queensborough
-253 Oak Bay-Gordon Head (won by Greens, does not count)
-170 Coquitlam-Burke Mountain
9 Courtenay-Comox
120 Maple Ridge-Mission
989 Vancouver-Fraserview
1022 Cowichan Valley (won by Greens, does not count)
1040 Stikine
1107 Saanich North and the Islands (won by Greens, does not count)
1340 Maple Ridge-Pitt Meadows
1420 Port Moody-Coquitlam
1445 Burnaby North
1450 North Vancouver-Lonsdale
1784 Surrey-Panorama
1805 Burnaby-Deer Lake
1841 Delta North
1957 Surrey-Guildford
2119 North Coast
2147 Burnaby-Lougheed
2268 Nelson-Creston
2446 Coquitlam-Maillardville
2791 Saanich South
2838 North Island
3292 Surrey-Fleetwood
3309 Nanaimo
3569 Surrey-Green Timbers
3761 Burnaby-Edmonds
4220 Surrey-Newton
4239 Esquimalt-Metchosin
4290 Nanaimo-North Cowichan
4318 Vancouver-Kensington
4348 Surrey-Whalley
4765 Vancouver-Point Grey
5466 Mid Island-Pacific Rim
5476 Vancouver-Fairview
5729 Vancouver-Kingsway
5836 Port Coquitlam
6082 Langford-Juan de Fuca
6129 Powell River-Sunshine Coast
6178 Kootenay West
7590 New Westminster
7612 Vancouver-West End
7776 Vancouver-Hastings
8539 Victoria-Swan Lake
10180 Victoria-Beacon Hill
10213 Vancouver-Mount Pleasant

BC Projection, what went wrong?

This will be a very short post because the answer is very simple; I thought I could outsmart the polls.

I can't.

BC Counting ongoing

44 BCL // 41 NDP // 02 GRN // (Weaver added to Greens) // 43% BCL // 39% NDP // 15% GRN

Lesson: The polls were right and I was wrong to second-guess them with fanciful mathematical adjustments for my projection.

Tuesday, May 9, 2017

Final BC Projection

I wasn't clear with my last projection that this, is, indeed, my final projection.

I've also included the numbers 1, 2, and 3 in the ridings I'm the least certain about; all could be Liberal victories.

Monday, May 8, 2017

Macron wins French elections

As noted earlier, Macron has won.

66% - Emmanuel Macron - 20,703,694
34% - Marine Le Pen - 10,637,120

Blank votes - 4,066,801

Non-voters - 12,041,278

Looking at the vote patterns and listening to the analysts, its clear that a lot of the 4 million blank voters (roughly 3 million) were Le Pen voters who decided their candidate is too nutty to actually back.

It's likely nearly all of the 7.7M Le Pen voters from the first round, and the 1.7M Dupont-Aignan voters, voted for Le Pen in the second round. The additional 1.2M voters likely came from Fillon, while 5M of his voters went to Macron, and 1M cast blank ballots. It is my guess that 2M Melenchon voters stayed home, 2M cast blank ballots, and 3M voted for Macron. Roughly, of course.

Interestingly, had the number of blank ballots remained at 1M, and not increased to 4M (and had the 3M cast their ballots for Le Pen) the results would have been 61%-39%, which was widely expected.

For Comparison:

(votes in millions)
18.0 Hollande
16.8 Sarkozy
2.1 Blank
9.0 Abstain (stayed home)

18.9 Sarkozy
16.8 Royal
1.6 Blank
7.1 Abstain

25.5 Chirac
5.5 Le Pen (sr)
1.8 Blank
8.4 Abstain

15.8 Chirac
14.2 Jospin
1.9 Blank
8.1 Abstain

16.7 Mitterrand
14.2 Chirac
1.2 Blank
6.1 Abstain

15.7 Mitterrand
14.6 Giscard d'Estaing
0.9 Blank
5.1 Abstain

13.4 Giscard d'Estaing
13.0 Mitterrand
0.4 Blank
3.9 Abstain

11.1 Pompidou
7.9 Poher
1.3 Blank
9.2 Abstain

13.1 de Gaulle
10.6 Mitterrand
0.7 Blank
4.5 Abstain

Sunday, May 7, 2017

The "Surprise Me" factor.

If I've calculate my timezones correctly, this post will publish 5 minutes before the polls close in France.

This is a simple and short post, and should nothing unusual happen in France, I will re-schedule this post for this evening, and, it will become hidden again, as to emphasize the QC125 post.

In short: I judge what will not happen by what would surprise me.

Would it surprise me for Le Pen to win? No. In fact it would not surprise me for her to take as much as 55% of the vote.

Nor would it surprise be for Macron to win by as much as 72%

These are, clearly, extreme ranges, but this is what I'm expecting.

In BC it would not surprise me for the NDP to win as few as 12 seats, nor would it surprise me if the NDP won a majority and the Liberals won as few as 20; nor would it surprise me if the Greens won as many as 15, or as few as 0 for that matter.

I will try to remember to include the surprise me factor in all future posts.

French election results

Roughly 12 million did not turn out to vote.

Of those who did, 4 million spoiled their ballot by casting a blank vote.

20 million voted for Macron

10 million voted for Le Pen

QC125 and Poll Averaging

QC125 is a new website that looks at elections and similar stats in Quebec. I highly recommend reading their latest article, which is in French. I thought I would summarize it in English for those of you, who like me, are unilingual.

