Saturday, October 31, 2015

Comparisons - United Right

I'm working on various spreadsheets to show historical results of elections. One of the things I'm compiling, is Federal elections, with popular vote by province.

I've done some math with that to show some interesting things that should be kept in mind when judging historical results compared to the 1990's and it's divided right.

First, lets start with Ontario.

At the top, we have the simplest way to present the results. The "Conservative" vote follows both the modern Tories, and the old PC Party. You can see during the 90's, there was a clear drop in popular support. At the highest level - 18.8% in 1997 - it comes nowhere near the next lowest - 32.0% in 1968.

Seen in the middle, switching the PC Party for the Reform Alliance does not help either. Our newest high, 23.6%, is still a long distance from our lowest low from outside the period.

In fact, these results only really follow a clear pattern if you merge the two, as seen at the bottom. Now we get results that fall in line with history, all right about 38%, within 0.3% to be exact. You'll find many results above this, and many below, which tells me that the "historic" voting patterns of Ontario are best shown, during the 1990s, as a "divided right".

So, we've concluded that in order to best compare to history, we should simply add the PC+CA vote, right? Not so fast. Next, look at Alberta.

At the top we can clearly see the "Tory" line makes little sense.

The middle line is far better, and though there is a drop down to the 50's from the 60's, both before and after, this line fits in much better with our historical trends.

However, the bottom shows why it is not a simple addition. 72.4% would be, by far, the largest we've seen. Even the "low" of 66.9% only compares to the massive Mulroney majorities, or the Harper majority of 2011, at a time when the country had a Liberal majority.

As such, in Alberta, it clearly makes more sense, in context, to "replace" the PC vote totals with that of the Reform Alliance for the period in question.

I don't have a PEI graphic ready just yet, but, I have created tables like this before. In that table we see a pattern at the top; in that it is the PC Party that retained the "vote pattern" of right-wing voters in the province, with the Reform Alliance never able to establish itself.

As such, this post is sort of a "heads up" for what to expect when I do finish transcribing all this data. While I do plan to have a "clean" version of votes from 1988-2000, I plan to "merge" the numbers for comparison purposes where needed.

Friday, October 30, 2015

Maps question

I've been working on possible new maps. Here are the rough designs. Feedback welcome:

Tuesday, October 27, 2015

Validation and Recounts

Not all ridings have validated their results. This is a process where - if I understand it correctly - the central office for the riding takes a look at all the information to ensure everything is in order.

As expected, this means some of the more remote ridings have yet to validate.

In addition, there are 3 recounts that have been ordered.

As such, I've prepared maps of only the uncontested validated results, meaning unvalidated ridings, or, ridings in recount, are shown as empty.

Monday, October 26, 2015

Vote Patterns, Alternate history, and the Canadian Alliance

Now that the election has passed, I can finally use some of my time to post about things beyond just current Canadian Federal politics.

One thing I've always been keen on is alternate history. What if this or that happened, or, did not happen. In this case, I'd like to look at what would have happened if, in the 2000 election, the Alliance was able to displace the PC Party.

The assumption is that sometime between 2000 and now, the federal PC Party has effectively ceased to exist.

What is most interesting is what happens when you examine the vote patterns of the Alliance. Here is Ontario, in the 2000 election, with a boosted Canadian Alliance vote total:

As you can see, while not perfect, the vote patterns are very close to what happened this election, and, that is my point.

The current Conservative Party has finally lost any element of the former vote pattern of the old PC Party. In fact, especially within Ontario (416 and 905 in particular) that entire vote pattern is now in the hands of the Liberals.

One of the only areas of the country the PC Party retained it's vote in, no matter how bad things got, was in the Atlantic. As such, I present the following alternate history. One which would see the Liberals still retain a majority, but, a very narrow one (172 seats) and has a PC Party with status (12 seats) all from the Atlantic, and, a Canadian Alliance with realistic and historical vote patterns, presented in 2015.

Not terribly exciting, perhaps, but something interesting to think about.

If the NDP had held on

There is some evidence that the Liberal-NDP vote this past election was very fluid.

As such, I decided to look at what could have happened if the NDP had held on right to the very end, and it was they, not the Liberals, who won a majority.

If such, the results would likely have been very much like this:

Projection model VS Results

In this experiment I compare the projection model, with proper polling numbers, to the actual results. Maps are below; note that in each case, the first map shows the real results with the second showing the model.







