Federal polling - started March 2, 2011

105 posts / 0 new
Last post
Federal polling - started March 2, 2011


2011 Federal Polling

Date / Pollster / Cons / NDP / Libs / Bloc

Jun 04 / GE / 29.6% / 15.7% / 36.7% / 12.4%

Jan 06 / GE / 36.3% / 17.5% / 30.2% / 10.5%

Oct 08 / GE / 37.7% / 18.2% / 26.3% / 10%

Feb 14 / EK / 37.3% / 14.2% / 24.8% / 9.3%

Feb 14 / IR / 39% / 18% / 25% / 9%

Feb 15 / HD / 37% / 14% / 27% / 10%

Feb 20 / NR / 39.7% / 18.9% / 26.6% / 9.9%

Feb 23 / EK / 32.4% / 14.8% / 27.3% / 10.5%

Feb 25 / AD / 38% / 19% / 23% / 11%

Feb 26 / VC / 39%/ 18% / 26%/ 11%

Mar 1 / IR / 43% / 13% / 27% / 10%

Mar 2 / HD / 36% /15% / 28% /             New today



Boy oh boy: Allan Gregg just won't quit going after his rivals, comparing them to Suncor denying the ducks got covered in muck.



But Gregg slammed the association for effectively implying the polling industry is 100 per cent accurate, all the time.
"Their response is we have no problems whatsoever," he said.

"That's like Suncor saying, 'No, those ducks didn't go into the tailing pond or they just got a little dirty when they went in there.'"
Gregg said the proliferation of sometimes conflicting polls and the hypeventilating analysis that frequently accompanies them does not strengthen democracy. On the contrary, he said: "Rather than have a public that's informed, you have a public that's misinformed."

He said he's not arguing that polls should be ignored; only that their import needs to be interpreted much more cautiously. Rather than pontificate on weekly fluctuations in individual polls, he said it makes more sense to average the results of various surveys and look at the trends over longer periods of time.

"It's not so inaccurate that we should be just saying, 'Oh, polling is in complete disrepute and the discipline is so fatally flawed that we should never look at polls."


From OO's above article

"While I would love to claim that negative advertising, charges by Elections Canada against Conservative officials or 'the Oda effect' is having a great impact on Canadian voter preference, we believe that within the bounds of statistical error, there has been no change this week," he said.

Lens Solution

In that article, pro-Conservative pollster Darrell Bricker/Ipsos-Reid refuses to accept any criticism.  It is time that journalists begin to question Mr. Bricker about his suspicious polls, particularly his bizarrely low NDP numbers.


In the last polling thread, Stockholm mentioned that Ipsos no longer prompts with the Green Party in their polls. Well...won't over 95% of Canadians be "prompted" with the Green Party's name when they see it on the ballot when they go into the voting booth?


It is a different psychologocal and cognitive dynamic, taking place within radically different contexts, even if it sounds the same.

Only political junkies and other of the more engaged folks behave the same in both circumstances.

Sean in Ottawa

Increasingly people hate polls and know that for that 10-20 minutes of their lives nothing will change other than to influence a poll. By far most people refuse to do polls in one way or another. Depending on how you count them between 95% and over 99% refuse. Thousands of numbers get dialed in order to get a small sample of results. I am increasingly convinced that it is the political junkies that do polls and they are not representative of the wider population.

This has become so obvious that there is a fight in the industry between those who want to acknowledge the problem and claim polls are still okay and those so scared they don't even want to admit this. Polls are increasingly ridiculous with now huge swings only a week apart that the pollsters to protect their fading credibility strain to explain.

Boom Boom Boom Boom's picture

Why not just ban polling altogether? (I know, I know, easier said than done Frown )

Sean in Ottawa

Two responses-- there are uses in developping policies and making political parties more responsive to citizens. That is an argument to only ban public polls. The trouble is we would not want the parties having information like this that is not disclosed. So no we can't ban them.

It might be possible to set a minimum standard for public polls that is very high-- in terms of sample size and methodology. Still there are issues there as much of this is proprietary. But we could require minimum sample sizes for political polls that reduce the number of huge MOE polls that seem to be popular now. We could also require reexamination of margin of error as I assume these are using out-of-date methods and likely due to changes we have spoken about here the MOE may be much higher than it used to be.


