Federal Election Polling Thread

[octothorp]

In the past, we've had separate threads for general political discussion, and for discussing polling.

So, consider this thread your place to discuss the latest polls, make predictions.

Try not to get into debates about how different parties will be good or bad for the country/province/you, how much you like or dislike parties or leaders or your local candidate, etc.

Some helpful links:

http://www.cbc.ca/news2/interactives...015/index.html - A projection based on an aggregate of polls, done by Eric Grenier.

http://www.threehundredeight.com/ - Eric Grenier's website, which provides more context than the cbc site, but similar data/focus.

http://www.electionprediction.org/ - an election predictor with more of a riding-level focus. If you're interested in a specific riding, there's often some good discussion that sheds more light than the polls do.

https://en.wikipedia.org/wiki/Opinio...deral_election I'm finding wikipedia as one of the best sites to get updates and links to actual polls.
http://news.nationalpost.com/news/ca...nteractive-map Interactive map on NP showing most competitive races last time around.

Post a link that you think is a good reference and I'll add it to this post. (I admit I almost exclusively go to 308, so I don't know a lot of what else is out there.)

[octothorp]

As of today, August 4th, here's where we're at in the aggregate:
NDP: 33.2
CPC: 30.9
LPC: 25.9
BQ: 4.7
GP: 4.7


The NDP got a big boost in a Forum Research poll that came out yesterday and had them at 39%. Lots of reasons to be skeptical about that, including the NDP jumping up to 45% in Atlantic Canada, which is about 12% higher than they typically poll. Also a huge NDP jump in the Prairies. Neither of those are out of the realm of possibility, but I'll be skeptical until other polls show a similar trend.

[octothorp]

That was a really quiet week in the polls following the debate, but we're starting to see a few more now.

The only unmistakeable trend is a slight uptick in Liberal support, which makes sense with the idea that their numbers were deflated on the Trudeau at least beat the low expectations that this narrative set for him.

What's interesting is where the Liberal support seems to be coming from. In most regions (Alberta, BC, Atlantic, Quebec), there's been a slight NDP downturn to go along with the Liberal gains. The Prairies are the one region where the PCs have improved since the debate. In Ontario, the CPC has had a notable downturn while the Liberals rise. This suggests to me that there's some fluidity in most of Canada between NDP and Liberal support, but in Ontario there might be more fluidity between Liberal and Conservatives. It's notable that even in Alberta, where predictably NDP numbers are down following McCuaig's comments, the Liberals look to be more a beneficiary of that than the Conservatives.

Anyway, not a lot of data to go on right now, many of those trends might turn out to be nothing. But if it was ever the plan of either the NDP or Conservatives to turn this into a two horse race, I'd say that has failed to happen.

[Iowa_Flames_Fan]

Yeah, I was getting ready to conclude that the debate didn't really move the needle, and I now think (after seeing a few more polls since then) that it just didn't move it all that much. However, there does seem to be an appreciable, but slight, movement toward the liberals, likely due to Trudeau exceeding expectations.

On a riding-specific level, I haven't driven around town or anything--but I'm starting to see a lot of Kent Hehr signs in Calgary Centre, and nothing for Joan Crockatt, or for any other candidate.

That's a riding that I've thought for a while could be one the Liberals might pick up--keeping in mind that Crockatt barely eked out a victory in the byelection, and was assisted by a vote-split between the Liberals and Greens that may not be repeated. Whether the NDP makes a push in this riding could be an important factor, but so far no sign of them.

[octothorp]

Originally Posted by Iowa_Flames_Fan

Yeah, I was getting ready to conclude that the debate didn't really move the needle, and I now think (after seeing a few more polls since then) that it just didn't move it all that much. However, there does seem to be an appreciable, but slight, movement toward the liberals, likely due to Trudeau exceeding expectations.

On a riding-specific level, I haven't driven around town or anything--but I'm starting to see a lot of Kent Hehr signs in Calgary Centre, and nothing for Joan Crockatt, or for any other candidate.

That's a riding that I've thought for a while could be one the Liberals might pick up--keeping in mind that Crockatt barely eked out a victory in the byelection, and was assisted by a vote-split between the Liberals and Greens that may not be repeated. Whether the NDP makes a push in this riding could be an important factor, but so far no sign of them.

Yeah, I think that Centre, Skyview, and Confederation are all in play, and the fact that the NDP is so slow-starting in all of these ridings is going to help the Liberals immensely. Centre and Skyview should be close to straight-up two-way races between Liberals and CPC, while in Confederation, vote-splitting between Liberals and NDP could be much more of a factor and allow the CPCs to win with a mid-30s percentage of the vote.

