Momentum with Notley's NDP in Alberta

The numbers continue to amaze, as a new poll this morning puts Wildrose and the New Democrats neck-and-neck in Alberta's provincial election campaign. The governing Progressive Conservatives, gunning for their 13th consecutive electoral victory stretching back to the Nixon administration, have dropped out of the three-way race.

The projection still gives Wildrose the lead with 31% support, or between 29% and 33%. The New Democrats overlap a fair bit, with between 26% and 31% support (or 29% more precisely). The PCs are in third with between 24% and 28% (or 25%), while the Liberals trail at length with 10% to 12% support.

Wildrose also has the edge in seats, and is projected to take 35. The PCs come up second by a hair with 24, while the NDP takes 23. 

But the ranges tell a more nuanced story. While Wildrose is comfortably ahead with between 27 and 44 seats, the NDP is more solidly placed to form the Official Opposition with 18 to 34 seats (and they could even conceivably win a plurality). The PC range tops out at 33, similar to the NDP, but bottoms out at just nine.

The Liberals stand at four seats and a range of four to nine, though that is likely to change this weekend once the official candidate list is out. The model currently assumes every party is running a full slate. The numbers will be adjusted once it is known exactly where each of the parties will have a candidate running.

Note that the projection now gives the other parties a range of zero to one seat, and 5% support. That seat belongs to the Alberta Party.

I'd also like to address the polls being done by 1ABVote (or 1Question, as they appear on Wikipedia), since I have received a lot of inquiries about them. I've spoken with Brian Singh, the man behind the polls, in the past, and I don't doubt that his polls are genuine. But this site does not and never has included polls commissioned by interest groups or political parties. As 1ABVote is an organization that is trying to unite progressives in Alberta, it does not qualify for inclusion in the aggregate. 

But let's get to the poll that was added to the aggregate this morning. It was done by Mainstreet Technologies and published by the Calgary Herald.

Mainstreet was last in the field on April 7. It recorded no change for Wildrose since then, as the party remained in first with 31% support.

The NDP picked up four points and was second with 30%, while the PCs were down three points to 24%.

The Liberals were down two points to 10%, and the Alberta Party was up two points to 5%. The number of undecideds ticked down by a point to 23%.

Turnout is unlikely to boost the Tories, as Mainstreet finds only Wildrose gets a boost among those who say they are certain to vote. The party is bumped up to 35% among these Albertans, with the NDP dropping to 29% and the PCs holding at 24%.

The poll showed the Tories have weakness not only in voting intentions but on the issues, as they led in none of the categories investigated by Mainstreet.

Wildrose led on the issues of taxes and healthcare, while the NDP was ahead on the environment. The two parties were nearly tied on education, and the only three-way race was on job creation. That the PCs could not poll better than Wildrose or do much better than the NDP on this issue is particularly problematic for them.

At the regional level, the race is closest in Calgary. Wildrose was narrowly ahead with 29%, while the Tories were down six points to 27% in the city. The NDP was up to 25%, the Liberals to 13%.

The New Democrats dominated in Edmonton with 51%, followed at a distance by the Tories at 21%, the Liberals at 13%, and Wildrose at 10%. There were only marginal changes since April 7.

Wildrose held a wide lead in the rest of Alberta with 39%, with the NDP gaining six points to surge into second place at 26%. The Tories were down to 23%, while the Liberals were down six points to just 7%.

These polls still boggle the mind, but it is hard to deny that they are pointing to something very real in Alberta. The real question is whether it will endure until May 5, of course. It would also be useful to have a few more polls from more established outfits to help confirm the trend.

Uptick in Alberta for Mulcair's NDP?

The federal projection has been updated, incorporating the latest polling data from EKOS Research, Léger, and CROP. The overall results have not shifted much, though the Conservative seat count has dipped to the benefit of the Liberals. You can see all the details here.

An interesting development does seem to be underway in Alberta, however, where the New Democrats are polling at their highest level in a year. Is the NDP on the upswing in the province for real?

First, let's briefly go over the new polls added to the model.

In EKOS's poll for iPolitics, the Liberals moved ahead with a gain of 1.6 points since the previous week to 32.1%. The Conservatives dropped 2.7 points to 30.2%, while the NDP was up 1.5 points to 21.2%.

Only the Conservative decrease was outside the margin of error, though the Liberal bump does end a losing streak for the party. Overall, however, the poll is well within EKOS's usual range.

The Léger poll reported by CTV Montreal showed a similarly close race, with the Liberals at 35%, the Conservatives at 34%, and the NDP at 20%. Compared to Léger's previous poll of Jan. 30 to Feb. 2, the Liberals held firm, the Tories were up two, and the NDP was unchanged. Suffice to say, those shifts would not be outside the margin of error of similarly sized probabilistic samples.

Finally, the poll by CROP for La Presse gave the New Democrats the lead in Quebec with 30%, unchanged from CROP's previous two surveys in the province. The Liberals dropped four points to 29%, while the Conservatives were up two points to 18% and the Bloc Québécois was up one point to 18%.

None of these shifts would be statistically significant. Noteworthy, however, is that the Liberals have dropped in two consecutive surveys (from 37%) while the Conservatives have increased their support over that time (from 13%). Also note that the results from CROP are broadly in agreement with those from Léger for the province.

Alberta turning a little orange

The longest and largest streak (both in real terms and proportionately) in the country currently belongs to the New Democrats in Alberta. Their numbers have been increasing over five consecutive weeks in the province, from 11.9% at the beginning of February to 16.3% in today's update. That is a gain worth 4.4 points, or an increase of more than one-third since that low ebb.

This upswing for the NDP in Alberta has coincided with a period of Conservative decline, as the party has slipped 4.2 points over the last five weeks from 54.7% to 50.5%.

(The third-longest streak at the moment is in Quebec, where the Liberals have fallen over four consecutive weeks from 30.3% to 27.7%.)

This jump for the NDP has pushed the party to two seats in the projection (Edmonton Strathcona and Edmonton Griesbach, though the latter is by a very slim margin). It has also put the New Democrats in play in Lethbridge.

The shift has been registered in most polls. In the two newest surveys, Léger has the NDP up five points in the province and EKOS has them up almost 10 points since the poll it conducted at the same time as Léger's last outing. Over the last eight polls, the NDP has managed between 13% and 19%. In the previous eight, that range was from 10% to 17%.

Whether or not the increase has come directly from the Conservatives is harder to determine. In those same most recent eight polls, the Conservatives have ranged between 49% and 54%. In the previous eight, that range was 47% to 61%. That suggests the Tories' numbers have just become more volatile, rather than necessarily dropping in the face of NDP gains.

The reality is rarely ever linear, and in this instance we may be seeing voters heading in every direction, the net result being an NDP uptick.

But why would this be taking place? We cannot definitively chalk it up to an improved performance by Thomas Mulcair, whose personal numbers in Alberta show no such clear trend line as those of his party.