In this day and age of blogs and 24 hour a day news cycles - polls are here to stay and no one is ever going to put toothpaste back into a toothpaste tube. In almost every case, the polling companies themselves put out press releases that give all kinds of information about sample sizes and margins of error etc...its not their fault if journalists then choose to ignore it all and just run with some splashy headlines. I suppose you could demand that the media observe certain rules about reporting polls - but look at all the other things that are reported on with no rules. Routinely a majotr demonstration happens and some media reporst say there were 100,000 people there and other say there were 10,000 - do we pass a law saying it is illegal to estimate the size of a crowd? Also, what happens in the blogosphere - if i am a private citizen, am I not entitled to look at a poll and interpret it any way i want and post my ideas wherever I want - and others are free to agree or disagree?

I really don't see what the problem is anyways. To me the REAL problam is that the calibre of political reporting in Canada is at an all-time low and most journalists who cover politics don't know what they talking about - and the way they cover polls is just one of many manifestations of this. So, maybe the real solution is to pass a law stating that only CERTIFIED Canadian political journalists should be allowed to report on canadian politics and to be certified they must pass an exam where they demonstrate among other things that they understand how to report on public opinion polls.


Sean in Ottawa wrote:

Depending on how you count them between 95% and over 99% refuse. Thousands of numbers get dialed in order to get a small sample of results.

...and yet in almost every case, the final polls before election have each party pegged to within about 1% of what it gets at the ballot box - so clearly that 5% of the population who took part in the poll turned out to have opinions that were virtually identical to those of the 95% of the people who refused.


I do agree that the larger problem is the calibre of political reporting, though part of that is their obsession with "what the polls are saying" to the exclusion of most everything else.

For example, the latest pair of polls which are causing everyone to despair, were in the field up to Sunday. Yet, they're offered as proof that nothing can unseat the Conservatives. I for one would like to see what happens after a few weeks of the NDP ads being in rotation, and news about the in-and-out charges starts to settle in to the general understanding.


Agreed that for all the problems with the polls themselves, it is the political reporting that is so appaling. Quoting pollsters who havent a clue what they are talking about is just part of the flow. I find that the same pollsters being quoted as airheads, nonetheless do at least serviceable work.


Greg L. is far and away the pollster with the broadest knowledge of politics- and he says the least in public.

I can think of at least one other senior polling person who is the same, and I have heard quote once in about 10 years.

Then there is that pollster the Liberals use- M....  You hear his work being referenced, but not him talking. Strong hunch he has a solid background in politics and campaign dynamics.

There is a pattern here.

Boom Boom Boom Boom's picture

I supposed polling is good for discovering trends in voting intentions, but I still don't take them very seriously.


Michael Marzolini does polling for the federal Liberals - I think he would be only too happy to shoot his mouth off with half baked theories - but one of the pre-conditions of being a party pollster is absoluet discretion and a non-existent media profile. If he started talking to the press, the liberals would probably fire him and the same goes for whoever polls for the other parties.


Still you can get a pretty good clue about what they're thinking at Pollara if you follow the Calgary Grit. For one thing, you can tell that they've already build a demographic profile of the typical Liberal voter vs. NDP voter vs. Conservative voter from the last election.

Sean in Ottawa

Stockholm wrote:

Sean in Ottawa wrote:

Depending on how you count them between 95% and over 99% refuse. Thousands of numbers get dialed in order to get a small sample of results.

...and yet in almost every case, the final polls before election have each party pegged to within about 1% of what it gets at the ballot box - so clearly that 5% of the population who took part in the poll turned out to have opinions that were virtually identical to those of the 95% of the people who refused.

I have worked for a pollster doing the final polls. It is like night and day. Those are the polls with accountability. Everyone is told future contracts and their jobs are on the line. The don't take the short cuts with sample--- they put their best people (the worst people do political polling at all other times because even though more people refuse political polls than ever before it is even harder to get people to do a poll on shampoo -- and client supplied sample is more critical and in some cases specialized expensive sample is required). They add depth to the survey to get closer usually much higher numbers are surveyed but it is the quality that is the huge change. More calls get listened to, better training etc.

Do not compare those final polls with the crap we get now. No comparison. And they just explain the difference away with some story about how things changed during the campaign. That might be the case but it doesn't have to be.

then there is the objective-- people knowing there is no accountability use various techniques to bias polls that are not used for final polls when the pollster wants them accurate. At other times they will accommodate a client that desires a specific result (and that is a gray area with the more ethical ones making fewer concessions to fair polling and the less ethical doing whatever they have to to get the contract.