[Slava]

I think the NDP botched Calgary. Basically they're so far behind in nominating candidates and getting things going that in those ridings in particular they've made the Liberals the de facto choice for a progressive voter. Calgary Centre is a prime example; progressives recognize that the vote split cost them before and aren't going to make that same mistake again. I don't know if the NDP has nominated a candidate yet, but it's obvious that Kent Hehr has been working hard for a year there.

Then you look at Confederation where Matt Grant has been door knocking and fundraising for a year or longer and the NDP just nominated a candidate. They're just behind the eight ball already.

The thing is its more than just a candidate nomination. It means fundraising, volunteer recruitment, signs, brochures and everything else are delayed. It's not insurmountable, but a big blow for a party that was hoping to capitalize on the provincial win.

[SebC]

I've seen it suggested - can't remember where, but probably the other thread - that some of the people who would've run for the NDP federally are currently unavailable due to their surprise elections as MLAs.

And to get us back on topic... Leger today has NDP 33%, Lib 28%, Con 27%.

[Slava]

Originally Posted by SebC

I've seen it suggested - can't remember where, but probably the other thread - that some of the people who would've run for the NDP federally are currently unavailable due to their surprise elections as MLAs.

And to get us back on topic... Leger today has NDP 33%, Lib 28%, Con 27%.

I also saw that, about the MLAs who were likely candidates.

And yeah, to stay on topic here, that poll is not a great start for the CPC campaign. I still think that the NDP will falter though over the longer term, especially when the other two parties set their sights on them.

[octothorp]

Biggest thing about that Leger poll, IMO, is that it confirms what a couple other polls have had recently, that Ontario is a complete three horse race (31, 31, 30).

One of the interesting questions in the Leger poll is something we've speculated on in this forum before:

Q- Does the recent election of an NDP government in Alberta make you more inclined, less inclined, or neither more nor less inclined to vote
for Thomas Mulcair's NDP in the upcoming federal elections?

Naturally, for most Canadians, it's a non-issue, with 52% saying it would not change their voting likelihood. But interestingly, 23% (most in Canada) of Atlantic Canada voters said that Notley's win would make them more likely to vote NDP. Not quite sure how to make sense of responses to that question.

[Iowa_Flames_Fan]

The Leger numbers are interesting, but I think it's worth waiting to see if they are confirmed by any other pollster. Last week there was no-one, I don't think, who had the Tories that low. It's not a big enough variation to call it an outlier, but it does paint a different picture than the roughly 31-31-29 split we are seeing everywhere else.

[Iowa_Flames_Fan]

I just looked at the list of polls on the Wikipedia site above, and realized I was wrong. Leger's numbers are actually pretty consistent with some of the other numbers out there--Forum, for instance, has the NDP at 34 and the CPC at 28.

I don't have a clear recollection--and I'm confident octothorp can set me straight--but am I right in thinking that Forum has a slight liberal house-effect relative to other pollsters, like Nanos and Mainstreet? If that's true, then I think all the polls really are telling us the same story, which is that this is a close election and the NDP has a slight lead.

[Vulcan]

If the NDP are leading in the polls, it's interesting that most of the CPC attack ads are aimed at Trudeau. Maybe they have their own polls saying the Liberals are their main competition or it's from Liberal voters where they have their best chance of attracting votes. I guess it could also mean that they think they can pick up the most seats where they are in a fight with a Liberal contestant.

[LChoy]

Originally Posted by Vulcan

If the NDP are leading in the polls, it's interesting that most of the CPC attack ads are aimed at Trudeau. Maybe they have their own polls saying the Liberals are their main competition or it's from Liberal voters where they have their best chance of attracting votes. I guess it could also mean that they think they can pick up the most seats where they are in a fight with a Liberal contestant.

I posed this question too in the other election thread and I think you are right.
I think even when looking at this board, it's unlikely that an NDP or Conservative supporter would cross over to the other side, their differences are just too great.
However, the ads are targeting the Liberals because their supporters, as well as the undecided, can potentially move over to the Conservative side. It is by attracting these voters that the Conservatives have a chance in winning the election. Also, there are probably conservative voters who may be considering in voting for the Liberals, so these ads hammering the Liberals may help prevent too many from jumping ship

LChoy

[Iowa_Flames_Fan]

I suspect that is Harper's thinking. It is a high-risk strategy, because the risk is you drive voters away from Trudeau and into the arms of Mulcair, but given their low ceiling and low approval ratings, the CPC may not have much choice but to take that risk.

[CaptainCrunch]

My guess is that people who are for the most part the hardcore NDP voters are never going to change their vote, whereas the people that are Liberal supporters are more likely to vote for the PC's if things start falling apart for the Liberals.

So they'll focus on splitting off the Lib vote for now.

[CroFlames]

Originally Posted by CaptainCrunch

My guess is that people who are for the most part the hardcore NDP voters are never going to change their vote, whereas the people that are Liberal supporters are more likely to vote for the PC's if things start falling apart for the Liberals.