We instead might be seeing a bit of cross-pollination between the provincial and federal scenes. Talk in Alberta is about the possibility of the New Democrats, under new leader Rachel Notley, taking over from Wildrose as the Official Opposition after the next election. The better press for the provincial NDP may be having a positive impact on the federal party.

The real question, as is always the case, is whether or not these numbers will prove to be sustainable or a mere flash in the pan. In the grand scheme of things, it does not have much impact on the national portrait. But with the race looking as close as it is, even losing an extra seat or two to the NDP in Alberta can have an important effect on the Conservatives' electoral calculations.

Liberal recovery or blip?

The projection was updated yesterday, reversing the first instance of the Conservatives leading in both the vote and seat projection. The Liberals are now back on top in the polling average, and have closed the gap in the seat count to just five. Does this mean the Liberal slide has come to a halt, or is it just a blip?

The Liberals lead again with 34.2%, up a point from the last update incorporating polls done to Feb. 3. The Conservatives dropped 0.6 points to 32.9%, while the NDP was down 0.7 points to 19.3%.

In terms of seats, the Conservatives have dropped three to 140, and their range has gone from 128-162 to 125-158. The Liberals have increased their seat tally by 13 to 135, and their range from 106-140 to 114-152.

The New Democrats have dropped 12 in the seat count, and their range has fallen from 53-81 to 43-77. This was due primarily to a drop in Quebec, where the party is down seven seats from the Feb. 3 update.

The bulk of the Liberals' gain, however, came in Ontario, where the gap between them and the Conservatives widened from 0.5 points to 2.9.

Let's look at the two polls that prompted the update. They were by EKOS Research for iPolitics and Forum Research for the Toronto Star. Both left the field on Feb. 10, but the EKOS poll had a longer tail.

Forum was last in the field from Jan. 27-28, and recorded a relatively large shift in voting intentions since then. The Liberals were up five points to 39%, re-taking the lead from the Conservatives, who were down three points to 32%. The NDP was down three points as well to just 17%.

The increase in Liberal support is outside the margin of error, whereas the drop for the Conservatives and NDP is not. But before Liberals get too excited, consider their recent trend in Forum polling: 36% - 41% - 37% - 34% - 39%. On average, Forum has had the Liberals wobbling by about four points from one poll to the next. This would suggest to me that this bump in Liberal support is more house effect than anything real.

That is, at least in terms of its size. EKOS also recorded a modest rebound by the Liberals, which argues that perhaps the party is indeed recovering from their recent slide.

EKOS was last in the field Jan. 28-Feb. 3, and put the Liberals up 1.6 points to 33.8%. The Conservatives were down three points to 32%, while the NDP was up one point to 18.9%.

Only the Conservative drop was outside the margin of error, but both Forum and EKOS recorded similar trends for the Tories and Liberals. For EKOS, this poll halts a series of surveys showing Liberal/NDP decline and Conservative growth.

Can we reconcile the two polls? One wonders whether we should bother, considering Forum's history of volatility. But if we extend the polls' results to their respective margins of error, we get the following:

As you can see, the ranges for the Conservatives and NDP overlap quite comfortably. The ranges for the Greens do not, and the Liberals only meet at 36%. I do not think that this is coincidental. The Greens are generally higher in EKOS's polls than in any others, while the Liberals are generally higher in Forum's.

The two polls suggest that the shift in voting intentions may have occurred among men. Forum recorded a jump of six points for the Liberals among men, outside the margin of error, while EKOS also showed a statistically significant jump of 4.6 points for the party. Both showed the Conservatives sliding by four to six points. Together, it puts the Liberals at either 34% or 40% and the Conservatives between 34% and 35% among this demographic. There was no similarly consistent shift among female voters.

One anomaly of the Forum poll was in British Columbia, where the Liberals were up 15 points to 47%, the Conservatives were down 11 points to 24%, the NDP was up six to 23%, and the Greens were down nine to 5%. That is a lot of movement, particularly in the context of EKOS finding no party shifting by more than a handful of points. EKOS put the parties at 31%, 29%, 21%, and 16%, respectively.

In Quebec, the province everyone is keeping an eye on, both Forum and EKOS showed a bit of a reversal for the Conservatives. Forum had the party down to 21% and EKOS had them down to 23%, though both of those shifts were inside the margin of error. They are both still, however, relatively high numbers for the party in Quebec. The Liberals appeared to take advantage in both polls, Forum recording a significant jump of eight points (but EKOS just two).

The Liberals increase nationwide coincided with a drop in Justin Trudeau's disapproval rating, from 44% to 38%. His approval rating was up slightly to 48%.

The Conservatives' decrease, meanwhile, coincided with worsening numbers for Stephen Harper. His approval rating was down four points to 36%, while his disapproval rating was up five points to 56%, the worst numbers EKOS has recorded since November. This may not be too worrying for the Prime Minister, though, as his approval rating was up in Ontario and down in the West, where he is at little risk.

Thomas Mulcair's numbers were fairly steady, at 51% approval to 36% disapproval. Regionally, his numbers improved west of Quebec and worsened east of Ontario.

It would be helpful to have a third (and fourth?) poll weigh-in on this, preferably one which uses a different mode of contact. Has the Liberal slide been reversed? Is the Conservatives' momentum coming to a halt? Or will this prove to be just a pause in the trends that have been building for the last few months? We'll see.

Conservatives and Liberals effectively tied

Two new polls, both showing the Conservatives and Liberals only one point apart (and to the advantage of the Tories), have nudged the vote projection into a virtual tie. That means the Conservative seat lead, which stood at 10 on Jan. 13, has now increased to 17. Nevertheless, the Conservatives are no closer to majority territory, and in fact still overlap a great deal with the Liberals.

The Liberals are still narrowly ahead in the aggregate, with 33.4% support to 33.2% for the Conservatives. That is a small drop of 0.3 points for the Liberals since the last update, and a gain of 0.8 for the Conservatives. The New Democrats are up 0.7 points to 21.3%.

In terms of seats, the Conservatives still lead with 139 (up three), or a range of between 122 and 155. Interestingly, the high range for the Tories has not moved at all. Instead, only the lower range has increased, from 117. It means the odds that the Conservatives would win fewer seats than the Liberals on current polling levels have lessened, but the Tories are no closer to the magic number of 170.

The Liberals have dropped four seats and their high range has fallen from 144 to 138. Their low range of 107 seats, however, is steady.

The NDP is up two seats to 74, but its low and high ranges have narrowed from 54 to 87 seats on Jan. 13 to 60 to 84 seats now.

The Bloc's high range has slipped from nine to four seats.

Let's take a look at the two new polls added to the model.

The poll by Forum, published in the Toronto Star on the weekend, put the Conservatives up two points (since Forum's last poll of Jan. 5-6) to 35% and the Liberals down three points to 34%.

The NDP was unchanged at 20%, while the Greens and Bloc were each up one point to 6% and 5%, respectively.

None of these shifts were outside the margin of error.

