2015 Federal Election Prediction

[cybercoma]

None of the polls in Canada can be trusted because popular support is a weak indicator of seat distribution. Our government is based on winning seats by a plurality of votes. Our polling is based on proportional representation. It translates loosely, but not without losing something in the conversion.

[Accountability Now]

I'm looking at a poll that was conducted that says you're wrong. The polls may be misleading as to the results of elections, for exactly the reason I alluded to about how the system works. However, that doesn't negate popular support. Harper's Conservatives got less than 40% popular support in the last election and hold a majority of the seats in the House. When you look at the popularity of Nenshi, the fact that the NDP holds a seat in Calgary, and what scientifically conducted polls show, it becomes quite clear that there is a significant amount of Leftwing support in Alberta. You care to show me any evidence to the contrary, other than your opinionated hunch?

Many scientific polls suggested Wildrose would be in power. We have a Conservative majority. Any other evidence required?

Btw...polls are taken on a small sample size. So don't bring the word scientific into it as anyone in science know the probability for error increases with smaller samples.

[Accountability Now]

1. Where on the poilitical spectrum do you think the mayors of Calgary and Edmonton would be?

2. The Liberals are not "Left wing"

3. You are right about provincial polling in Albera - (and BC), they can't always be trusted.

1. Don't know as I can't vote in those elections so I don't really follow their platforms. Regardless...mayors do not follow party politics. My point is that Alberta has always been and for the conceivable future, will always be Conservative.

2. Are you kidding me? Then what are they and please don't say Center as I don't want to laugh myself to sleep.

3. I think polling everywhere is faulty as it is limited to sample size. It becomes particularly faulty when certain parts of the population are chosen over others....with the others being the majority.

[cybercoma]

Many scientific polls suggested Wildrose would be in power. We have a Conservative majority. Any other evidence required?

Did you read my post?

Btw...polls are taken on a small sample size. So don't bring the word scientific into it as anyone in science know the probability for error increases with smaller samples.

All of that information is available in the methodology of the polls. That is if you take the time to actually go to the polling firm's publication and not just look at the headlines from a news agency.

The problem here isn't margin of error, however. The problem is exactly what I said. Polls look at popular support, which isn't how our system works.

[Accountability Now]

None of the polls in Canada can be trusted because popular support is a weak indicator of seat distribution. Our government is based on winning seats by a plurality of votes. Our polling is based on proportional representation. It translates loosely, but not without losing something in the conversion.

The poll that you seem to support disagrees with you. As per the last time these guys did a federal election poll:

http://www.threehundredeight.com/2011/05/projection-vs-results.html

However, it was not the seat projection model that failed. The seat projection model actually performed very well - or would have had the popular vote projection model not missed the mark so completely.

In fact, inputting the actual provincial-level vote results into the model, treating them as the results of a poll, churns out a very accurate result.

Their error last time is that they felt popular support was higher or lower than what it is when in fact they didn't get popular support right at all. As is what I am saying is the case here.

ThreeHundredEight.com projected two things: the popular vote and how that would transform into seats. It failed at the first, which means it failed at the second.

[Accountability Now]

Did you read my post?

All of that information is available in the methodology of the polls. That is if you take the time to actually go to the polling firm's publication and not just look at the headlines from a news agency.

The problem here isn't margin of error, however. The problem is exactly what I said. Polls look at popular support, which isn't how our system works.

As per my last post....they didn't even get the popular vote right which you are saying is what this poll is about. I said its not. Anymore to discuss?

[cybercoma]

Ok. I'm convinced that that you're more interested in trying to teach me something than actually reading what I said, since you're repeating back to me the problem with 308's modelling, which is exactly what I said.

[cybercoma]

Oh and 308 couldn't even explain coherently why his modelling failed. He blamed regional support polling, but also went on to say, "I should be clear that the polls were not terribly inaccurate...But they under-estimated the Conservatives." He goes on to say in the comments, "Had the polls not shifted so late in the campaign, the model would have likely given an accurate result. It failed at capturing the swing in voting intentions. For the first three weeks of the campaign, the projection was almost certainly very close to what would have happened." So like I said, the problem lies between translating the popular support polls into seat projections. His modelling isn't capturing the swing in voter intention that happened because it's placing too much weight on earlier polling.