In a nut shell-- I believe the final polls mostly but have no faith in the crap sold now. And I have seen up close the difference in how they are done.

Boom Boom Boom Boom's picture

The way the Cons are fumbling, they may get down to 30% yet! One can only hope.

Lens Solution

Hopefully the Cons will drop in the polls over all the scandals this week, although according to some of the commentators, Canadians only care about the economy and aren't interested in any of these other issues.


308.blogspot.com has developed a new projection model which comes up with significantly different seat estimates from its former model.




The writer admits that he previously significantly overestimted Liberal projections at the expense of the NDP whose seat estimate rises from 20 to 31.


He also presents vote estimtes for every individual riding.




Which just goes to show what a huge difference it makes to tweak the assumptions built into the model.

There are two ways to look at this: the search for the holy grail of the thoroughly grounded model, or that the whole exercise has inherent fatal flaws.

To my mind, the best take on such quests is Umberto Eco's narrative that focuses on the search for Prester John. the name of the book escapes me at the moment. Bodalino?

Life, the unive...

nicky wrote:

308.blogspot.com has developed a new projection model which comes up with significantly different seat estimates from its former model.




The writer admits that he previously significantly overestimted Liberal projections at the expense of the NDP whose seat estimate rises from 20 to 31.


He also presents vote estimtes for every individual riding.



This just shows the absolute foolishness of these kinds of projects when they are done remotely with no attention or knowledge of actual ridings.

I'll use two ridings I know well.  Huron-Bruce and Bruce-Grey-Owen Sound.  This nonsense has the Conservatives as almost equal in strength (a 2 point difference) despite the reality that the BGOS Con MP is very popular and the HB Con MP is a non-entity and at best most people are indifferent to him.   And it just gets worse from there.

The 'analysis' has the Liberal candidate in HB at 33 %.  One assumes that this is based on past performance.  But let's look at the reality on the ground.  First we had a popular Liberal incumbent MP.  Then in the last election the Liberals ran a really well known and liked candidate.   This time however the Liberals are running a very weak, invisible candidate who in the most recent municipal election had a record number of voters come out and voted him out of the mayor's chair by a margin of almost 3-1.  Not a lot of strength there.  Last election the NDP ran a lovely, well-meaning but very fringish candidate.  This time though the NDP has a very well known, high profile candidate who was on radio every week here and featured regularly in our weekly newspapers while the Ontario president of the NFU (Ontario's fastest growing farm organization).  That coverage has mostly continued and just this week he was in the news at least twice.  Suggesting he is under 13 per cent is frankly unbelievable- as many Liberals I know are saying they will be voting for him (not NDP mind) because he represents their values and they are so unispired by the Liberal canddiate.  He will by the end of the campaign be challenging for the riding I have no doubts because he takes votes away from both the Liberals AND the Conservatives.  As it is I expect if a poll was done of the riding that included local names he would be comfortably in 2nd with the Cons clearly in sight.

Over in BGOS in the last election the Greens ran a well known local personality.  This time out though they are running someone who is clearly talented, but has zero profile.   So suggesting a 27 % vote for the Greens is very unlikely.  The Liberals though have gone from a hopeless sad-sack candidate last time to a crackerjack who has been working the riding really hard.  Her profile rivals the MP and she gets frequent media coverage, including a front page story in the only riding weekly last week.   Suggesting the Liberals will come in at 15 per cent is as much nonsense as the prediction for the NDP in HB.  Again if a poll was done of this riding with local names the Liberals would be very comfortably in 2nd, although a fair ways back of Miller.

These types of exercises are worth than useless, because they are profoundly misleading and poorly executed.  You simply can not predict riding by riding outcomes based on past results and broadband polling numbers.  It is trash pure and simple.

Arthur Cramer Arthur Cramer's picture

I don't like this 308.com. I think it has its own agenda. LTU, you are right to distrust this whole process. I think these are all focused on manipulationg results to create impressions. I wish this projection and polling stuff would go away. It does a disservice to the national debate.