Abacus showed similar stability, with the Conservatives and Liberals each down one point to 33% and 32%, respectively, since Abacus's last poll of Dec. 18-20. The NDP was up two points to 24%, while the Greens were down one to 5% and the Bloc was unchanged at 4%.

All of these shifts were also within the margin of error (of a probabilistic sample of similar size), but it is clear from both Forum's and Abacus's trend lines that the Liberals are drooping to the benefit of the Conservatives (if not in absolute terms, at least in relative terms).

This is also the first time since Justin Trudeau became Liberal leader that two consecutive polls by two different pollsters have shown a Conservative edge.

Forum does have a regional oddity in its numbers, though. In Quebec, Forum gave the Conservatives 26% support, putting them in second place ahead of the NDP (25%) and behind the Liberals (27%). That is absurdly high for the Conservatives (you need to go back to before the 2008 election to find the Conservatives routinely polling at that level), but Forum has often had higher numbers than usual for the Conservatives in Quebec. And broadly speaking, the Tories have been experiencing an uptick in the province, so perhaps this is a product of that.

Both Forum and Abacus showed a close race in Ontario, the Conservatives ahead in B.C., Alberta, and the Prairies, and the Liberals in front in Atlantic Canada. Quebec was the bone of contention.

The projection model is showing some interesting regional fluctuations as well. In British Columbia, the Liberals have now fallen to second place, though they are unchanged at 31%. The Conservatives are up 3.1 points to 31% as well, and are narrowly ahead of the Liberals. Their seat range has improved from 13-18 to 15-22. The NDP has dropped by 2.2 points to 24%, and their range from 10-13 to 6-11.

The Liberals are sliding in the Prairies. They are now at 29%, and have been consistently dropping since mid-December, when they were at 34%. The NDP has benefited, improving from 17% in mid-December to 23% now.

Vote projection in Quebec

In Quebec, the Liberals dropped 1.2 points (and from 20-30 seats to 19-26 seats) to 28%, putting them narrowly behind the NDP, which was up 1.6 points to 28% (and from 29-47 seats to 34-46). The Liberals have been sliding here as well, having had 34% support in mid-October.

But it is the Conservatives who have taken advantage, up from 13% in early November to 20% now. They have pushed the Bloc into fourth place, and are now estimated to be in play in 11-16 seats.

A couple regions to keep an eye on, then. Overall, things remain quite close. The Conservatives are inching up in the national tally, but have actually dropped a little in Ontario. That is what is keeping them from pulling ahead more decisively in the seat projection.

Introducing the 2015 federal election projection model

It is hard to believe that 2015 is now finally upon us. The 2011 federal election seems like eons ago. At the time, the prospect of a majority government, after seven years in which four elections were held, seemed like an endless expanse. But we're now finally in the election year, so it is time to get ThreeHundredEight.com's projection model up and running.

I'll get into analyzing the numbers themselves in the coming days. For now, you can click on the link above or the table above to see the breakdown. For now, let me go over the features of the model, the changes that have been made, and how we got here.

What's been learned since 2011

The roots of the model that will be used for the 2015 federal election, and which has been used in 12 provincial elections that have occurred from coast-to-coast, was first employed in the 2011 federal vote. In some ways, that election was a successful proof of concept. In other ways, it was a failure.

Those who doubt the usefulness of projection models and other critics will undoubtedly point to that failure in 2011. They would not be wrong in doing so, but it would be dishonest. Why? Because the projection model worked very well in 2011.

At least, the seat projection model did. After the election was over, I plugged the results in each region of the country into the model, as if the polls had nailed the call. The outcome was exceedingly close to what actually happened:

The model would have missed the Conservative count by just five seats, and pegged the Liberals and New Democrats to within three. Even the huge breakthrough in Quebec for the NDP would not have thrown the model for a loop - the New Democrats would have been awarded 60 seats and the Bloc Québécois just four (the actual result was 59 and four, respectively).

So the seat projection model proved its ability to translate regional support levels into accurate numbers of seats. This has been proven again and again in provincial elections since. But the call in 2011 was nevertheless missed.

There were two reasons for that. The first was entirely my fault. At the time, I had been basing my model on past elections like those in 2006 and 2008. In each of those elections, the polls in the last weeks of the campaign hardly budged. The swing that occurred over the holidays in 2005-2006 had settled in for some time by the end of that campaign. With those elections as my guide, I calculated that I needed to decay the weight of a poll by just 7% with each passing day.

That was a grave error. It is a system that would have worked well in 2006 or 2008, but utterly failed to capture the late swing away from the Liberals and towards the New Democrats that occurred in 2011. What happened was that, by election day, the projection model was still roughly one week behind where it should have been. Accordingly, it projected a Conservative minority government with 78 seats going to the NDP, 60 to the Liberals, and 27 to the Bloc.

After that disaster, I changed the way the weighting system worked and increased the decay to 35% per day. This has been employed in the past dozen election campaigns and has had no trouble capturing the shifting mood of the public. But that is, of course, if the polls capture it as well. And that is the second reason why 2011 did not go well.

The polls did a moderately good job of things in 2011. They were on the money in Quebec, where the electorate had shifted so dramatically. But at the national level, the polls missed out on the Conservative majority government. With the system in place now, the vote projection model would have given the Conservatives 36% of the vote, instead of the 39.6% they actually got. Both the NDP (31.5% instead of 30.6%) and the Liberals (20.1% instead of 18.9%) would have been over-estimated. The seat projection would have put the Conservatives about a dozen seats shy of a majority government.

This is why, starting with the 2012 Alberta provincial election, I developed a system to estimate likely ranges of support. This system takes into account the potential for the polls to be wrong, and is reflected in the table at the top of the page by the low/high and minimum/maximum ranges. Nineteen times out of 20, the result should fall within the maximum/minimum ranges. There are varying levels of likelihood of the results falling within the other ranges, as explained on the methodological page.

These ranges allowed me to be one of the only people to consider the possibility that Wildrose would not win that 2012 vote.

Starting in 2013, I added another element to the projection model. I found that, too often, people were taking the projected results in individual ridings too literally, as if they were a real indication of support. For that reason I have long thought about not publishing riding projections. But I think it is important to 'show my work', and that is what the riding projections do. They also include ranges that should give people a decent idea of who is and who isn't in play in a given riding.

Still, I wanted to make sure people considered the potential for error. I added probabilities to the projection model, measuring the likelihood that the call the projection model makes in each individual riding will be right.

To take London North Centre, where the Liberals held their two-day caucus meeting last week, as an example, the model currently says that the Liberals would win it with between 43% and 50% of the vote. The Conservatives, at 29% to 34%, are apparently not in play. But the model also says that the Liberals have an 82% chance of winning it. That means that, despite the ranges projected by the model, there is still a roughly 1-in-5 chance that the Liberals will not win the riding.

These probabilities were the last major addition made to the basic projection model, which has been employed with very few changes in the most recent provincial elections in British Columbia, Nova Scotia, Quebec, Ontario, and New Brunswick. A few new minor features, however, have been added for 2015.

