And Just Like That..... All the Polls Say it's a TIE! Whodathunkit?

[WestCanMan]

@ Left4rds:

Picture a graph representing Candidate A's popularity in a 2-party race, with a line on it that stays consistent at 51-54% over the course of four months. Never really changes. 

Can you weave a story around that to help your candidate? Certainly. You can weave a whopper. Like this:

Your own candidate, Candidate B is at 47%, and the other party's voters are highly motivated. The race isn't even close. Then your guy has a terrible debate, where his dementia can't be hidden, so he has to go.

It's 4 months from the election, everyone knows your guy is leaving, so you put up poll numbers that show the other guy at 70%. HE'S SOARING. It doesn't matter because like I said, the election is 4 months away. Not one ballot is being cast. (remember, the polls are really peaking at 54%, because 45% of voters are cultists who aren't going to budge even if their leader says they need to be nuked). 

Then you bring in a new candidate, and she's a complete dunce with a proven track record of whorishness and failure, but you say "OMG CANDIDATE C IS SO AMAZING!" and then your fake polls come in, and lo and behold "THE POLLS SAY THAT CANDIDATE C IS AMAZING, JUST LIKE WE SAID!!! CANDIDATE C HAS TAKEN THE LEAD, WITH 55%!!! THIS IS ALSO PROOF THAT EVERYONE HATES CANDIDATE A!!!"

(so the polls are at about 53% for Candidate A, but you're workin' it hard) "LOOK AT C JUST CRUSHING IT! SHE COMES FROM A MIDDLE-CLASS NEIGHBOURHOOD, DID YOU KNOW THAT? EVERYBODY LOVES THAT, BECAUSE IT SHOWS THAT SHE CARES ABOUT THEM!" Awwwwww.

Now as you get closer to the election, you have to dial in your accuracy so that you still have credibility for the next election cycle, so you say 53-47 for C a week out. Then by election night you say 51-49 for C. Then the election goes like this:

"Oh, wow. Our polling was just off by 1.5% for Candidate A and 3.1% for Candidate C. That's pretty good. Well within the margin of error, on average. Clearly what happened on election night is the misogynists and racists came out in force, so the really nice people were scared to go to the polling stations."

The MSM got to use fake polling to back their massively false narratives and pump up Candidate C for 4 months. 

 

CBC: "You lead those dumb-ass horses to water if you want to WCM, but they ain't gonna drink it, cuz these horses only drink Kool-aid 😉."

Edited Wednesday at 12:44 AM by WestCanMan

[CdnFox]

4 minutes ago, WestCanMan said:

@ Left4rds:

Picture a graph representing Candidate A's popularity in a 2-party race, with a line on it that stays consistent at 51-53% over the course of four months. Never really changes. 

Can you weave a story around that to help your candidate? Certainly. You can weave a whopper. Like this:

Your own candidate, Candidate B is at 47%, and the other party's voters are highly motivated. The race isn't even close. Then your guy has a terrible debate, where his dementia can't be hidden, so he has to go.

It's 4 months from the election, everyone knows your guy is leaving, so you put up poll numbers that show the other guy at 70%. HE'S SOARING. It doesn't matter because like I said, the election is 4 months away. Not one ballot is being cast. (remember, the polls are really still at 52%, because the 48% that are cultists aren't going to budge even if their leader says they need to be nuked). 

Then you bring in a new candidate, and she's a complete dunce with a proven track record of whorishness and failure, but you say "OMG CANDIDATE C IS SO AMAZING!" and then your fake polls come in, and lo and behold "THE POLLS SAY THAT CANDIDATE C IS AMAZING, JUST LIKE WE SAID!!! CANDIDATE C HAS TAKEN THE LEAD, WITH 55%!!! THIS IS ALSO PROOF THAT EVERYONE HATES CANDIDATE A!!!"

(remember, the polls are really still at 52% for Candidate A) "LOOK AT C JUST CRUSHING IT! SHE COMES FROM A MIDDLE-CLASS NEIGHBOURHOOD, DID YOU KNOW THAT? EVERYBODY LOVES THAT, BECAUSE IT SHOWS THAT SHE CARES ABOUT THEM!" Awwwwww.

Now as you get closer to the election, you have to dial in your accuracy so that you still have credibility for the next election cycle, so you say 53-47 for C a week out. Then by election night you say 51-49 for C. Then the election goes like this:

"Oh, wow. Our polling was just off by 1.5% for Candidate A and 3.1% for Candidate C. That's pretty good. Well within the margin of error, on average. Clearly what happened on election night is the misogynists and racists came out in force, so the really nice people were scared to go to the polling stations."