Arthur Cramer, Winnipeg


Of course every individual riding has its own context and strong and weak candidates etc...but I don't think 308.com is claiming or even trying to accurately predict the result of each individual riding in Canada. It is trying to create a model that shows how various popular vote scenarios might produce various seat distributions. Of course some ridings will swing much more strongly than the province-wide swing and some won't swing and at all - but usually these outliers cancel each other out. So, what's the harm in it. We all gossip and speculate about which party will win which seat etc...308 is trying to use some more objective factors (ie: polling data). So what? He has as much right to do that as any of us.

Life, the unive...

No Stockholm they are trying to predict the results in individual ridings.  That is what the link goes to.  It is harmful because some idiot(s) will grab onto it as some proof that we must have strategic voting, they will get it all wrong based on these models and we will end up with progressive politics being the loser.  It happens every single frikkin time.  That's the problem.  Not whether he has the right to post as if he knows something he doesn't.  Weren't you just bemoaning the lack of quality political journalism in Canada- well here's a part of it.


The link shows what the model "projects" in each riding to add up to the overall predicted total. But so what - we all prognosticate on election results anyone is free to start their own blog. Before 308, there was electionprediction.org - where some Liberal staffer "predicted" results in each riding based on gossip submitted to the site (most of which was clearly written by staffers from all three parties!).

"Strategic voting" is nothing new in Canada. Its been around for as long as there has been more than 2 parties (ie: almost 100 years). It usually takes the form of - I can't vote for so and so because they have no chance. In many cases the NDP benefits from tactical voting - it helps Linda Duncan win in Edmonton and we are only too happy to tell people thinking of voting Green to vote NDP instead since the Green party has no chance of winning any seats. So what? Whatever information is out there it there for people to consider when they decide how to vote. Whether its a seat projection or just national polling horse race numbers - there will always be people who will not vote for one party or another because they think it can't win in their area or nationally.

If people DO want to vote strategically, then i hope that they do study 308.com because it clearly shows that there are dozens and dozens of ridings where the NDP is the "strategic choice". Better that than not having this tool and being left with the message of "just vote Liberal no matter where you live".

Life, the unive...

and we wonder why people give up on the NDP in an awful lot of ridings. 


There are lots of reasons why potential NDP voters "give up" on voting NDP. It could be because the NDP has no active campaign in their riding and they get no calls, no door knocks and no literature. It could be because they look at the results of the last election and figure" hmmmm...I live in Thornhill, I wonder how the NDP did here in the last election - OH! they got 4% of the vote - well so much more that!" (what's your solution to that - making it illegal for anyone to look up past election results in any ridings?). People might also look at national polls and assume that since the NDP is at 18% nationally - it has no chance PERIOD no matter where you live. Or they might react to pundits on the At Issue panel telling them that the NDP can't win etc...

In the end, voters have access to information - there is no point kvetching about that. There is no way that we can put 30 million Canadians into solitary confinment for the duration of a campaign and brainwash them from any knowledge of past election results, and stop them from reading any polls or seat projections or any punditry speculating on the results and have them just spend 35 days doing nothing but watch free time political broadcasts from each party and candidate. it ain't gonna happen. So live with it.

In any case, I suspect that the only people who ever even look at 308.com are a couple of hundred political junkies who already know how they are going to vote anyways - so why bother complaining. I will leave it to mathematicians or amateur psephologists to critique the guys'a actual model - but I honestly don't think that he's purposely conspiring to make any party look good or bad. If the NDP suddenly soared to 30% in a series of national polls - all of sudden we would all FALL IN LOVE with seat projection models like 308 wnhich would suddenly start projecting over 100 seats for the NDP.

Sean in Ottawa

Sorry Stock-- I too see the difference between past results and projected current ones based on current polling.

Flawed methodology built on flawed methodology.

By posing as science this makes it worse.

Let people look at last results because they know it is last results-- this hybrid Frankenstein partly based on past results and partly on cruddy polls with huge margins of error claiming to be the current situation is a big problem for me.

If people would see it for what it is that would be fine-- something akin to horoscopes provided for entertainment but presented as any more than that it is dangerous deceptive and leads to continued warping of our electoral system.


Its not actually based on "huge margins of error" since he is aggregating dozens of polls which makes the sample by region quite robust. In any case, i don't know what everyone's complaining about - the 308.com projections are actually quite favourable to the NDP these days. According to him, the latest Harris Decima poll would give the NDP 35 seats - even though it has the NDP at 15% which is a 3 points drop from the last election. So, i say don't look a gift horse in the mouth!