What's new in 2015

While none of the basic mechanisms of the model have changed (see the methodological page for complete details on how it works), there are some differences between the 2015 model and the model that was used in prior elections.

First is something that is gone. In past elections, I've applied a "floor-crosser" factor in ridings where a sitting MP has crossed the floor to run with another party. It has had a mixed record. In some cases, the "floor-crosser" factor was quite effective. In others, it was greatly off the mark. I have decided to drop it as a factor that is taken into account. In its place, the party losing a floor-crosser suffers a 'no incumbent' penalty, and the party picking up the floor-crosser gets a 'star candidate' bonus. In past elections, this would have worked better than the system that had been used.

A new factor that has been added to the model, however, takes into account party leaders. My research has suggested that leaders are far more difficult to defeat than normal incumbents (though it can always still happen, of course) and that an MP running as a leader for the first time perform especially better than other incumbents. The loss of a leader as a candidate in a given riding, in addition, is far more penalizing than the loss of the average MP. So, for example, Gilles Duceppe's old riding will be harder to win back for the Bloc Québécois than the average riding they lost in 2011.

The particularities of the 2015 election

As always, an election has its own particularities that need to be taken into account. The idea behind the projection model is that it can be applied uniformly in all jurisdictions in Canada. But the fact is there are always oddities in individual elections that have to be considered.

The new federal riding boundaries is one thing that will make the election potentially harder to call than would have otherwise been the case. But the model has dealt with new boundaries before. In the 2011 Manitoba provincial election, for example, 56 of 57 ridings were called correctly despite the shifting boundaries. For this election, the model merely uses the transposed results that were calculated by Elections Canada. MPs that have shifted ridings are treated as incumbents. Ridings with no incumbents are left unadjusted by the incumbency factor.

The presence of Forces et Démocratie complicates matters. Normally, new parties have been handled by the model without much difficulty. Their support is awarded equally in all ridings within a given region (with the correct regional results, the model would have been within one seat of Wildrose's result in 2012, despite the party fielding candidates for the first time in many ridings).

But no pollster has included FeD in its surveys yet, so we cannot gauge their support. What I have done, until better data becomes available, is to treat Jean-François Larose in Repentigny as I would any sitting MP running as an independent (he retains a portion of his 2011 vote). For Jean-François Fortin, I have simply decided to adjust his 2011 support levels with the Bloc exactly as Jean-Martin Aussant's PQ support in the 2012 election shifted when he led Option Nationale. He seems like an appropriate example to use as a guide.

The floor-crossing of Bruce Hyer from the NDP to the Greens (via a stint as an independent) is also hard to model. Floor-crossing between the 'establishment' parties is relatively straight forward. Not so when it is the Greens, because their base in Hyer's riding was so small in 2011. Treating him like any other floor-crosser would, I think, under-state his likely support. What I've done for him, then, is to adjust his support levels in the same way that Blair Wilson's support shifted when he crossed to the Greens before the 2008 election. Hopefully a riding poll will be done in the riding to give us a better idea of what is going on, as I fear the system in place is also inadequate.

The last particularity of the 2015 election is the presence of an independent candidate like Inky Mark. In 2011, the model did not make any adjustments for the independent candidacy of Hec Clouthier, a former Liberal MP, in Renfrew-Nipissing-Pembroke, who captured a respectable share of the vote. Mark used to be an MP for the riding of Dauphin-Swan River-Marquette (now known as Dauphin-Swan River-Neepawa) and is running as an independent. So, based on past cases of former MPs attempting a comeback as an independent, Mark has been awarded a portion of the support he had in the last election.

Don't miss the forest

As always, I urge readers to exercise caution when looking at the projections. Consider the ranges very carefully. Don't take the individual riding projections as fact. The model simply cannot take into account the effects of local issues in all 338 ridings, and some will be wrong.

Take, for example, the riding of Outremont. It is currently projected to go Liberal, with the NDP in range of winning it. That is Thomas Mulcair's riding. Do I think he is actually going to lose it? Not for a second. But the model does, simply because of the gains the Liberals have made province wide in Quebec.

And that brings up something I am concerned about in the 2015 election. Polls have shown strong gains for the Liberals in Quebec. They have roughly doubled their support in the province since 2011. The model, then, doubles their support in every riding. That means the party is in a good position to sweep Montreal, where they had a decent base in 2011.

But regional polling by CROP has shown disproportionate growth for the Liberals outside of Montreal, meaning that the model may be inflating the party's numbers too much in Montreal and not enough outside of it. This is why, for example, it could be very wrong about Outremont.

The overall seat count for the province, however, is likely to be accurate. Using the polls from CROP, I made projections for the province at the sub-regional level. The end result was not much different. At the riding level, there would be some important differences - more Liberal wins outside of Montreal, fewer around the city. But the sum total of the seat count remains the same - particularly when we consider the ranges. This is why I beg readers not to miss the forest for the trees. The individual riding calls are not nearly as important as the overall regional and national projections.

You might wonder why I don't adjust the model to project sub-regionally in Quebec. The issue with that is that no other pollster is releasing sub-regional numbers for the province. And in 2011, CROP exited the field over a week before the end of the campaign. It would seem unwise to turn the Quebec portion of the model into a sub-regional one, when in all likelihood I won't have information to plug into it.

That about covers everything. I reserve the right to make some adjustments between now and the start of the campaign, though I don't suspect anything but tiny tweaks (perhaps to the values of the 'factors', for example, if I have time to plug some more data points into my calculations). If you notice anything that looks fishy, or even typos, please do alert me to them as errors could have crept in.

Though the model is fully automated, the graphics aren't. They can take quite some time to pull together, so at this stage I'm not sure how frequently I will update the projection. At least weekly, certainly, but updates with every individual poll, as I was doing for the running poll averages, may be too much. And just to make things clear: no, I am not changing the name of the site to ThreeThirtyEight or some variant of that. Consider the name of the site as a historical monument to when it was launched, during an era of seemingly endless minority governments.

Are we heading back to that this year? The current projection, with the Conservatives at between 117 and 155 seats and the Liberals between 107 and 144, would seem to say yes. But more on that later this week.

Final New Brunswick projection: Toss-up

Both David Alward's governing Progressive Conservatives and Brian Gallant's opposition Liberals have a shot at forming government in tonight's provincial election in New Brunswick.

The final polls have suggested that a tightening of the race has occurred in the final week of the campaign, and the very last poll conducted yesterday indicates that it could very well be a tie. We will see tonight whether that is the case, but because of the uncertainty in the polls the result is, right now, a toss-up.

The likely outcome

The projection model considers it almost equally likely that the Liberals or the Progressive Conservatives will win tonight's election.

The Liberals are projected to take between 39.6% and 45.4% of the vote, while the Tories are projected to take between 38.6% and 42.5%. There is quite a bit of overlap there, and the seat projection bears that out. The Liberals could take between 17 and 31 seats, while the PCs could take between 18 and 32 seats.