Edited December 7, 2013 by cybercoma

[Accountability Now]

Ok. I'm convinced that that you're more interested in trying to teach me something than actually reading what I said, since you're repeating back to me the problem with 308's modelling, which is exactly what I said.

You keep saying that the polls are an indicaiton of popular support. I disagreed and have shown you, as per the last 308 federal election polls, that they don't.

I have been reading everything you have posted.

[Accountability Now]

Oh and 308 couldn't even explain coherently why his modelling failed. He blamed regional support polling, but also went on to say, "I should be clear that the polls were not terribly inaccurate...But they under-estimated the Conservatives."

If you read the rest of it you will see it shows that IF they had the popular support correct, then the seat projection would have been accurate. It is fairly clear that the popular support on the last federal poll was NOT correct. What's to believe its correct this time?

[cybercoma]

Ok. I'm obviously not getting through to you.

You're right. There's absolutely no way whatsoever to understand what party people support.

[Accountability Now]

Ok. I'm obviously not getting through to you.

You're right. There's absolutely no way whatsoever to understand what party people support.

You're not getting through to me? Don't get upset because I have pointed out a fallacy in your postion. I undestand that you have expressed these models aren't 100% accurate. And in theory, popular support doesn't equate to seats. I get that 100%. My point is that these polls are nothing more than a conversation piece.

edit: They are a convesation piece because even 308 has shown they can easily get the popular vote wrong. They have also shown that when they do get the popular vote right that it does equate to accurate seat projections. Something you clearly disagreed with.

Edited December 7, 2013 by Accountability Now

[cybercoma]

And you're wrong. Even Eric admitted in his comments at the bottom of the page that you linked to that the polls captured the shift in support, but his modelling did not. He said explicitly, "had the polls not shifted so late" then his projections would have been accurate. So he's saying the polls weren't inaccurate after all. It's the fact that his modelling weighted earlier polls too much, but he didn't admit that in his article for some reason. This is why he blames the popular polls early on, then turns around and says they weren't terrible inaccurate and then in the comments below says the popular polls shifted later (obviously capturing voter intention). What his modelling doesn't account for is people changing their mind during an election cycle.

And on that point, what this poll above doesn't account for is anything other than a snapshot in time when the poll was taken. At that time, popular support in the province is around what they say. Is it really that farfetched that 40% of Albertans support Leftwing parties? As has been pointed out to you, the mayors of both major cities are left leaning, the NDP has a seat in Calgary. Despite this, the polling still shows that more than half of the voters support the Conservatives. Is it really so far fetched to imagine that 2 in 5 voters would vote for a leftwing party? They're still outvoted 3 to 2. That's despite waning support across the country for Harper due to the scandals that are plaguing him.

Edited December 7, 2013 by cybercoma

[Accountability Now]

And on that point, what this poll above doesn't account for is anything other than a snapshot in time when the poll was taken. At that time, popular support in the province is around what they say. Is it really that farfetched that 40% of Albertans support Leftwing parties? As has been pointed out to you, the mayors of both major cities are left leaning, the NDP has a seat in Calgary. Despite this, the polling still shows that more than half of the voters support the Conservatives. Is it really so far fetched to imagine that 2 in 5 voters would vote for a leftwing party? They're still outvoted 3 to 2. That's despite waning support across the country for Harper due to the scandals that are plaguing him.

Alberta is the only province that does NOT have a single Liberal seat. And they have one NDP seat....ONE measly seat out of 28 seats total.And you're making a big deal about this? Hell the Green Party has a seat in BC...I guess they're going to win that province...hey? I get that seats might not equate to popular support but come on...what is more likely....that 40% of Alberta support leftwing when they currently have one seat to show for it OR that the poll is not correct. I choose the lattter but believe what you want.