The MSM got to use fake polling to back their massively false narratives and pump up Candidate C for 4 months. 

 

CBC: "You lead those dumb-ass horses to water if you want to WCM, but they ain't gonna drink it, cuz these horses only drink Kool-aid 😉."

Yep. 

Generally speaking accurate polling will reflect in certain other areas as well if one party is genuinely doing better than another party it tends to show up in other activities and things that go on.

When there is a significant disconnect between the polling you're seeing and other indicators that is wise to question. That does not always mean that the polls are wrong by any stretch, and it's quite possible that the polls are correct. But it's usually a sign that something is off.  And this election cycle it certainly seems like there is a strong disconnect between many of the polls and other indicators such as union endorsements, the motivation of the voting blocks, the turnout at the rallies, the performances at debates etc. 

And that sudden tightening in the last week is another indicator. We may very well find that this turns out to be one of Canada's famous surprise election results.

 

[WestCanMan]

3 minutes ago, CdnFox said:

Yep. 

Generally speaking accurate polling will reflect in certain other areas as well if one party is genuinely doing better than another party it tends to show up in other activities and things that go on.

When there is a significant disconnect between the polling you're seeing and other indicators that is wise to question. That does not always mean that the polls are wrong by any stretch, and it's quite possible that the polls are correct. But it's usually a sign that something is off.  And this election cycle it certainly seems like there is a strong disconnect between many of the polls and other indicators such as union endorsements, the motivation of the voting blocks, the turnout at the rallies, the performances at debates etc. 

And that sudden tightening in the last week is another indicator. We may very well find that this turns out to be one of Canada's famous surprise election results.

What I don't see is legitimate reasons to like Carney, or to trust him, and I don't see Canadians being warm to him. 

There are now half a dozen solid reasons to consider Carney a massive liar, he's a global elite with no track record of working for the little guy, and he's not even a Canadian. He actually calls himself a European when he's in Europe. 

Aside from fake polling and the CBC saying "Carney is awesome", no one here can come up with any reasons to say what's awesome about him. And they won't even try to defend his integrity. He's a total zero in that regard.  

[CdnFox]

10 minutes ago, WestCanMan said:

What I don't see is legitimate reasons to like Carney, or to trust him, and I don't see Canadians being warm to him. 

There are now half a dozen solid reasons to consider Carney a massive liar, he's a global elite with no track record of working for the little guy, and he's not even a Canadian. He actually calls himself a European when he's in Europe. 

Aside from fake polling and the CBC saying "Carney is awesome", no one here can come up with any reasons to say what's awesome about him. And they won't even try to defend his integrity. He's a total zero in that regard.  

The real head scratcher is the ndp support that it supposedly flooding to him. I get that trump scared them and they are abandoning the whole third party premise to try and get someone elected, but they're going for a rich industrialist banker and capitalist whose job was to evict people and hide taxes from the poor? That's their big play here?

Edited Wednesday at 01:09 AM by CdnFox

[SpankyMcFarland]

18 hours ago, August1991 said:

These Anglo guys can speak French like Turner - but not like Mulroney.

Mulroney grew up in a French-speaking town. It would be weird if his French wasn’t better than these guys. 

[SpankyMcFarland]

29 minutes ago, WestCanMan said:

CBC: "You lead those dumb-ass horses to water if you want to WCM, but they ain't gonna drink it, cuz these horses only drink Kool-aid 😉."

So are pollsters deliberately falsifying their data in Canada? That would leave a yawning opportunity for firms that aren’t. 

[CdnFox]

54 minutes ago, SpankyMcFarland said:

So are pollsters deliberately falsifying their data in Canada? That would leave a yawning opportunity for firms that aren’t. 

It would be no opportunity at all for those who aren't. First off It would be hard to prove that people didn't just change their minds. Second off as long as the final days prediction is accurate they will be able to advertise next year that they were very accurate all along even if that's not true. 

As to intent, a few may be putting their thumbs on the scales.  Definitely ekos is doing it deliberately. Their polling asks questions that are absolutely horrible about the CPC before asking the polling question, like basically "is PP a horrible person or just an unmitigated fool.  a) horrible person b) unmitigated fool c) none of the above", that kind of question. Not even kidding, they're that bad.  That skews the odds. 

Also how they choose to 'weight' their polling radically changes the outcome so if they make minor tweaks they can easily case the liberals in a better light. 

Honestly I think for a number of them the problem is strictly in their methodology, it's getting harder and harder to come up with affected methodology that you can rely on. Coupler clearly skewing the odds but for the most part I think it's less about intent.