The precise projection gives the Liberals the edge in the vote with 41.7% to 38.6% for the Tories, but gives the PCs the edge in seats with 26 to 23.

But this is literally as close as it gets. Three seats are projected to go to the PCs with 50% confidence - in other words, a true coin flip. If all of those seats instead go to the Liberals, then the seat count would be reversed, with 26 going to the Liberals and 23 to the PCs. The model is giving the PCs the edge in seats by a collective margin of two percentage points across those three seats.

And they aren't the only close calls. There are three more seats projected with 53% confidence (two PC, one Liberal) and three more with 56% confidence (again two PC, one Liberal). It will not take much to reverse the seat count entirely. This is a projection made with very little certainty of the final outcome.

This might be surprising, considering that the Liberals have led in the polls for over a year - and by nine points in the most recent survey from the Corporate Research Associates. But it is the final poll of the campaign that has injected a large dose of uncertainty into today's result. More on that later.

The projection does not consider a seat for the New Democrats to be likely. With 11.7% support, or between 8.3% and 14.7%, the party is only marginally above where it was on election night in 2010.

Neither are the Greens or the People's Alliance expected to win any seats. The Greens end the campaign projected to take between 4.3% and 7.5%, or 6% more precisely, a relatively good result for them.

Expecting the unexpected

Really, the entire night is an exercise in surprise. But if we extend the ranges to the minimum and maximum levels, encompassing 95% of all likely outcomes, we still see a lot of overlap between the Liberals (7-36 seats) and PCs (12-42 seats).

The NDP's higher range extends to two seats, so the potential for a minority government exists if the seat count is otherwise as close between the Liberals and Tories as the projection currently has it. I would not count out the possibility of an NDP seat, as the party has made major strides forward compared to 2010, when it was almost an afterthought. Though the polling numbers don't suggest the party has made significant inroads, it does seem like the party has done good work to legitimize itself a great deal.

The polls

There have been precious few polls in this campaign. In all, just five were published. In Nova Scotia, a similarly sized province, last fall, there had been 27 polls published (three rolling polls).

In the early stages of the campaign, it did not seem to matter too much. The Liberals were polling around 45%, the PCs around 31%, and the NDP around 15%. There was no major variation that couldn't be explained by the margin of error. But the last two polls have changed things considerably.

The CRA poll of September 15-18 did not show major Liberal slippage, but did suggest that the Tories were making gains with the NDP faltering. From 29%, the PCs were then at 36%. And from 17%, the NDP was down to 11%. Was there something going on?

The poll conducted and published yesterday by Forum seems to suggest there is change afoot. Forum had the Liberals with a 10-point lead on September 11,  but yesterday Forum reported the lead to have vanished: a tie at 40% apiece, with the NDP slipping again to just 12%.

The trends, as you can see above, have pointed at PC gains. (The chart shows the error bars taking into account the margin of error, the trends for CRA and Forum, and the days each of them were in the field.) The Liberals have been slowly dropping - the NDP as well - while the PCs have been moving forward. So this could be for real.

Forum's record in election-eve polling is a generally decent one. In Alberta in 2012, its election eve poll was the only one that suggested Wildrose might not win easily. In Ontario earlier this year, Forum had the Liberals winning a majority (but the Tories too high). They were close to the mark in Nova Scotia in 2013, though so were other polls in the field.

It makes it hard to discount the poll, particularly since CRA was hinting at gains. And this poll's sample of decided voters is almost four times as large: 1,269.

How the leaders fared

It is a rare campaign where all three major party leaders have negative approval ratings, but that has been the case in New Brunswick.

David Alward started the campaign with a -33 rating (27% approval, 60% disapproval) but has ended it at -21 (35% approval, 56% disapproval). This is somewhat in line with CRA's 'best premier' numbers. Alward was at 22% at the beginning of the campaign and 13 points behind Gallant. He ended it at 27% and just five points behind.

Gallant has not had a successful campaign, at least according to the polls. He was a +1 in Forum's first poll, with one-third of New Brunswickers not having an opinion of him (34% approval, 33% disapproval). Yesterday, he was at -10, with an approval rating of 36% and a disapproval rating of 46%. His 'don't knows' dropped by 15 points, but 87% of those people who made up their minds about Gallant went over to the disapproval side. That does not bode well for tonight.

And Dominic Cardy started the campaign as the leader with the best numbers, at +5 (34% approval, 29% disapproval), transitioning to +14 at mid-point (38% approval, 24% disapproval) but plummeting to -6 in Forum's final poll (34% approval, 40% disapproval). This suggests he may have trouble getting a ballot box bonus. By the end of the campaign, only 7% told CRA that he was their choice for premier. He was at 17% in May 2013.

Alward should have no trouble being elected in his riding of Carleton. The model gives it to him with 96% confidence, with between 58% and 67% support vs. 24% to 27% for the Liberal candidate.

Gallant, too, should be elected without issue in Shediac Bay-Dieppe. It is projected to go to him with 90% confidence, with between 51% and 58% of the vote to 28% to 32% for the PC candidate.

The model does not consider Dominic Cardy a likely winner in Fredericton West-Hanwell, which is considered to be a PC seat with 67% confidence. The model puts Cardy in third with between 15% and 26% of the vote, against 37% to 43% for the PC candidate. But I would consider this riding one that could go over to the NDP nevertheless.

David Coon is projected to take between 13% and 22% of the vote in Fredericton South, potentially placing as high as second. As Coon was in the leaders' debate for the Green Party, he could even outperform this tally.

Few polls, much uncertainty

This election was supposed to be an easy one. In the year before the vote, the Liberals were enjoying a lead of almost 20 points, and that lead seemed to be holding through to the end of August. The race tightened one the campaign got going, as is usually the case, but the last set of polls has thrown all expectations to the wind. And it shows why at least one more poll would have been helpful to give us a clue of whether CRA or Forum is on the mark.

Consider two scenarios. The first assumes that CRA was accurate and that voting intentions remained static over the last few days. Forum's final poll, then, was perhaps a victim of the margin of error. My model, based solely on the CRA poll, would give the Liberals 31 seats and the PCs just 18 (or a range of 23-35 for the Liberals and 14-26 for the PCs). In other words, a relatively certain result.

The second scenario assumes that Forum is right, and that CRA was registering the first signs of PC gains. The model then gives the PCs 28 seats to 21 for the Liberals (or a range of 21-35 for the Tories and 14-28 for the Liberals). A closer result, but more clearly a PC victory than the current projection.

Instead, we are somewhere in between in the worst of worlds. Considering the nine ridings projected with 56% confidence or less, a gust of wind could easily make the outcome change from a 26 to 23 split in favour of the Tories to a 30 to 19 split in favour of the Liberals. A gust of wind in the other direction could instead transform the close result into a comfortable PC victory of 28 seats to 21. It makes for an exciting night, but a terribly nerve-wracking one for me - and more importantly the candidates with their futures at stake tonight.

Liberals lead in final days as PCs close gap

Though the race is tightening, the Liberals under Brian Gallant appear on track to form New Brunswick's next government when voters go to the polls on Sept. 22.