As for the mayors...they may have left wing tendancies when compared to the people they run against however does that make them left wing. Of course, people in civic elections don't vote for parties....they vote for people. The same is not the case in provincial or federal as people will vote party over people.

The scandal? Lol....people in Alberta don't give a crap about $90k. They care about Keystone, Gateway and any other pipeline or project that will create revenue and jobs. Any announcement of these pipelines will secure a Conservative vote in Alberta...maybe not the rest of the country but definitely here.

Edited December 7, 2013 by Accountability Now

[waldo]

I don't think you quite understand how our electoral system works. There could be 65% Left support in Alberta and the Right could win every single seat in the province.

at that provincial level, Alberta got a taste of real right-wing the last provincial election... Albertan's showed they didn't like it; didn't like the social conservative side and that's why the former Wildrose has now morphed into the Mildrose. Apparently someone doesn't know what a Progressive Conservative actually is... doesn't know what a 'red tory' is.

[cybercoma]

Alberta is the only province that does NOT have a single Liberal seat. And they have one NDP seat....ONE measly seat out of 28 seats total.And you're making a big deal about this? Hell the Green Party has a seat in BC...I guess they're going to win that province...hey? I get that seats might not equate to popular support but come on...what is more likely....that 40% of Alberta support leftwing when they currently have one seat to show for it OR that the poll is not correct. I choose the lattter but believe what you want.

So you know seats don't equate to popular support, but you're going to equate it anyway. Harper's government had less than 40% popular support at the polls and held nearly 65% of the seats after the election. So I don't need to tell you how silly it is to make the leap in logic that you're doing here. You already know.

So let's look at popular support in Alberta at the time of the election. The Conservatives got 2/3 of the popular vote in 2011. Are you still shocked at those numbers above that have a 60/40 split, instead of 66/33?

There's nothing even remotely provocative about those numbers and you're acting as though they're impossible.

[cybercoma]

at that provincial level, Alberta got a taste of real right-wing the last provincial election... Albertan's showed they didn't like it; didn't like the social conservative side and that's why the former Wildrose has now morphed into the Mildrose. Apparently someone doesn't know what a Progressive Conservative actually is... doesn't know what a 'red tory' is.

Exactly. It's a myth that Alberta is Canada's Texas (which is the Inception of myths, since Texas is becoming increasingly liberal). There's solid Red Tory support out there and there are some Hard Right voters, but it's not as cut and dry as Accountability Now wants it to be. The seat counts in Alberta do not at all reflect the popular support in the province. If it did, then the Conservatives would have only won 2/3 of the seats. That's not how our system works however. Edited December 7, 2013 by cybercoma

[overthere]

The Alberta PCs are not a right wing party and neither are the people who voted for them.

They are solidly in the center, like nearly all successful parties in Canada. Credit Farmer Ed Stelmach for much of that, he quietly engineered a winning scenario from his own political execution.

The provinical Liberals are done here, they sealed their fate by electing the certified idiot Raj Sherman as their leader. Sherman bleated relentlessly about queue jumping scandals, bullied Redford in having a full and expensive public enquiry, and has had to hide ever since when the enquiry came up with a big fat nothing for $10 million. in expenses. Nobody knows or cares where the Liberals are in the political spectrum, I doubt they know themselves. Larry Decore must be rolling in his grave, he came pretty close 20 years ago.

I don't see Wildrose gaining much traction in the middle. To do it they'd have to do exactly what the PCs did a few years ago: jettison all the hardcore Godbotherers and right wing lunatic fringe from their ranks. What would be left of Wildrose? The PC party.... Won't happen, not soon anyway.

[overthere]

The shocking number in the last Alberta election was the huge jump in voter turnout. All the people that didn't vote in the last Tory landslide came out for this won, and they all voted.... Tory. They'd seen enough from Danielle Smith and her cast of human dinosaurs. Nary a pundit predicted it. Puzzling.

