Which polls are most accurate?

Here's an interesting post from Worthwhile Canadian Initiative, a keen blogger who's analyzed the seemingly inexplicable gaps in party support that we see between pollsters' numbers, particulary between the Liberals and Greens. He illustrates why poll watchers will be wise to trust Nanos's daily releases most.

Free Press Insiders polls

Is it just me or is the Winnipeg Free Press taking some liberties in the way that it's publishing the results of its Free Press Insiders polls?

The Free Press seems to be portraying the results as if they're representative of voting intentions in Manitoba and in specific Manitoba ridings. However, the poll isn't a random sample of eligible voters; it's a large Internet panel of Free Press readers. As such, the poll results are, at best, only representative of Free Press readers.

If the panelists differ from electors in their voting intentions -- something that's certainly possible given that the panelists are disproportionately Internet-savvy newspaper readers -- then the published numbers aren't indicative of party standings.

Elections Canada has some very specific rules associated with the publishing of election polls. These include indicating a poll's margin of error and whether the survey was not conducted using recognized statistical methods. I don't see either in the Free Press article.

Clearly, if a margin of error applies to these results, it should be indicated along with a note that riding-by-riding numbers are less robust than the province-wide numbers. If a margin of error doesn't apply because the poll isn't scientific, then that should be made clear in the article.

Even if Elections Canada's requirements were met in an adjoining article I missed, omitting the standard caveats from a presentation of the poll results threatens to mislead the voting public. For those thinking of voting strategically and relying on the Free Press to inform themselves, that's a big problem.


SEPT 26 UPDATE: Accompanying today's Free Press Insider poll results is a cautionary note that the poll results "do not necessarily reflect the views of the entire Winnipeg or Manitoba poplation" along with more information about the sample size and methodology. The inclusion of this information represents a big improvement -- it's clear and allows readers to put the poll results in context.

Dreaming big

Following a stream of historically poor Liberal poll numbers, some campaign observers are beginning to ask whether the NDP is about to surpass the Liberals (see here, here and here) and form Her Majesty's Loyal Opposition in the next parliament.

Indeed, the big fantasy of the federal NDP, and likely a prerequisite for the party to ever become a true national contender for power, has always been to surpass the Liberal Party. New Democrtas believe this milestone would give rise to a new political polarization between the right and left, thus damning the Liberals to the political wilderness forever.

This thinking is never far from the minds of NDP strategists, who face the tricky task of fighting the Tories without inadvertently sinking themselves by helping the Liberals who, for their part, adeptly hug the middle, adopting the rhetoric and policies of the right or the left in whatever combination is most likely to win them power.

The model of political polarization the NDP seeks emerged in Britain in the first half of the 20th century as the Labour Party surpassed the Liberals, who were never again to taste power. Canadian examples can be found in Saskatchewan, Manitoba and BC, though sometimes this is in a two-party system with the Liberals on the right (e.g., Saskatchewan's anti-medicare Liberals of the 1960s and 1970s and the current right-leaning Liberals in BC, who replaced the Social Credit Party there). Polarization is arguably occurring in Nova Scotia where the NDP appears close to winning power and the Liberals slip further into third place every election.

The big question: can it happen federally? As things stand, the New Democrats are likely to make gains and the Liberals likely to suffer losses, but can the shift be dramatic enough for the NDP to form Official Opposition? Some polls show the gap between the two parties to be as little as 3%, a level of competitiveness not seen since around 1990.

So what would it take?

First, it's important to note that the NDP need not necessarily surpass the Liberals in popular vote to surpass them in seats. This is because, generally-speaking, the Liberals' votes tend to be more evenly spread than either the Conservatives' or the NDPs' votes, which tend to be concentrated in particular areas. This tends to make the Liberals more efficient than other parties at converting votes into seats when they're popular but less inefficient than other parties when they're unpopular. In 1984, the Liberal-NDP gap was 28% to 19%, but the seat gap was only 40-30. Another percentage point or two might have pushed the NDP into second.