They're also seems to be problems with these running averages polls. The theory was that if you pull 500 people a day and then average the results over 3 days dropping the third day with the results for the next day then you should theoretically have a running average with a sample size that would be equivalent to 1500 people. Nanos has been doing this for several elections now. But it doesn't seem to be all that accurate for some reason, it doesn't seem to track the trends as well as a 'normal' poll does.  But nobody can afford to do a 1500 person poll every night. 

On the last day or three all the pollsters will pull out all the stops and spend the money to  be as accurate as possible so they can say they 'accurately guessed ' the polls. We'll see how close their 'final' prediction is to some of their campaign predictions. 

[WestCanMan]

53 minutes ago, SpankyMcFarland said:

So are pollsters deliberately falsifying their data in Canada? That would leave a yawning opportunity for firms that aren’t. 

A yawning opportunity for them to do what, exactly? Get ignored by the MSM outlets that got hundreds of millions of dollars from Trudeau, courtesy of the taxpayers? 

They don't have to falsify their data at election time, they can dial it in close for election day, as they always do. "The Libs are just a little bit ahead now in the popular vote..." is the election eve quote, followed by "Awwwww, the pollsters were so close. the CPC was up by 2%. But the pollsters really noticed how the CPC were closing in at the end there 😉😉."

**The job of the pollsters that CBC chooses to give credence to is to have polls that back CBC's narratives**

When it's 1 month from election time, do you think CBC wants to say:

• "Poilievre's campaign has all the momentum right now, his message is clearly resonating with mainstream Canadians", or
• "Poilievre is still only appealing to the hard-core conservatives: minorities and people in the centre are choosing the LPOC candidate because they think he is better at the thing on this other chart. Look at the gap in the polls on this subject, which voters consider the key to this election." Then they'll point to a chart that says "Everyone thinks that this guy will be better at dealing with that thing". And not one bit of it has to be true. CBC just needs to able to speak glowingly of the LPOC candidate, and their policies, and polls are the key to doing that. "The cool kids all think that our, I mean the LPOC candidate is better at this."

Do you think that CBC wants to admit that Carney's constant, massive lies are having an effect on polling? Are they gonna allow a legitimate poll that actually shows "which candidate voters see as having the most integrity"? And then have a discussion about why that would be the case? 

They'll talk about straws, abortion, global warming, how Carney singlehandedly saved Canada in '08 while Harper and Flaherty were standing there in awe, etc. The topic of Carney's laundry list of lies never made the debates and it won't be a topic in the news either. Nor will the Chinese gov't's support, the Lib MPs who worked with China, etc, etc. That's all backburner sh1t. Straws, man. This election is about straws now. 

CBC News page right now:

There's actually nothing at all about Carney on that page right now. Nothing. 

If you scroll down it's more not-so-positive stuff about Piolievre:

[TreeBeard]

1 hour ago, SpankyMcFarland said:

So are pollsters deliberately falsifying their data in Canada? That would leave a yawning opportunity for firms that aren’t. 

Every single poll had the Cons up by 20+ points a few weeks ago.   It’s rather funny for people to think they were accurate then, but fake now. 

[WestCanMan]

Top story about Carney right now on CBC:

 

 

 

[WestCanMan]

6 minutes ago, TreeBeard said:

Every single poll had the Cons up by 20+ points a few weeks ago.   It’s rather funny for people to think they were accurate then, but fake now. 

It was all part of the plan to make it look like bringing in Carney was a massive change. 

LPOC with Trudeau: 😭🤯

LPOC with Carney: 💪🥰

"OMG, LOOK! IT'S A TOTALLY DIFFERENT PARTY NOW! 100%!!! I DON'T EVEN RECOGNIZE ANYONE!!!"

[CdnFox]

11 minutes ago, TreeBeard said:

Every single poll had the Cons up by 20+ points a few weeks ago.   It’s rather funny for people to think they were accurate then, but fake now. 

If by a few weeks ago you mean a few months ago that might be accurate. A few weeks ago they had the liberals up.

And frankly if you look back a few weeks ago I was saying the polls didn't seem accurate then either so that kind of shoots your whole theory down   

[CdnFox]

8 minutes ago, WestCanMan said:

It was all part of the plan to make it look like bringing in Carney was a massive change. 

LPOC with Trudeau: 😭🤯

LPOC with Carney: 💪🥰

"OMG, LOOK! IT'S A TOTALLY DIFFERENT PARTY NOW! 100%!!! I DON'T EVEN RECOGNIZE ANYONE!!!"

wimpy wimpy wiiimpy  HEFTY HEFTY HEEEFTY!!! 