The latest numbers put the Liberals in the lead with 45 per cent, against 36 per cent for David Alward's governing Progressive Conservatives. The New Democrats were at 11 per cent, with the Greens in fourth at 6 per cent.

You can read the rest of the article on CBC.ca.

The projection has changed only a little with the addition of the CRA poll, and primarily at the margins. Liberal support upticked slightly from 44.3% to 44.8%, while the Tories were up 1.6 points to 35.3%. The NDP was down 1.8 points to 11.9%.

In terms of the seats, the Liberals and Tories swapped one in the projection. The Liberals are now projected to win 31 and the Tories to win 18.

But the likely ranges are more interesting. From 28 to 37 seats for the Liberals, they have dropped to between 25 and 35. That puts them just above the minimum threshold for a majority government. The Tories have gone from 11 to 21 seats up to 13 to 24 seats.

And the drop in NDP support has resulted in the party's maximum falling from five seats to two seats.

We should expect to hear from Forum on Sunday night, as Forum tends to put out numbers on the eve of an election. If that occurs, I will try to update on Sunday night as well and have a full analysis on Monday. I will be looking for signs that the PCs are indeed making gains.

Liberals still lead in New Brunswick, and some news

The projection for the upcoming New Brunswick election, approaching fast on Monday, has been updated with the latest poll from Forum Research. The Liberals continue to lead, though the gap is narrowing. But before getting to that, a little news.

I'm excited to announce that I am joining the CBC News political team. I'll be writing for CBC.ca as well as appearing on television and the radio, covering polls and projections in the run-up to the 2015 federal election. It will be a lot of fun, and as the vote gets closer and closer we should have some great stuff for you.

My first article for the CBC, looking at the lay of the land as the House returns, can be found here.

It will be business as usual for the time-being on ThreeHundredEight.com. I'll be posting links to my articles and appearances here, in addition to keeping you up-to-date on the latest polls, as always.

I'd also like to thank The Globe and Mail and The Huffington Post Canada for the great opportunity they gave me to write for them since 2010 and 2011, respectively.

Now to New Brunswick.

It has been a quiet campaign on the polling front. Forum's poll, conducted on September 11, is the first set of numbers we've seen since the Corporate Research Associates poll that left the field on August 31.

The projection has narrowed as a result of this poll, with the Liberals down 3.1 points to 44.3%, or between 42% and 48% support. The Progressive Conservatives have gained four points, hitting 33.7%, or between 32% and 37%. The New Democrats took a hit of 2.7 points, falling to 13.7%, or between 10% and 17% support.

The Liberals are now projected to win between 28 and 37 seats, down from the 32 to 41 seats from the last projection. The Tories are now up to 11 to 21 seats from eight to 15, while the NDP is now projected to win only zero to one seat.

Of note is that the maximum and minimum ranges now overlap between the Liberals and Tories, meaning a PC victory is now plausible. It requires a big miss by the polls, however.

The Forum poll showed the Liberals still leading with 42%, down four points from their previous survey of August 25. The PCs were up just one point to 32%, while the NDP was down two points to 13% and the Greens one point to 6%.

This poll pegged support for other parties to be 7%, a gain of six points and the only shift that is outside the margin of error.

There is only one other party on the ballot, that being the People's Alliance of New Brunswick (PANB). There are a smattering of independents as well. Has the PANB made significant gains?

We can't say for certain whether respondents were intending to vote for the PANB or an independent candidate, though in all likelihood the majority of them intend to cast a ballot for the PANB (the party has 18 candidates, compared to nine independents). But therein lies a problem. Most New Brunswickers will not be able to cast a ballot for the PANB. Were the poll respondents aware of that?

With a PANB candidate in just 37% of New Brunswick's 49 ridings, many voters may not have yet realized that they will not have the option to vote for the PANB. That alone, then, could drop that 7% support to just 2.6%, or 3.6% if we make allowance for the 25 ridings that have either a PANB or independent candidate on the ballot.

This is one reason why the projection shows such low support for other parties. My methodology for estimating the support of smaller parties is to focus on how they performed in the last election and the number of candidates running in the current one. This method has been very successful in the past. The PANB does not have enough candidates to net 6% or 7% of the vote. That would require them to average about 19% in every riding. In 2010, the party averaged 4.6% per riding, or 3.4% if we exclude leader Kris Austin's notable performance.

Perhaps the debate, in which Austin took part, has boosted his party considerably. But even that looks unlikely. Support for other parties stood at 7% among anglophones, but also 5% among francophones. How plausible is it that the PANB, a party that has a reputation (deserved or otherwise) of lacking respect for francophones, and which is led by a leader who does not speak French and so could not participate in the French-language debate, is polling almost as strongly among francophones as it is among anglophones?

The minimum/maximum range for "Others" in the projection should suffice to capture the PANB's support.

Also of interest in this poll were the results for the three main leaders. Brian Gallant, whose party is ahead by a wide margin, is not polling as well as he was at the end of August.

His approval rating dropped by five points to just 29%, while his disapproval rating ballooned to 46% from 33%. Even among his party's supporters, his approval rating is just 62% and his disapproval rating is 19%.

David Alward's numbers held steady, at 26% approval and 63% disapproval. Those are not enviable numbers by any stretch, but they are no longer so horrible compared to Gallant's.

Dominic Cardy of the NDP is doing better. His approval rating was up to 38%, but more importantly his disapproval rating was down five points to 24%. And for all the noise about the direction that Cardy is taking the NDP, just 5% of NDP voters said they disapproved of him.

But liking Cardy is not the same as voting for him, as the party's drop in support attests. And when respondents were asked how they rated the leaders on specific issues, Cardy was still behind the others.

On who would best handle the budget, Gallant edged out Alward with 27% to 26%, with Cardy at 18%. On who could best bring jobs and growth to the province, it was Gallant at 31% to Alward with 28% and Cardy at 12%. Only on being able to cut waste was Cardy competitive.

Though Gallant was more comfortably ahead on ethics and vision, that he was only a few points up on Alward on economic questions should be of some concern for the party. Alward may be too unpopular to make major inroads, but if Cardy can eat into Gallant's support the race could get a lot closer.

Final Quebec projection: Liberal victory

Philippe Couillard's Quebec Liberal Party is on track to win tonight's provincial election in Quebec and form a new government. Pauline Marois's Parti Québécois should form the Official Opposition, while François Legault's Coalition Avenir Québec and Françoise David's Québec Solidaire should retain their positions as the third and fourth parties, respectively, in the National Assembly.

The final projection suggests that the Liberals are very likely to win a majority government, though a minority government is still a distinct possibility. All of the charts below can be magnified by clicking or tapping on them.

The likely outcome

The Liberals are projected to win between 60 and 78 seats, putting them mostly over the 63-seat mark needed to form a majority government. They should take between 38.5% and 44.2% of the popular vote. Their best performance since 2003, when Jean Charest first won a majority government, is thus possible. The precise projection is for the Liberals to take 40.1% of the vote and win 69 seats, their best result since 2008 and 2003, respectively.