To come in second in this election, the NDP would likely have to win at least 60 seats or so, with the Liberals falling below that number. The losses for the Liberals in Ontario, where they won 54 seats in 2006, would have to especially large. Specifically, we'd need to see the following:

Ontario

Liberal numbers would have to drop below 30% from the mid-to-high 30s where they are now, while the NDP would have to climb 5 points or so to the low 20s, with noted strength in northern Ontario. The Tories would break 40%. Liberal losses to both the Conservatives and NDP, plus bleeding to the Greens in vulnerable suburban ridings, would result in a seat count of something like this:
Con 55 / Lib 28 / NDP 23 (2006: Con 40 / Lib 54 / NDP 12)

Quebec

Bloc numbers would have to stay over 30%, around where they're hovering now, as continued Bloc leakage to the Tories and NDP will produce new victories for the Liberals. The Tories would have to stay around their current 30% while the the NDP approached 20% (up from about 15% now). This might yield something like this:
Bloc 33 / Con 23 / Lib 14 / NDP 4 / Ind 1
(2006: Bloc 51 / Con 10 / Lib 13 / Ind 1)

Atlantic Canada

The Tories would need to gain only a few points (to maybe 36%) at the expense of the Liberals for them to steal a number of close Liberal seats in Nova Scotia, PEI and especially New Brunswick. This would offset Tory losses that are all but guaranteed in NL (a victim of Danny Williams's ABC campaign). The NDP, who are already up a few points over their 22% in 2006, would have to eke out a few new victories in St. John's and Nova Scotia. In seats, that would probably come out to something like this:
Con 14 / Lib 10 / NDP 7 / Ind 1
(2006: Con 9 / Lib 20 / NDP 3)

Prairies

On the prairies, a couple of points gained for the NDP over their 2006 numbers could give them up to five more seats (I can't see the party winning more than three in Saskatchewan). The Tories would have to win some, lose some, leaving the Liberals with 1-2 seats in total on the Prairies. That would leave something like the following seat count:
Con 46 / NDP 8 / Lib 2 (2006: Con 48 / NDP 3 / Lib 5)

BC

NDP numbers have been strengthening in BC in the last week. Two or three more points would put them over 30% and win them a few new seats at the expense of both other parties. The Liberals, who are bleeding to the Greens in suburban Vancouver, would see some seat losses to the Tories. That could leave the situation something like this:
Con 18 / NDP 17 / Lib 1 (2006: Con 17 / NDP 10 / Lib 9)

North

A territorial seat for each party as the Cons pick up Nunavut:
Con 1 / Lib 1 / NDP 1 (2006: Lib 2 / NDP 1)

Nationally

In order to make the above happen, the NDP would need something in the range of 22%, while the Liberals could have no more than about 26%. In addition, the shift in numbers would have to be in the right regions. The above seat counts total as follows:
Con 157 / NDP 60 / Lib 56 / Bloc 33 / Ind 2
(2006: Con 124 / Lib 103 / NDP 29 / Bloc 51 / Ind 1)

The above exercise, apart from being fun (it's still more fun than realistic at this point), suggests it's narrowly possible for the NDP to surpass the Liberals. I would suggest that it remains an uphill battle for the NDP and that they shouldn't underestimate the Liberals. Still, a lot can happen in the one month of campaign that remains.

If the Liberals somehow do tank and make a historically poor showing on October 14, they really shouldn't blame Dion, who seems to be a rather decent, intelligent politician. Rather, for a number of reasons, they can place the blame squarely on this fellow.

Photo: The House of Commons (facing the government side)

If I were a rich man...

I wonder if this scene from the musical Fiddler On The Roof is what Dion had in mind when he announced on Tuesday that he's not a rich man.

It does make me wonder: did the Liberals not see this campaign coming? First, they were forced to scramble to secure a campaign plane (and ended up with a real gas guzzler). Now, they've suddenly realized that Dion's image needs a makeover and are trying to recast him as just a regular guy -- why didn't they do that a year ago?