[CouchPotato]

1 hour ago, CdnFox said:

As to intent, a few may be putting their thumbs on the scales.  Definitely ekos is doing it deliberately. Their polling asks questions that are absolutely horrible about the CPC before asking the polling question, like basically "is PP a horrible person or just an unmitigated fool.  a) horrible person b) unmitigated fool c) none of the above", that kind of question. Not even kidding, they're that bad.  That skews the odds.

I've seen people posting online the Ekos is asking people their vax status.

[Moonlight Graham]

Polls show the CPC is getting their butts kicked in Ontario and QC.  It's over.

CPC even won the most votes the last 2 elections but their vote distribution is not efficient, lots of blowouts in the prairies.

Apparently 7 million votes were cast in advance polls.  They could count those and we'd already know the winner.

Edited Wednesday at 06:49 AM by Moonlight Graham

[Michael Hardner]

3 hours ago, Moonlight Graham said:

Polls show the CPC is getting their butts kicked in Ontario and QC.  It's over.

CPC even won the most votes the last 2 elections but their vote distribution is not efficient, lots of blowouts in the prairies.

Apparently 7 million votes were cast in advance polls.  They could count those and we'd already know the wwinner

You're probably right, but there's still the  chance that CPC voters aren't responding to polls.

9 hours ago, SpankyMcFarland said:

So are pollsters deliberately falsifying their data in Canada? That would leave a yawning opportunity for firms that aren’t. 

338 publishes the accuracy of their model for past elections. It was pretty close.

Edited Wednesday at 10:47 AM by Michael Hardner

[myata]

Sure pigs and cows are flying high in the maga-conservative alt universe. To the dumb infinity and way beyond!

[WestCanMan]

3 hours ago, myata said:

Sure pigs and cows are flying high in the maga-conservative alt universe. To the dumb infinity and way beyond!

Even in the age of fibre-optics and 5G, your posts are a serious waste of bandwidth. 

[WestCanMan]

7 hours ago, Moonlight Graham said:

Polls show the CPC is getting their butts kicked in Ontario and QC.  It's over.

If so, then Trump will have won two elections in just 5 months. 

[Moonbox]

15 hours ago, CdnFox said:

Here you go   This explains clearly why you effed up.  You go the math wrong   

Understanding the margin of error in election polls

It is not enough for one candidate to be ahead by more than the margin of error that is reported for individual candidates (i.e., ahead by more than 3 points, in our example). To determine whether or not the race is too close to call, we need to calculate a new margin of error for the difference between the two candidates’ levels of support. The size of this margin is generally about twice that of the margin for an individual candidate. 

Kudos to you for actually citing something specific and directly for a change, even if it's not great.  At least it's real, and it's true for the USA, just not Canada:

While this is the correct conclusion when there are only two possible survey responses, it is not correct when there are more than two possible responses, which is in fact virtually always the case. How much difference this makes depends on how many responses are outside the two categories of interest.

-Charles H. Franklin University of Wisconsin, Madison October 27, 2002 (Revised, February 9, 2007) 

(It's off a PDF that I don't know how to hyperlink) 

For a 5.4% lead to be a a "statistical tie", you'd need the Liberal vote to be overestimated by at least half that, and for PP to be the 100% beneficiary of it (which is already unlikely).  You also need the NDP/Green/Bloc numbers to be (at minimum) almost dead-on.

As before, you're assuming all of the variability leans heavily in the Conservatives' favor, and that the Bloc/NDP/Greens somehow don't exist.  

TLDR:

Sorry kiddo/muffin , nice try, good effort, but you're still wrong.  

Edited Wednesday at 02:56 PM by Moonbox

[SpankyMcFarland]

Given that this geezer golf ad does not feature Poilievre, the Tories’ own polling must show that older men aren’t as keen on them or the leader as they usually are. 

https://www.cbc.ca/news/politics/last-week-ad-campaigns-conservative-ads-no-pierre-poilievre-1.7514897

 

[Moonbox]

25 minutes ago, SpankyMcFarland said:

Given that this geezer golf ad does not feature Poilievre, the Tories’ own polling must show that older men aren’t as keen on them or the leader as they usually are. 

https://www.cbc.ca/news/politics/last-week-ad-campaigns-conservative-ads-no-pierre-poilievre-1.7514897

 

Yeah those kept coming up during the Leafs game.  Seems like the Conservatives are finally acknowledging him as the liability he is, and put him in the candidate protection program.  😐

[CdnFox]

2 hours ago, Moonbox said:

Kudos to you for actually citing something specific and directly for a change, even if it's not great.  At least it's real, and it's true for the USA, just not Canada:

Which i do all the time, except those times you cite something that proves you wrong first   I know how much you love to pretend otherwise but fact is i cite more than you. Except when you've provided one that proves you're wrong. 