Click to magnify

The Parti Québécois should finish second with between 38 and 54 seats, capturing between 26.1% and 29.6% of the vote. In terms of seats, the party should not do worse than its 2007 performance under André Boisclair, but could take its lowest share of the vote since 1970, when the PQ took 23% in its first electoral outing. The projection gives them a precise total of 26.9% of the vote and 45 seats. The former would be their worst performance since 1970.

Remaining in third place should be the Coalition Avenir Québec, with between five and 13 seats and between 21% and 24.4% of the vote. More precisely, the CAQ is expected to win 22.8% of the vote and nine seats. That is a drop from their 2012 performance of 19 seats and 27.1% of the vote, but relative to the past results of the Action Démocratique du Québec, its predecessor party, it still represents their third-best result since 1994.

Québec Solidaire will finish fourth with two seats and between 7.2% and 8.4% of the vote (or 7.9% more precisely), which would make this campaign their best result on record.

The provincial Greens, who have fielded only 44 candidates out of a possible 125, are projected to take between 0.5% and 0.7% of the vote. Option Nationale, which has put up almost a full slate, should take between 0.4% and 0.7% of the vote. Other parties and independents are projected to take between 0.5% and 1.5% of the vote (the Conservative Party of Quebec, the party fielding the most candidates in this category, will likely finish first in this group). No seats are projected to be won by any candidate other than those running for the four major parties.

Expecting the unexpected

Recent polling performances have taught us to approach elections with a great deal of caution, and tonight's vote should be no different. In 2012, the Liberals were under-estimated in the polls to a great degree, enough to turn a potential slim PQ majority into a very slim PQ minority. An exact repeat of 2012 is unlikely, as the Liberals are no longer a government in free fall, a role the PQ is now filling. Nevertheless, some sort of miss is still more than possible.

Ranges tracking

The maximum and minimum ranges take that into account (though even here, they are designed to capture 95% of potential results). At these ranges, the Liberals could win between 48 and 83 seats and between 36.1% and 51.4% of the vote, while the PQ could win between 29 and 63 seats and between 25.8% and 34.4% of the vote.

This does open up the possibility of the PQ finishing first in the seat count and second in the popular vote, but it would require a miss of the same proportions as we saw in Alberta and British Columbia. Considering how the campaign unfolded in Quebec, such a result would not only be surprising from a statistical point of view but from a completely subjective one as well. The uncertainty that prevails at the end of this campaign is more in terms of how far the PQ may fall, rather than whether a Liberal victory is in the cards.

That is because of the gains the CAQ has been making in the last two weeks of the campaign. They could win between four and 23 seats and between 18.7% and 26.7% of the vote. As the party has seen its support increase in every new poll since March 19, it makes it very likely that the CAQ will end up on the high side of the projection. Indeed, the CAQ could end up with more seats than they had at dissolution.

For Québec Solidaire, there is far less potential for a major surprise, ranging at between 6.5% and 9.2% of the vote and between two and three seats. That all points to improvement over 2012's result, however.

The Greens and Option Nationale, even at their maximums, are not expected to be able to do much better than their projected result (and could even fall to as little as 0.3% of the vote). Fatima Houda-Pepin in the riding of La Pinière, however, is considered capable of pulling off an upset at the extreme edges of the projection.

Regional breakdown

The Liberals are projected to finish first in the four regions defined by the model: the island of Montreal, the suburbs of Montreal, the Quebec City region, and the rest of the province.

Regional breakdown

On the island of Montreal, the Liberals are projected to have between 50% and 57.3% support, enough to give them 20 or 21 seats. The island is a rather static place in Quebec, where few seats change hands. Even at the extremes, the Liberals are projected to still win between 19 and 21 seats. The PQ should finish second with between 19.1% and 21.6% of the vote, taking five to six seats, while QS should take between 11.7% and 13.6% of the vote and two seats. The CAQ could edge out QS for third in the popular vote, with between 11.9% and 13.9%. The region is where QS could win its third seat, and it will be interesting to see how much of the vote the party takes here. Polls have given them as much as 19% support, which would easily net them four seats, but also as little as 9%, which is less than the 12% the party captured in 2012.

Tracking regional support

In the off-island suburbs of Montreal, the Liberals hold their narrowest lead with between 32.9% and 37.7% of the vote to between 29.5% and 33.5% for the PQ. This should net the PQ between 14 and 17 seats, while the Liberals take between 12 and 14 and the CAQ between zero and four. This is a bit of a swing region, as the extremes expand to 9-21 seats for the PQ, 8-16 for the Liberals, and 0-8 for the CAQ. The results for the CAQ are worth keeping an eye on, as this is the region of Quebec where they have made the most gains since the mid-point of the campaign.

Riding projections, A-L

The Quebec City region is a two-horse race, with the Liberals between 37.2% and 42.7% of the vote, compared to between 31.5% and 36.7% for the CAQ. The Liberals should win between six and eight seats, while the CAQ should take between three and four. The PQ can win none or one. This is another swing region, but primarily for the CAQ. They could win as many as seven seats, dropping the Liberals to three.

Riding projections, L-Z

In the regions of Quebec, the Liberals should take between 35.8% and 41% of the vote and win between 22 and 35 seats. The PQ should capture between 30% and 34% of the vote and between 19 and 30 seats, while the CAQ should take between 21% and 24.5% of the vote and between two and five seats. This is the most important 'region' of Quebec in terms of deciding the electoral outcome (not surprisingly, it has the most seats in the model). This region alone could drop the Liberals to 62 seats and into a minority situation. The major electoral battles will occur in the swathe of territory between Montreal and Quebec City: the Mauricie, Centre-du-Quebec, and Estrie. More than a few three-way races could occur here.

The polls

Polling in Quebec has been quite consistent, with pollsters generally agreeing on the state of the race and the trends that were developing. At the beginning of the campaign, the PQ and Liberals were neck and neck as the CAQ was well behind. But at the mid-point of the campaign the PQ began to falter, as the three other parties inched upwards. The Liberals held steady but the PQ plunged, as the CAQ increased its support significantly. If the campaign had been a week longer, the PQ may have dropped into third. That could potentially still happen.

Polls representing at least 99% of the weighted average

The last set of polls have been rather consistent. With the exception of the Forum Research poll conducted on April 3, all of the final polls (looking at likely voters, when available) have had the Liberals between 38% and 40% of the vote, a rather tight grouping. The PQ has been between 27% and 29%, which is just as tight. But the CAQ has been harder to peg. Ipsos Reid had the party at 18% among likely voters in their poll ending April 1, while the polls from EKOS, Léger, and Forum ending on April 3 had the CAQ at either 21% or 23%. The poll by Angus Reid finishing April 4 had the party at 25%. Is that a normal spread caused by the margin of error, or are these steady gains being made by the CAQ? Did those gains plateau, or would polling conducted on April 6 have shown another two or three point gain for the party?