In any case a seat projection is only as good as the validity of the polling data that you feed into it - if you think that the polls are inaccurate, then fine - we are back to bitching about the fact that people talk about polls (including all of us) - what are you going to do about it. Are you going to suggest that people be banned from babble if they dare to write something like "gee, looking at the last couple of polls it looks like the NDP is averaging 20% and by my reckoning that will probably yield about 45 seats"

There is nothing new about seat projections - its been a feature of every election night in the UK since the 1950s to have a "swingometer" that tries to extrapolate the average swing from polls in terms of how many seats are likely to change hands. i just don't see what's wrong with people playing this harmless parlour game.

I'd be surprised if the existence of 308.com results in one single solitary person in Canada voting any differently.


The BBC's UK election swingometer is light years ahead of anything we have in Canada. The tv networks in the UK have impressive graphics showing swingometers that indicate the marginal seats between different parties. When the polls close the swingometer quickly shows which parties are gaining seats from the last election and which ones are losing.  When I've watched BBC election night telecatsts I've wondered why we don't have these sophisticated metric tools in Canada.

Maybe Britain has more swingers then we do?

Only in Britain you say! pity...


There is actually a simple reason why they use those "swingo-meters" on election night and we don't in Canada. In the UK, there is a long lag between the time the polls close and when results first come in because they take all the ballot boxes to a central "counting hall", lock the doors and only announce the results when the whole count for a riding is complete. The polls close at 10pm there. Usually the first riding out of 650 reports at about midnight and then three or four more report over the next hour and the flood of results only comes in between 2 and 4 am! As a result after the first riding reports, you have about an hour to kill before the next one reports and so all the talking heads on TV can do is spend an hour playing games with the swingometer to see what happens if the 4.5% Lab to Cons swing in "Sunderland South" gets replicated across the country etc.... By British standards, election night in Canada is all over in a flash - usually 45 minutes after the polls close in Central Canada the networks can declare a winner based on all the partial results flooding in.

PS: This is also why we don't bother with exit polls in Canada the way they do in the UK - we get a flood of real results within half an hour of the polls closing - in the UK if they didn't have exit polls to report on - there would be NOTHING for the pundits to talk about for the first couple of hours of the election night broadcast.

Pogo Pogo's picture

This thread reminds me of the SNL - Buckwheat Assasination show.  They interview a professor who says that some people only kill celebrities for the notoriety and we just encourage copycats with the media circus.  The interviewer tells him that he said some profound things that we should all consider, then moves on to an in depth story on the man who shot Buckwheat.

The arguments to ignore between election polls are overwhelming, but hey did you see the latest numbers...

Sean in Ottawa

Centrist party-- you are spamming.

You have made this point in two threads that resulted in a discussion and a complaint that you posted two threads -- this latest post is nothing more than an advertisement and it has been placed in three threads that have nothing to do with your party -- with the two threads you created that is five altogether. You have only placed one post outside of this promotion.

We have a lot of different political views here from NDP to Green to smaller parties to people who have no party and even some left Liberals. What we have in common is a complete lack of interest in a new right wing party between the Liberals and the Cons. People are progressives here. Now your continued flogging of this without regard for the content of a thread has moved from shameless self promotion to spam.

Your post was seen here as well:


And here:


If you want to advertise-- support this site by buying a banner ad otherwise stop using threads for ads. I see others have made the same point. I'd be surprised if you were still here a day from now if you do this again.

On the other hand if you want to bring a perspective bring it on topic and give your new party a rest for a bit.




Fighting election on content of Flaherty's budget would be foolish





The election WON'T be fought on Flaherty's budget. That will simply be the occasion on which the government will be defeated - what the main issue of the election campaign will be is anyone's guess. The opposition can easily vote the government down on the budget and then the very next day make the theme of the campaign Harper's autocratic, dictatorial style.

Life, the unive...

For instance, anyone want to tell me off the top of their head the specific motion and wording that brought the Matin government done.   The budget will be the trigger, not the base of the election.  By e-29 no one will even be talking about it other than as an example of good goverance or all that is wrong with the Haper Cons depending on who is talking.


Not the wording specifically but.....