Some other thoughts and observations on the campaign thus far:

• You have to wonder about Ontario Liberal Premier McGuinty's decision not to endorse his federal counterparts. Apparently, he doesn't want to worsen his already abysmally poor relationship with feds. If that doesn't tell us how far ahead the Tories are in this campaign and how unlikely the Liberals are to win, I don't know what does.
  
  
• The puffin poop Tory ad that was all over the new today blew up in the Tories' faces, as it well should have. Have they forgotten how their ad making fun of Chretien's paralysis backfired several years ago? Perhaps the Tories' punishment should fit the crime -- straight to bed with no dessert, plus grounding for one week.
  
  
• I was somewhat surprised that May was barred from the televised debates given that the criteria for entry has always been to have a minimum of one MP, which the Greens have now met. Ironically, the Liberals, who pushed for May's inclusion, would likely have the most to fear from May's participation, as polls show it's them bleeding the most to the Greens, particularly in Ontario (confirming some comments I made previously).
  
  
• One new EKOS poll puts the Liberals at the historically low level of 24%, only narrowly ahead of the NDP, at 19%. Similarly, a second poll (by pollster Segma Unimarketing, which I've never heard of until now) put them at 25% to the Conservatives' whopping 43%. The federal Liberals, one of the most successful political parties in the western world, have only received that low a share of vote once -- in 1867! In 1984, their next worst performance, the Liberals converted 28% of the popular vote into just 40 seats.
  
  
• The NDP might be the only party to run a full slate of 308 candidates in this election. The Tories have bowed out of Portneuf-Jacques-Cartier to help Independent MP André Arthur in Quebec, while the Liberals are of course not running against Elizabeth May in Central Nova, whose party isn't fielding candidates against either Stephane Dion in Saint-Laurent-Cartierville or former Conservative Bill Casey in Cumberland-Colchester-Musquodoboit Valley.
  
  
• On Monday, Young Liberals heckled Jack Layton for borrowing Liberal votes in the last election instead of focusing on Harper. They claim that attacking the Liberals in 2006 split the anti-Harper vote and helped elect him. Now, if they really believed that, would they not be busy heckling at Tory rallies instead?
  
  
• With PQ leader Pauline Marois undergoing surgery, the Bloc loses an important campaign ally (Marois had planned to stump for Duceppe). That party is desperately struggling to stop the loss of support to the Conservatives and, to a lesser extent, the NDP (a just-posted Globe article suggests the Bloc may have slowed the Tory momentum in Quebec).
  
  
• New Stephane Dion website: http://www.thisisdion.ca/. I wonder how many fans looking for this website will accidentally wind up at the Liberal leader's site.
  
  
• Finally, there are some great online resources for political junkies who just can't get enough. Pundit's Guide is an amazing database of candidates, contests and results. Paulitics and niXtuff provide regular tracking of polls. The Globe's, Canwest's and the CBC's sites provide additional articles, polls, and punditry. Independent sites nodice.ca, Election Prediction Project are also great resources. Democratic Space will launch its election coverage on Sept. 14.
  
  
  

A two-horse race and other pre-election myths

Today's Globe and Mail pretty much stole the headline I was planning for a blog post: Harper Tories On The Brink Of Majority. This might come as a shock to some pundits who keep looking to tied polling numbers as the be all-end all source of election analysis, so let's bust a few pre-election myths.


Myth: Harper's calling an election, but even he doesn't think he'll get a majority

Reality: The Canadian public are majority-wary and the Tories know this, so of course they're going to downplay their chances. Even if some polls show them hovering somewhere between 32% and 35%, that's only a few points short of majority territory, and they're extraordinarily well-organized and flush with cash. While they sure wouldn't mind a stronger minority and sticking it to the Liberals, as Tom Flanagan suggests, make no mistake -- they're gunning for a majority government and they just might get it.