Smith is going to have to dial down the volume on her entire party including herself to have any chance next time. She';ll have to strap muzzles on a majority of her candidates and choke diseent to an extent that would make Harper look loquacious to succeed- and they won't stand for it. She is going to be in a very tough spot.

[Accountability Now]

So you know seats don't equate to popular support, but you're going to equate it anyway. Harper's government had less than 40% popular support at the polls and held nearly 65% of the seats after the election. So I don't need to tell you how silly it is to make the leap in logic that you're doing here. You already know.

So let's look at popular support in Alberta at the time of the election. The Conservatives got 2/3 of the popular vote in 2011. Are you still shocked at those numbers above that have a 60/40 split, instead of 66/33?

There's nothing even remotely provocative about those numbers and you're acting as though they're impossible.

Pardon the pun....but you are quite liberal with your numbers. You say they are claiming a 60/40 split when its actually its actually 54.6/43.7 with 1.7% going to other. Why is this a big deal? Because the highest poplular vote for the left wing (in Alberta) since the 1997 federal election was 37.6% which happened during the Stockwell Day- Harper transition (In other words, who would support the right wing at a time like that?). Over the six elections, the average has been 31.8% popular vote for left wing parties. BUT now....this poll is saying they have 43.7%? And I'm the one taking the leap? LOL. So its roughly 12% higher than the six election average (31.8%) and the 2011 election (31.3%). To me that is a significant difference which is why I say these polls don't seem right especially considering the extenuating circumstances of having Trudeau Jr (ie anti-Alberta himself) running against Harper (an MP from Calgary). Of course don't let the historical numbers get in the what you feel is a realistic projection.

The funny part is you claim in the article that the polls weren't inaccurate. What he actually said is that the polls weren't TERRIBLY inaccurate....meaning they were still off. He goes on to say about how they underestimated the strength of the conservatives especailly in BC, Ontario and Atlantic Canada. That is quite the underestimate considering those three areas alone win the Conservatives a minority government. He also goes on to say that the weighted average polls were less accurate than the polls closer to the election. Yet....what is the graphic that was posted? Weighted average. Hmmm.

I have never said that the left wing in Alberta is dead or that it doesn't have some support. I just don't believe its at 43.7%.

[Accountability Now]

Exactly. It's a myth that Alberta is Canada's Texas (which is the Inception of myths, since Texas is becoming increasingly liberal). There's solid Red Tory support out there and there are some Hard Right voters, but it's not as cut and dry as Accountability Now wants it to be. The seat counts in Alberta do not at all reflect the popular support in the province. If it did, then the Conservatives would have only won 2/3 of the seats. That's not how our system works however.

My original post was to show disagreement with the suggested 43.7% left wing popular vote. I have never said we are hard nose red wingers through and through. I am very aware that the left has some support. I just think that number is off and the historical numbers tend to agree with me.

[BC_chick]

Dec 2 reference - (before the Harper Conservative "truth" attack ads begin in the new year):threehundredeight.com weighted average of the latest federal polls

It never ceases to amaze me how differently BC polls compared to my everyday experiences with the people I meet. People love Harper out here. 1 in 4 seems way too low and I even live in the city. Edited December 8, 2013 by BC_chick

[waldo]

It never ceases to amaze me how differently BC polls compared to my everyday experiences with the people I meet. People love Harper out here. 1 in 4 seems way too low and I even live in the city.

not my 'everyday' experience... even my buds/family in the interior, some with former long-standing allegiance to Conservatives, now want nothing to do with Harper Conservatives. In any case, the Northern Gateway focus is about to ramp up... there's been a "lull". Expect that to rally the anti-Con vote in short order.

[BC_chick]

I hope so.

As for my personal experiences, maybe it's my neighbourhood (above average income) or the industry I work in (finance), but it seems to me that it's more than 1 in 4 people that I meet that have drank the Harper Kool-Aid.

[ReeferMadness]

It never ceases to amaze me how differently BC polls compared to my everyday experiences with the people I meet. People love Harper out here. 1 in 4 seems way too low and I even live in the city.

Maybe you just run with a bad crowd.