Quote

 

While this is the correct conclusion when there are only two possible survey responses, it is not correct when there are more than two possible responses, which is in fact virtually always the case. How much difference this makes depends on how many responses are outside the two categories of interest.

-Charles H. Franklin University of Wisconsin, Madison October 27, 2002 (Revised, February 9, 2007) 

 

Not really relevant is it.

(

Quote

It's off a PDF that I don't know how to hyperlink) 

I'll spot you this one.

Quote

For a 5.4% lead to be a a "statistical tie", you'd need the Liberal vote to be overestimated by at least half that, and for PP to be the 100% beneficiary of it (which is already unlikely).  You also need the NDP/Green/Bloc numbers to be (at minimum) almost dead-on.

 

No, that's not how it works. Sigh. I posted a detailed explanation of this. 

The number given to either party is the center of a range of likely probabilities based on the data.  It is just as likely that the actual number is in the upper portion of that range, or the lower.  

if two ranges overlap then it is JUST as likely that one is at the top of the range and one at the bottom.  Which is why we call it a 'statistical' tie, rather than an actual tie which is what we call it when they BOTH have the same probability numbers. 

So ALL that's required is that the ranges overlap. And this has been the convention for many many years. 

 

Quote

As before, you're assuming all of the variability leans heavily in the Conservatives' favor, and that the Bloc/NDP/Greens somehow don't exist.

There is no 'assumption'.  You still don't get how this works. 

As posted above, the margin of error applies to each number and is the range of possible outcomes based on the data. If those two overlap at all then it is a statistical tie.  

When the margin of error of two estimates overlaps, it suggests a statistical tie, meaning the difference between the two is not statistically significant. This is because the overlap indicates a possibility that the true values could be the same. Specifically, when the 95% confidence intervals of two means overlap, it suggests that the difference between those means is not statistically significant,

TLDR:

Moonbox, you're not only a loser, you're one of the most persistent and thorough losers I've ever met   Well done  

 

Your comments boil down to one thing. You did not understand what you were talking about. And I did so I knew I was right. I would have thought that after all these times looking foolish you would have looked it up first but whatever.

You can make an argument that technically speaking if you wanted to hire a lawyer the margins don't truly overlap they are just touching each other but for the purposes of the discussion and the point being made it would be utterly childish and pathetic to try and make that argument.

Better luck next time, i'll probably see you in the next 10 threads where you try and jump in and salvage your ego by attacking me randomly

[Moonlight Graham]

6 hours ago, myata said:

Sure pigs and cows are flying high in the maga-conservative alt universe. To the dumb infinity and way beyond!

All the choices this election are nothing short of terrible.  I'm hoping for a minority government.

[taxme]

On 4/21/2025 at 10:36 PM, CdnFox said:

even posters like ecos with the absolutely detestable Frank Greaves who swore he would use his powers to defeat poilievre now show the liberals and conservatives at basically a tie after weeks of insisting that there was double digit spreads between them

I don't know why but a lot of posters this election put their thumbs on the scales I think. You look at their methodologies and you look at what they said and there were a few that were fairly honest but most were less so.

As I predicted a few days ago now that we're getting close to the end they're all going to pretend that they're pulling is saying it's magically a tie. Last night, everybody changed their minds and that hugely leads vanished and wow it's actually a tie now who'da thunk it

This will allow them to pretend that they were at least close no matter who wins.

It's pretty obvious from some of the commentary that they believe that the conservatives are trending right now and have momentum. The poles are currently showing a liberal minority. I have a funny feeling we're going to see a conservative minority in reality but you never know.

If that happens I expect that mark carney will drop out of politics and go do something else despite having promised that he was in it for the long haul win or lose. Hopefully at that point the conservatives will call a snap election and say seeing as they got A free ride in the last leadership race there's no point in waiting for them to finish this one  

The polls in Alberta show that 30% of Albertans are ready to separate from Canada. If the liberals win, even a minority, it is believed that the number will go up higher. If the liberals become a minority, then the block will surely back them up.

After all, both party's do hate Canada and both are ready to get together to screw the conservatives anyway they can. We must get a majority conservative government or the conservatives will have a tough time trying to get thru some of their agendas. Just saying.