Myth: The Liberals are neck-and-neck with the Tories and have a decent shot at winning

Reality: Party leaders rarely make gains in their first election campaign, and the odds are even worse for a leader going into the campaign with baggage in the form of a poor image, an unpopular campaign plank, fundraising challenges (now that the days of handouts from corporate Canada are gone), and close to a quarter of the caucus opting not to run again. Those who think the Liberals are in the game to win might want to ask why the party has done absolutely all it could in the last year to avoid an election.


Myth: The NDP is losing ground to the Greens

Reality: Most polls show the NDP holding all or nearly all of its 2006 levels of support and, while the Greens are still up a little over their 2006 levels, most of this added support is in Liberal-friendly suburban Ontario (though even that may vanish as the Liberals continue their fine tradition of "borrowing" the policy planks of other parties).

The NDP hopes to build on its 2006 support with a now-seasoned and well-recognized leader (no campaign gaffes as in 2004), the best-financed and most competitve campaign in the history of the party, a surprising number of "star" candidates who have emerged to run for the party in non-traditional areas, and Liberal slippage to the right (Tory success tends to throw NDP-Liberal battles to the NDP). This should produce new victories in Montreal and Gatineau, northern ridings, industrial Ontario (especially in the wake of news like this), and pockets of support in the West.


Myth: Winning or not, the Greens are more principled than the other parties

Reality: Desperate to do anything to meet the one MP threshold to join the leader's debates, the Greens recently welcomed into their ranks an MP who was kicked out of the Liberal Party for spending irregularities. With their hug-a-Liberal strategy backfiring as the Liberals steal their carbon tax plank, expect the Greens to make a lot of noise and gain a few points in the popular vote, but wind up with a big donut for seats.

A very rough early prediction:

Cons ~152
Libs ~85
NDP ~39
Bloc ~32

Image: Thomas Kelly, "The False Start," 1858. Hand-colored lithograph depicting Jerome Park Racetrack, New York City.

Polling mysticism

So today's poll has the Tories "flirting with a majority." Now wait... didn't yesterday's poll have them in "a dogfight" with the Liberals?

Apparently, yesterday's competitive Liberals are now suddenly today's underdogs fighting even to stay above historic party lows (of which their reported 27% would certainly be). The NDP, whose 12% in today's poll put them well below their 2006 support levels, were just two weeks ago at 19% and apparently well-positioned to make gains. Back then, the Greens found themselves at 7%, a standing barely above their 2006 election result and far short of their apparent rapid ascendance to 13% in the latest poll.

As political observers, we all ooh and ahh at each new horse race figure that's printed in the papers, trying to read into the numbers some sort of truth about who's ahead and who's behind and who's going to be sorry and who's going to be rewarded. The art of poll watching has taken on an almost mystical quality, with its high priests quick to ascribe some dramatic trend or, in some cases, validate the various notions and myths their particular political tribe feels a need to hold on to (which includes partisan false bravado).

For political junkies, prognosticating election results after every poll makes for fun exercise, but it's fairly pointless, for a number of reasons:

• First, most of us who read about polls regularly know some of the reasons that poll numbers jump around: sampling (± a certain percentage, 19 times out of 20) and various sorts of non-sampling errors, and the way survey questions are asked (debates continue about what question wording best replicates how voters are likely to respond upon arriving at the polling booth, something that Blackberry Addicts were never exactly shy about raising).


• Second, polls taken in between elections are often more about party brands than they are about the true competitive standing of each party relative to one another. It's what I often call a "parking lot" effect, in which poll respondents instinctively "park" their vote in between elections. Asked by a pollster who they'll vote for and they'll tend to blurt out whatever party brand they roughly associate with or that happens to be top of mind. The Liberals and Greens tend to be beneficiaries of this tendency in most parts of Canada.


• Third, pre-election campaign polls don't take into account party organization, finances, gaffes or infighting, strategic positioning, the name recognition of incumbent "star" candidates, media exposure, and party messaging, all of which play a huge role in that complex game of chess that plays itself out in each election campaign.
  