Fall of government

The first volume of the Gomery Report, released on November 1, 2005, cleared Martin of any wrong doing while placing some blame for the scandal on former Prime Minister Jean Chrétien for lack of oversight, although it acknowledged that Chrétien had no knowledge of the scandal.[23] On June 26, 2008, Chrétien was subsequently cleared of all allegations of involvement in the scandal.[24] However, many have criticized the Gomery Inquiry as not having the scope to assign criminal responsibility for the Scandal or to investigate Martin's role, and indeed some have accused Martin of purposely "tying Gomery's hands." Gomery specifically said that Martin "is entitled, like other ministers from the Quebec caucus, to be exonerated from any blame for carelessness or misconduct", as the Department of Finance's role was not oversight, but setting the "fiscal framework".

A Canadian judge issued a gag order that barred Canadian media from covering the hearings of the Gomery Inquiry. Despite this ban, leaked information circulated after being published in an American blog.[25]

After the Gomery findings, NDP leader Jack Layton notified the Liberals of conditions for the NDP's continued support, one of which included a ban on private health care. Martin turned down the offer, as well as rejecting an opposition proposal schedule an election for February 2006, in return for passing several pieces of legislation.[26] The Conservatives, supported by the other two opposition parties (the NDP and Bloc Québécois), introduced a motion of non confidence against the Martin government. The motion passed on November 28 by a count of 171-133, defeating the government, after which the Governor General issued the election writs for a vote to be held on January 23, 2006.[27][28]

The motion was notable in that it was the first time a ruling government had been defeated on a non-confidence motion not associated with any legislation; previous defeats of minority governments in Canada had occurred in connection with the failure of financial legislation, such as budget bills in the case of Joe Clark and Pierre Trudeau or supply in the case of Arthur Meighen.

Some commentators described his tenure as Prime Minister as unfocused and indecisive, with The Economist referring to him as "Mr. Dithers".[29][30]




  1. The vote of no confidence against Fraser was in unique circumstances. The Governor-General, Sir John Kerr, had dismissed Gough Whitlam and his Labor government after their failure to secure supply in the Senate in the constitutional crisis of 1975, despite a clear, stable Labor majority in the House of Representatives, and appointed Liberal leader Malcolm Fraser Prime Minister. Fraser then advised the dissolution of both houses and the calling of an early election. Between Fraser's advice and the formal dissolution of parliament, Whitlam responded by going to the House of Representatives and passing a motion of no confidence in Fraser, but by the time the motion was formally received by the Governor-General he had formally enacted the dissolution and so this gesture made no difference, particularly as Fraser had already advised an immediate election. At the ensuing elections Fraser and the Liberals won a landslide victory, and the events confirmed that a government requires the approval of supply in both houses to govern.
  2. Trudeau lost a motion of confidence when he failed to pass the 1974 budget. However, it was later revealed that this was done purposely by Prime Minister Trudeau in a successful attempt to win a majority government. This is the only time the tactic has been used in federal Canadian politics, but it established a precedent. Such a tactic is now called "engineering the defeat of one's own government", and the practice is widely frowned upon.
  3. While Meighen, Diefenbaker and Trudeau were toppled by loss of supply, and Joe Clark was defeated by the passage of a subamendment to a budget bill that read "that this House has lost confidence in the government," only Paul Martin lost an actual motion of no confidence put forward by the opposition parties.
  4. The Grand Council of Fascism passed a resolution asking the king to resume his full constitutional powers, which amounted to a vote of no confidence in Mussolini.



alan smithee alan smithee's picture

CentristParty should be banned...He/she no longer debates or comments ...He/she is a spammer..And a bad one at that.

alan smithee alan smithee's picture

I used to take polls alot more seriously...Now each pollster have alot of disproportionate numbers and it's becoming blatantly partisan with an agenda angle.

Maybe a series of explicit questions to a random 100,000 people evenly spread across the country would give everyone a more accurate picture and perspective.

A thousand people cannot speak for a million people.

Seeing how there are so many different poll firms,each with totally different numbers make it obvious that these polls are a tool to influence public opinion.


Malcolm Malcolm's picture

It's very nice that Eric Grenier is adjusting his model so that it won't inflate Liberaal support.

He's still a hack.