PJF, who runs the site, ran a little simulation. He generated 5,000,000 votes (a roughly quebec-sized electorate) and then randomly picked 1000 of these voters to be "polled"

The results of the polls are here

None of the parties actually moved up, or down, during this period. Red was assigned 32%, Blue 28%, Green 24%, and Orange 12%.

This is why we poll average. This is also why those of us who use polls to analyze politics, go out of our way to ignore single polls. A single poll showing Red and Blue tied means nothing, and this shows exactly why, just as a single poll showing a lead of 8 points for Red means nothing, nor a poll showing Green ahead of Blue.

Poll averaging is critically important as it gets rid of the "margin of error" each poll comes with naturally. With a margin of error of 3% (as in the above) a poll showing, for example, 43% and 43% is identical to one showing 49% and 37%. Poll averaging helps us weed out this margin of error and figure out what is really going on.

Next time someone tries to show you an individual poll and suggest one candidate or another is therefore winning, show them these two posts, and remind them that error in polling is part of the process and not a mistake.

Saturday, May 6, 2017

UK Projection Update

As well some local election results.

The Tories increased their seat share by a ratio of 1.42 as they now hold 1899 seats, and previously held 1336. Labour's ratio is 0.75, while the LibDems managed 0.91. The SNP is at 1.01 while UKIP is way down at 0.007 as they lost all but 1 seat.

Friday, May 5, 2017

Quickie: English and the EU

This is actually a topic I've addressed before.

A reminder that even without the UK, more people in Europe will speak English than any other single language by a wide margin. 42% vs 30% for German, and 25% for French.

Consider Sweden, Finland, Denmark, and the Netherlands. Each has their own home language, different from the home language of the others, but all have high proficiency in English. Any move to get rid of English as a language will result in these countries getting irritated, at minimum.

There are, in fact, more English speakers than German speakers in every EU country except Austria and Germany, where the language is Official and Native, and, more English speakers than French in every country except France, Luxembourg, and Belgium, where the language is either Official or Native. The remaining 22 countries all speak more English. You can drop that number to 20 if you count both Italian and Spanish as separate languages for this purpose, and 17 if you count Russian, and 16 if you also count Polish. Even at these levels (English vs German, French, Italian, Spanish, Polish, and Russian) there are still 16 countries that speak English more VS 11 that speak one of these languages more.

In short; even without the UK, and even without Ireland and Malta, who are staying in the UK, and where English is official, English is the #1 most used language in Europe, and will remain so for the time being.

BC projection update

UK Local elections are not fully counted at this point, so in the interim, I've updated my BC projection to be more sensitive to regional impacts.

Thursday, May 4, 2017

UK Local elections today

Councils across most of Britain go to the polls today for local elections. This will be a good preview of party strengths and weaknesses in the General election, and, will have a pulling effect on the trendline momentum, speeding up the shifting of parties in the polls.

Not much else to say at this point; but will report on results tomorrow, and what it may mean.

Wednesday, May 3, 2017

Boldest Projection I've made yet - BC

I've mentioned before that I am able to tell when certain polls are wrong, but not by how much.

On a fluke I tested out using ratio rates of responses to modify polls, and, it seemed to work very well in 2013. As such, using that same math, I am going to make a projection for BC. A very bold one.

That is all at this time.

Tuesday, May 2, 2017

Point form Conservative Party Leadership

None of the tiny candidates really can overtake the larger ones at any point.

Andrew Saxton
Deepak Obhrai
Rick Peterson

The smaller candidates are also too diverse to make overtake any of the medium sized candidates

Chris Alexander
Brad Trost
Steven Blaney
Pierre Lemieux

Given that they share some ideological points, the medium sized candidates might be able to support one another; but, in reality, neither of them can overtake the next largest candidate.

Lisa Raitt
Michael Chong

Finally this brings us to

Kellie Leitch

Who is an ideological candidate, someone with "ideas" and will thus be not as transfer friendly as otherwise might be the case. As such, she will be 4th from last.

This brings us to the 3 largest candidates

Erin O'Toole

O'Toole might be able to hoover up moderate votes and pass Scheer, but I doubt it.

Andrew Scheer

Scheer has positioned himself as the "Harper" of the race, and would basically continue Harper's policies. The problem is this still leaves him behind, the eventual winner.

Maxime Bernier

With strong support from Quebec, and the riding-based system used in the party, Bernier, an "ideas" candidate (Libertarian in nature) will likely win the leadership, and, as a result, move the Conservative Party to a more Libertarian viewpoint.

Monday, May 1, 2017

May, the month to come

Busy month. On the 4th the UK holds local elections, while on the 7th France holds the final round of their Presidential elections. The 9th sees an election in British Columbia (my gut says the NDP will win) before we get some time off. On the 30th there is an election in Nova Scotia, where the Liberals hold a commanding lead. This is all in advance of the UK elections on the 8th of June. France will hold Parliamentary elections on the 11th and 18th of June as well.

Of course I'll keep you all up to date, and may even find time to look into Algeria (4th of may), Schleswig-Holstein (7th), South Korea (9th), the Bahamas (10th), North Rhine-Westphalia (14th), Iran's Presidential (19th), Lebanon (21st) Lesotho (June 3rd), Puerto Rico (11th), Albania (18th) and Mongolia's Presidential (26th)