For these reasons, the hype that follows most polls tends to be pretty hollow. A better analysis might take into account some of the above factors.

The Greens, currently buoyant in most polls, have considerably less ground organization, campaign experience, cash resources, star candidates, or media exposure than the other three national parties and will almost surely sink when the next election campaign goes live. Nik Nanos, of SES Research, suggests that the Greens typically drop 1/3 of their vote from the last poll to the actual election. That's after the deflation of the Green balloon that's likely to occur during the campaign itself.

The Liberals, in their current state of endless infighting, fundraising difficulties, mass retirement of MPs and loss of valuable potential candidates, to name just a few of their problems, had best shape up and fast if they expect to hold the 27% to 35% the polls currently peg them at.

Worse for the party is the unenviable position they keep backing themselves into: critiquing the government but then doing all they can to make sure it isn't defeated. For them, voting to defeat the government in a confidence motion means risking catastrophic losses in an election (for some of the reasons noted above), yet supporting the government or abstaining from confidence motions means branding themselves as a "weak" opposition that has little alternative agenda to that of the government.

A weak, indistinct, vision-less Liberal Party is one that's going to find itself wedged in an election between the Conservatives and the NDP when centre-right Liberals find they have no reason to vote for a "me too" Liberal rather than a Tory and when centre-left Liberals find that the alternative vision they're looking for lies only with the NDP. Is it any wonder, then, that the Tories and NDP are literally trying to goad the Liberals into an election, while the Liberals are finding ever new and interesting ways to contort themselves in an attempt to avoid just that?

We live in very interesting times.

Photo: a map of the 2006 federal election results.

May's Greens: rising tide?

Globe columnist Lawrence Martin's latest piece on Elizabeth May and the federal Green Party was quite interesting. As many know, May is fighting to participate in the leader's debate and hoping to knock off Peter MacKay and win the first seat for her party. It won't be an easy task, but the Greens are feeling boosted by recent polls showing her party as high as 13% nationally and as high as 17% in Toronto's "guilt-ridden" 905 belt.

The big question is whether May can hold on to her poll numbers and translate them into votes at election time. It's likely that the Greens will be far outspent and outorganized by the three national parties, all of whom will be spending the maximum amount, and likely the Bloc too. Electoral history is filled with stories of overly optimistic parties driven by buoyant but ultimately ephemeral support levels.

A lot of folks assume that if the Greens do gain support that it'll be at the expense of Layton's NDP. According to the Martin article, some Greens even talk about merging with the NDP to create a new GDP (apparently that's short for "Green Democratic Party" and not "Gross Domestic Product"). However, Green gain at NDP expense isn't entirely certain. If it were, how might we explain the fact that the GP's greatest strength, when looking at all 500,000+ sized urban areas, lies in Calgary (9% in 2006, on average) and the 905 belt (17%, according to some poll numbers). These are hardly traditional areas of support for the NDP, which, at 10%, came in barely ahead of the Greens in Calgary and, outside of Oshawa, Hamilton and the Niagara region, struggles for votes in the 905 area that surrounds Toronto. Meanwhile, the NDP's two strongest cities, Vancouver (27% in 2006) and Winnipeg (28%) are actually two of the GP's weakest (5% and 4%, respectively).

My hypothesis right now is that the Greens are emerging as a or perhaps even the protest party of choice in regions where the NDP is relatively absent. Apart from a few BC ridings, wherever the NDP has a strong, well-managed campaign, the Green vote tends to be a non-factor, staying in the 2% to 6% range.

For the Greens, it certainly doesn't hurt that their message is more palatable for some suburban voters who feel a need to support a platform they see being ecologically-conscious, but who may be uncomfortable with the NDP’s traditionally pro-labour, redistributive policies (which the Greens tend either not to share or to keep awfully quiet about). If that's the case, the Greens may well be usurping more Liberal vote than NDP. Is that the reason behind the Dion/May hug-a-Green/hug-a-Liberal strategy? It's unlikely that, over the long run, they're both going to emerge victors from their quasi-alliance.