So they'll focus on splitting off the Lib vote for now.

That's exactly what got the NDP Official Opposition status. Everyone thought Layton would align himself with the Libs, but instead, he attacked them.

He got official opposition, but Cons got a majority. So I'm not sure if the NDP qualify that as a win or not.

Layton wasn't my favorite politician, but he battled to the end. May he RIP.

[octothorp]

Originally Posted by Iowa_Flames_Fan

I just looked at the list of polls on the Wikipedia site above, and realized I was wrong. Leger's numbers are actually pretty consistent with some of the other numbers out there--Forum, for instance, has the NDP at 34 and the CPC at 28.

I don't have a clear recollection--and I'm confident octothorp can set me straight--but am I right in thinking that Forum has a slight liberal house-effect relative to other pollsters, like Nanos and Mainstreet? If that's true, then I think all the polls really are telling us the same story, which is that this is a close election and the NDP has a slight lead.

Yeah, I could be wrong but I believe that Forum and EKOS are the two that traditionally have a reputation for a liberal house effect, while Nanos is the other extreme. (I'm not sure I'd call Nanos an extreme, as even they tend to underestimate Conservative support, they just underestimate it less than any other pollster).

Regarding the Conservative strategy of staying on the Trudeau message, my guess is that they see themselves as in a lot of two-way battles with the Liberals, and relatively fewer two-way battles with the NDP. So driving voters from the Liberals is going to be the higher-reward strategy, even if they gain less than half of the defecting voters. But it's definitely higher risk, because there is the risk that you also end up turning some of those Liberal/NDP battles into NDP wins, and if it's a race between the NDP and Conservatives to form a minority, this strategy might play into the NDPs hands more. Which is why I think the Conservatives still think they have a shot at another majority.


Actually, that's worth looking into. When I have time, I'm going to go through threehundredeight's riding breakdowns, and see how many battleground ridings he has between each party.

[Jets4Life]

Although I despise Trudeau, I'm leaning towards the Liberal platform.

[CaptainCrunch]

Still not onboard with Trudeau and Mulcair.

Will be interesting to see what happens when more platform details are released.

[octothorp]

I've taken every riding where threehundredeight has the leading party with a less than 70% chance to win, and marked it as a competitive race. If it's got the 2nd and 3rd party within 5% of each other, I've marked it as a 3-way-race. Which are really just arbitrary thresholds, but I had to start somewhere. Here are all the matching competitive races in the country, broken down by province:

BC:
CPC/NDP:6
CPC/LPC:2
NDP/LPC:
NDP/LPC/CPC:4
Other: 1

AB:
CPC/NDP:3
CPC/LPC:2
NDP/LPC:
NDP/LPC/CPC:1

SK:
CPC/NDP:3
CPC/LPC:
NDP/LPC:
NDP/LPC/CPC:

MB:
CPC/NDP:1
CPC/LPC:
NDP/LPC:1
NDP/LPC/CPC:

ON:
CPC/NDP:8
CPC/LPC:20
NDP/LPC:5
NDP/LPC/CPC:5

QB:
CPC/NDP:5
CPC/LPC:
NDP/LPC:5
NDP/LPC/CPC:
Other: 4

NB:
CPC/NDP:1
CPC/LPC:2
NDP/LPC:
NDP/LPC/CPC:


NS:
CPC/NDP:1
CPC/LPC:1
NDP/LPC:
NDP/LPC/CPC:

PEI:
CPC/NDP:
CPC/LPC:1
NDP/LPC:
NDP/LPC/CPC:

NF:
CPC/NDP:
CPC/LPC:1
NDP/LPC:1
NDP/LPC/CPC:

Territories:
CPC/NDP:
CPC/LPC:1
NDP/LPC:
NDP/LPC/CPC:

Total:
CPC/NDP:25
CPC/LPC:29
NDP/LPC:12
NDP/LPC/CPC:10


So, based on that, the Conservatives are involved in a lot of two-party battles with the Liberals in Ontario, but outside Ontario they're actually involved in more battles with the NDP. I would guess that whatever any Trudeau messaging we're getting here, voters in Ontario are being hammered with way more than that. Those 20 seats they're battling the LPC for in Ontario is probably their top focus right now. Also makes sense that when Harper spoke in Quebec he was more focused on the NDP than the Liberals.

As well, this may explain the fact that the Liberals and NDP haven't gone after each other that much. They're not involved in a lot of battles with each other, but have more battles with the Conservatives. For the Liberals, attacking the Conservatives is by far the better policy because any defecting Conservative voters are more likely to go to the Liberals. While for the NDP, not many Conservative voters are likely to switch to NDP, but if they attack the Liberals, they risk driving voters to the PCs, and with the number of PC/NDP races in the country, that's a risky strategy for them.