Such errors are commonplace, as, it's impossible to adjust for all local variations, nor will it ever be possible to do so.

Friday, October 23, 2015


There are many things to discuss about the Senate, that may take many posts to get though. I will start with an interesting fact, and begin by going over some numbers.

The "median" Prime Minister has appointed between 37 and 39 Senators, depending on weather or not you count the original 73 appointments as "a Prime minister" or not. Among the bottom half, the median has appointed at least a dozen senators.
For the purposes of the following statistics, I will be ignoring the 6 Prime Ministers who appointed under 12 Senators.

Of the 16 Prime Ministers we are looking at...
Only 3 appointed nothing but Senators from their own party. Mackenzie and Bowell, both from the 1800s, and Harper. It might be surprising for some to learn this, but, every Prime Minister (who we are looking at, IE excluding the 6 who served very short terms) during the 1900s appointed at least 1 Senator from outside the governing party.
Don't get too excited. 10 of the 16 did not appoint any Senators who were from the opposition. The other 7 appointed Senators who sat as Independents, either as true non-partisans, Independents who backed the government party to a lukewarm extent, or Independents who opposed the government.

Only 6 appointed Senators from the opposition. Who were those 6 and why?

Mulroney appointed 1 opposition Senator, a Reform Party member who won an election.

St. Laurent appointed 1 PC Senator after his advisors told him to, as, the PC Party had so few Senators, they were at real danger of losing party status in the Senate.
Borden appointed 3 Liberals, all in 1918. From what I can gather, all Unionists.

Macdonald appointed 9 Liberals, and, oversaw the Royal Proclamation that founded the Senate, which appointed 27 Liberals. At the time, however, it was not yet precedent to appoint only your own to the Senate.

And so, of the 16, 14 clearly had "partisan motives" for their Senate appointments. What of the other 2?

The most recent of them is Paul Martin. He appointed 2 Conservatives, Andree Champagne and Hugh Segal. While both moderates, they both supported Harper in a way. Martin had also appointed "Progressive Conservative" senators, but neither Champagne nor Segal quit the official party to be PC Senators. Part of the problem with judging Paul Martin is he only spent a few years in office, and we can not know if he would have simply appointed more and more Liberals had he spent longer in office.

Last is the Prime Minister who has done the most for Senate Reform since confederation. It was on his watch that Senators received their mandatory retirement age.

Pierre Trudeau.

Pierre Trudeau appointed 81 Senators. Of those 8 were from opposition parties, a full 10%. 7 of them were Tories, he appointed them throughout his entire term of office, starting in 1972, and ending in 1982, with the latter Senator serving right up to the year 2000. Trudeau also appointed a Social Credit Senator, Ernst Manning.

So why does any of this matter?

Simple. Justin Trudeau has already done more to reform the Senate than Harper ever did, when he released his Senate caucus.
Justin Trudeau has indicated, by various statements, that he wants the Senate to be much more politically independent and much less partisan.

At this time, there is a vacant seat in PEI in the Senate. Given that Mike Duffy is the only non-Liberal in the Senate from PEI, Justin might show he means what he says by appointing a Conservative to this job.

There are many options. Pat Binns was always a moderate, yet, had backed the idea of a united right from the get go. Binns, should he want the job, would not only be a moderate, yet also a loyal big C Conservative Senator. The problem with Binns is he'd retire a year before the next election, unless, Justin plans an early election using our new electoral system.

Many other provincial politicians from the PEI PC Party could fit the bill as well. I won't list them all, but the fact remains that this is 'the sort of thing he would do' and as such, I personally would not be surprised to see it happen.

Consider as well there as a vacant seat in BC, a province full of both big G and small g greens. Ontario has 7 vacancies, That's enough space for 3 Liberals, 2 true Independents, and, of course, myself! Though, I admit, the latter is a bit unlikely.

With 22 vacancies, and a commitment to changing the Senate, it is quite possible we could see appointments sooner rather than later, and if so, I would keep an eye open to those appointments not from the Liberal Party. It is quite possible that no Conservative would be appointed, and, also possible no Liberal will be either. Appointing 22 people who have never held a public political affiliation would certainly be a very bold statement indeed.


A quick map series showing which ridings were won with 50%+1 of the vote