The Internet, being the ever-glorious provider of election study and polling numbers that it is, has given me quite a bit of interesting data to play with. That's one source of the tables I have above, which are simply summed riding-by-riding totals for each region. I've also found that the 2006 Canadian Election Study data files are freely available; that study consisted of interviews with thousands of Canadians before and after the election to gauge such things as party momentum, preferred second choices, and reactions to party platforms, campaign announcements and party leaders. In the lead up to the next federal election, I hope to share of the interesting findings from this data.

---

At least one fellow Manitoba blogger, upon noticing my recent profile change (which states I've been a blogger since only January of this year), has concluded that my entire 2007 blogging history must have been written and posted all this month. It sort of conjures up the image of some wretched, hunched-over character in a dark, grungy basement cackling madly in between fits of wild typing to produce mass amounts of blog content (if only I was that prolific...). No harm done, but lest others make the same assumption, I thought I'd make it known that I merely changed the Gmail account I associate with the blog to a new one. For those using a Gmail address with blogger.com, it's easily done, as the instructions here, here, and here all attest. Of course, given the new profile date that'll appear by your name, you should be prepared to have any and all past election predictions challenged! Don't say I didn't warn you...

And the race is on...

Lots of posting today!

Tories fail to run full slate of candidates

There are sure to be some red faces in the Tory backrooms today. Nominations have closed and, after falling short in Flin Flon, the party was unable to field a full slate of candidates in the election.

Rightly or wrongly, the ability to run a full slate of candidates is almost always taken as a key indicator of whether a party is organized and serious about winning or whether it's a fringe party. Even the struggling Liberal Party, which was unable to run a full slate in 1999, was able to secure a name on the ballot in all 57 constituencies this time around. This is a big blow to Tories, indeed.

The race in Flin Flon will now be fought between NDP incumbent Gerrard Jennissen and Liberal challenger Garry Zamzow. That's if you can really call it a race: Jennissen took 73% of the vote in the last election.

The failure to run a full slate will be especially hard for the Tories' reputation in northern Manitoba. In some election campaigns, the Tory leader doesn't even visit the north. Now, long accused of neglecting and ignoring the north, the Tories won't even be offering residents of part of that region the chance to to vote for them.

Adieu, Monsieur Rocan

Longtime former Tory MLA Denis Rocan didn't follow through with his threat to run in the provincial election. He had suggested possibly running as an independent against Hugh McFadyen in Fort Whyte. The other obvious option for him would have been to run as an independent in Carman, where he was challenged and defeated for the Tory nomination. Frankly, running in Carman (where the Conservatives got 52% of the vote in 2003) would've been his best shot at winning.

I have to say I think Denis Rocan is a pretty decent guy, having met him once in the 90s. I could probably never vote for him myself, but I've heard he was a solid representative and hard worker in Carman. It didn't seem to do him much good within the Tory ranks, however -- he was booted out first as the candidate when he lost the nomination race and then from the caucus after he broke ranks and supported the NDP's budget.

Wellington recap

The race in Wellington continues to attract a lot of media attention, especially now that former New Democrats Conrad Santos and Joe Chan are running as independents against their old party. Flor Marcelino, the NDP's candidate and editor of the Philippine Times, looks like a great candidate. She'll do well.

The little parties

In addition to the three most established parties, the Green Party and the Communist Party are running candidates. The Greens will be on the ballot in 15 constituencies, while the Communists will be an option in six. The voters in Fort Rouge will have the most choice in this election, with six candidates to choose from: a New Democrat, a Conservative, a Liberal, a Green, a Communist, and an Independent.

The Green Party leader, Andrew Basham, is running directly against Premier Gary Doer in his constituency of Concordia. Andrew Basham's mother, whose basement he apparently still lives in, is running in Wolseley, where the Greens captured about 19% of the vote in 2003. After the infighting that's plagued the Greens since then and the NDP's general dominance of the environment as an issue, there's little hope they'll be able to even come close to repeating their 2003 result.