Ah, but now he's a hack with competition:

 * 2CloseToCall.Blogspot.com

 * CdnElectionWatch.Blogspot.com

 * And, of course, Democratic Space will be back into action once the campaign gets underway, though he and his wife just had twin girls in February. Here's his post disavowing strategic voting from last August, however: 




I think the obsession with polls is counterproductive. Here's why:

Partisan politics can be seen as a race between political parties. Party politics is akin to athletics where each party can be viewed as being a 100m racer. In racing, the racer's goal is to win races. In politics, the goal of a  party is to win elections. In order to win races, a racers has to be faster then their competition.  In order to win elections, a political parties has to have more popular policies then their competition.

So the goal of a racer ultimately not to win races but to run as fast as possible. Winning races is the biproduct of being fast. Running speed is the independent variable. Winning elections is the dependent variable.

Following this logic, the goal of a party is ultimately not to win elections but to have as popular policies as possible as winning elections is the biproduct of having popular policies. Popularity of policies is the independent variable. Winning elections is the dependent variable.

Spending too much time concentrating on party popularity polls is akin to a runner concentrating too much on handicapping their position vis a vis other racers. Spending too much time concentrating on the other runners takes away from the runners valuable training time and reduces the runners ability to increase their speed. In turn, spending too much time concentrating on the popularity of the other parties takes away from the party valuable time that could be spent on developing better policies. A runners top priority should be training to become faster as a parties top priority should be creating popular policies because the party with the most popular policies wins elections as the runner with the highest speed wins races.

Just as speed is the goal of a runner, popular policy should be the goal of political parties. Just as handicapping the race is a waste of time for the runner, concentrating on party polls is a waste of time for political parties.

In racing, racers use the stop watch primarily to track their own speed but in partisan politics, polls are not sed primarily to guage the populatiy of policies but mostly to guage the realtive popularity of the parties. This is akin to a racer not using a stop watch to guage their own speed. Instead of polling the public to determine the realtive popularity of the parties, polls should concentrate primarily on determining the relative popularity of political policies.

There is an opportunity cost to being obsessed with polls. Instead of using energy to establish popular policies, energy is wasted obsessing on fretting about the latest weekly poll.

So why is their such a dearth of issue based polling and issue based politics and such an excessive amount of party popularity polls and horse race politics?


Have we seen this poll yet - I love this guy!  Laughing


Poll of Canadian federal leaders and party in the Province of British Columbia

Feb 26, 2011

A targeted poll of 1,058 British Columbia voters from the 2008 general federal election in Canada. This poll was conducted between February 16, and February 23, 2011. Margin of error is 3.04%, 19 times out of 20 @ 95% confidence. Based on this poll there is a 100% probability that the federal Conservatives lead the federal New Democrats and federal Liberals in the Province of British Columbia. There is a 100% probability that the federal New Democrats lead the federal Liberals in B.C.

Question #1 If a federal election were held tomorrow for which leader and party would you caste your ballot?

Stephen Harper and Conservative Party of Canada
37.5 %

Jack Layton and New Democratic Party of Canada
29.5 %

Michael Ignatieff and Liberal Party of Canada
21.5 %

Elizabeth May and Green Party of Canada
8.5 %

Other Party
2.5 %




Sean in Ottawa

RE my above post:

Harper: 4 negative one positive

Ignatieff: 5 negative

Duceppe: 4 negative 1 positive

Layton: 5 positive

Sean in Ottawa

also this in the same poll


Canadians were asked to select up to six words or expressions from a list to describe the four party leaders sitting in the House of Commons. The top five results for each one of the leaders are:

• Stephen Harper - Secretive (39%), arrogant (38%), intelligent (33%), dishonest (30%), out of touch (29%)

• Michael Ignatieff - Arrogant (44%), out of touch (38%), boring (32%), inefficient (31%), dishonest (28%)

• Jack Layton - Intelligent (32%), down to earth (29%), honest (28%), compassionate (27%), open (27%)

• Gilles Duceppe - Arrogant (30%), out of touch (24%), intelligent (20%), boring (18%), inefficient (16%)


Let's play -- which one of these is not like the other....


Sean in Ottawa

Approval ratings:

Layton 34%

Harper 32%

Ignatieff 14%

Sean in Ottawa

Vision Critical


Cons 39%

Liberals 23%

NDP 17%

BQ 9%

Green 9%


Edited the article on it said Greens were at 3% -- actually at 9%


Topic locked