Manitoba's Communist Party, a perennial competitor and probably the most active such party in Canada, is also running. Most of its candidates are the same names that appear on the ballot election after election.

I once spent a little while chatting with Darrell Rankin, the party's leader, after he happened to knock on my door in one election. As a candidate, he struck me as quite interesting, engaging and intelligent. I wasn't quite as impressed with the party pamphlet -- complete with a headline praising the regime in North Korea -- that he left behind.

A day at the races

I'll be posting some seat-by-seat predictions in the very near future. However, I may wait to see what the next provincial poll numbers are before weighing in on the really close races.

My guess is that, with the relatively sleepy pace of the campaign so far and the inability of the Tories to land any punches, the next poll will show an increase in NDP support over the last poll, with the Liberals and others down a couple points each, and the Tories holding steady. Then I expect the NDP to come out a few points below those numbers on the night of May 22. My assumption is that there are many fickle and wishy-washy voters who, once alone in the dim light of the polling booth, will vote for change regardless of who's in power or how well they've done.

The last poll before the campaign started showed the NDP and Tories neck-and-neck at 40%, with the Liberals at 15%, and others holding what's left. To compare, the NDP received 49% in the last election, with the Tories getting 36%, the Liberals 13%, and others 1%. If the poll numbers stay as they are when people vote, it's likely the NDP would be re-elected with a majority by winning a pile of Winnipeg seats by a small margin, while the Tories stack up giant majorites in places like Steinbach, Emerson, and Pembina.

In the next poll, the Liberal numbers will be the ones to watch: provincially, the Liberals almost always fall during the campaign, which tends to benefit the NDP. The Tories usually have the best chance of winning when centre-left voters are split between the NDP and the Liberals. If the Liberals buck the trend and creep upward, it may cut into NDP votes in southern Winnipeg and toss those seats to the Tories. In contrast, if the Liberals sink below the 13% they got in the past two elections, the NDP may be able to pick up Inkster from the Liberals as well as Southdale and Kirkfield Park from the Tories.

To compare with 2003, the pre-campaign polls then put the NDP ahead at 44%, with the Conservatives at just 30% and the Liberals at 21%. Mid-campaign polls showed the NDP surge to 51% and 55%, while the Liberal numbers dropped off and the Tory numbers stayed static. On election day, the Tories "pulled" their vote well and ended up with 36%, still well behind the NDP's 49%.

A second thing to watch in the next poll will be the Winnipeg/non-Winnipeg split in the numbers. The last study showed the NDP down 8 points in Winnipeg and down 15 points outside of Winnipeg compared to the last election. If the party picks up, will it be in one region or both?

Astonishing facts

Emerson and Springfield, now both staunchly Tory seats, were once NDP seats.

In 1973, Steve Derewianchuk, the NDP candidate in Emerson, took 2,374 votes, beating out the Tory with 1,937 votes and the Liberal with 1,768 votes. The victory was relatively short-lived, however: Albert Driedger regained the seat for the Tories in 1977. Driedger later went on to represent Steinbach in the legislature. In Emerson in 2003, the Tories took 59% of the vote to 21% for the Liberals and 19% for the NDP.

Rene Toupin won Springfield for the NDP in 1969 and 1973, but lost it in 1977. Andy Anstett narrowly regained the seat for the NDP in 1981, only to lose it to Tory Gilles Roch by 55 votes in 1986. Almost immediately after being re-elected as a Tory in 1988 (and being subsequently denied a cabinet post), Roch crossed the floor to join Sharon Carstairs's Liberals, which had just become the Official Opposition. The Tories easily regained the seat in 1990 and have held it ever since with ever-increasing majorities, which come largely due to an influx of wealthy Tory voters into new subdivisions in places like Oakbank. In 2003, the Tories won the seat with 61% of the vote, to the NDP's 31%.