Campaign polling trends

Support for QS has been mixed as well, with the party at between 6% and 12% in the final polls. The bulk of the surveys, however, put the party under 10%, and considering its historical tendency to under-perform the polls, that is the safest bet.

But the polls have been generally even throughout the campaign, with little real controversy or outliers. It has been interesting, however, to see the polling being dominated by firms from outside Quebec. Léger was the only Quebec-based firm to be polling up to the end, while the other surveys in the last week came from Ipsos Reid, EKOS, Forum (all Ontario-based) and Angus Reid (B.C.-based). CROP dropped out of the field, publicly at least, after their poll of March 13-16. CROP had been very active in 2012, so that was unusual. There was also a great lack of riding-level polling, with only two surveys being released during the entire campaign, after one in five ridings were individually polled in 2012.

How the leaders fared

Both Philippe Couillard and François Legault have seen their numbers improve throughout the campaign. The most frequently polled question, on who would make the best premier, showed gains for both leaders.

Couillard had started the year at between 22% and 25% on this question, but as the campaign progressed that increased to between 26% and 27% to 30% in the last Angus Reid and 33% in the final Ipsos Reid. Legault went from 13% to 15% before the campaign began to between 23% and 25% at the end of it, putting him either in a tie or ahead of Pauline Marois. The incumbent premier dropped from 30% to between 20% and 25%.

Forum was the only firm to also ask approval rating questions, and here we see growth for Legault again. His approval rating was 32% at the beginning of the campaign, while his disapproval rating was 49%. In Forum's last poll, Legault had an approval rating of 57% to 27% disapproval. He really had a stellar campaign - but most of it in the last two weeks.

Couillard's numbers were stable after an early jump from 34% to 44% approval. His approval rating barely deviated from that to the end of the campaign, while his disapproval rating topped out at 42% in the final week.

Marois's approval ratings dropped insignificantly but consistently, from 34% to 32% and then 31% at the end of the campaign, as her disapproval rating increased from 58% to 65%.

These three sets of numbers tell the story of the election rather well. An early bump for Couillard that held steady (a good way to describe his campaign), and sustained decrease and increase for Marois and Legault, respectively.

Pauline Marois is given a 55% chance of winning her riding of Charlevoix-Côte-de-Beaupré, with between 36% and 41% of the vote against 33% to 38% for her Liberal challenger. One would expect a leadership bonus to make her safe, but we've seen many cases of leaders losing their seats in recent elections (Darrell Dexter in Nova Scotia, Christy Clark in British Columbia, Michael Ignatieff and Gilles Duceppe in the federal election of 2011).

It is a similar situation in Roberval, Philippe Couillard's chosen riding. He lives there, but he certainly did not choose an easy seat for himself. The Saguenay-Lac-Saint-Jean region is not known for its Liberal leanings. The model considers the riding a genuine toss-up, with the PQ given a 50% chance of winning it with 38% to 44% of the vote against 37% to 43% for Couillard.

Oddly enough, François Legault is slated to be the leader most likely to win his seat of the three main parties. And this after he spent most of the campaign expected to lose. He is given a 61% chance of winning L'Assomption, with between 38% and 44% of the vote against 35% to 39% for the PQ candidate.

Françoise David is the safest leader going into tonight's vote, with an 87% chance of winning her Montreal riding of Gouin. She is projected to take between 47% and 54% of the vote, against 27% to 30% for the PQ candidate. Her co-spokesperson, Andrès Fontecilla, is unlikely to be successful in the neighbouring riding of Laurier-Dorion. Gerry Skalvounos of the Liberals is given a 73% chance of winning, with between 38% and 43% of the vote against 25% to 28% for Fontecilla. I would bet on a better performance for Fontecilla, however.

A campaign that shows the importance of campaigns

Most commentators, including yours truly, expected the Parti Québécois to be successful in its quest for a majority government. The PQ was dominating among francophones and Couillard was starting to look like he had the leadership instincts of an Ignatieff. The CAQ was being squeezed out of the debate on the charter. But in the end, a mismanaged, chaotic, and inconsistent campaign sunk Pauline Marois and the PQ. As soon as the focus turned to a potential referendum, the Liberals made headway as the best anti-PQ option. And as soon as the PQ started to flail wildly as its poll numbers dropped, Legault began to look like the better anti-Liberal option. It was a disaster for the PQ from start to finish.

This is what makes the last polls of the campaign believable. One could argue in Alberta that Wildrose did not have the feeling of a government about to come to power, or that in B.C. the New Democrats had not really waged a winning campaign. It would be shocking if the PQ somehow managed to win this election tonight, after itself giving every indication that the party was on track for defeat. The behaviour of the party over the last few weeks has not been the behaviour of a party doing well.

That does not make tonight's vote a foregone conclusion. There are numerous factors to take into account:

The Liberals have only been out of power for 18 months. Are Quebecers ready to give them a majority government? Will voters who want nothing of a PQ victory but still feel uncomfortable with the Liberals go over to the CAQ, now that a PQ re-election is no longer likely?

Will the PQ manage to convince enough sovereigntists not to split the vote in order to prevent a majority victory by the Liberals?

The CAQ's voters have been recorded in poll after poll as being much less committed to their party than supporters of the PQ and Liberals. Will those voters waver when it comes time to cast a ballot? Does the CAQ have enough organization to get out its vote? Will the momentum that has been carrying Legault and the CAQ forward in the last few weeks continue and push the party to new heights?

And will QS be able to better its historical under-performance of the polls, concentrating its vote in the right places? Or will their high hopes be squashed again?

We'll find out tonight. It has been a whirlwind of a campaign, so plenty could still have occurred in the final days after the pollsters left the field. But Philippe Couillard and his Liberal Party appear set to prevail.

Angus Reid hints at further CAQ gains

What could end up being the last poll of the campaign (never say never) was released yesterday by Angus Reid, showing the same numbers for the Liberals as other polls but a tight race between the Parti Québécois and the Coalition Avenir Québec for second place. Was Angus Reid at the top end of the consensus of where the CAQ stands, or did Angus Reid record further gains for the CAQ? And, perhaps the most important question going into tomorrow's vote, can the CAQ continue its momentum to move into second?

The projection does not consider it a likely possibility. The maximum support for the CAQ and the minimum support for the PQ estimated by the model puts them in a tie at 26.5%. But that would still translate into no fewer than 32 seats for the PQ and no more than 21 for the CAQ. (Recall, however, that the maximum and minimum confidence bands include 95% of scenarios.) But that we're talking about the possibility that the CAQ could actually emerge out of this campaign with more seats than it had at dissolution speaks to the tremendous gains the party has made in the last two weeks.

The Liberals are still well ahead in the projection. While it leans towards the party capturing a majority, the likely ranges still envision as few as 59 seats for the Liberals. That puts them under the 63-seat mark needed for a majority government.

Let's move on to the Angus Reid poll. My detailed analysis of the final projection will  be posted tomorrow morning.

This is the first outing from Angus Reid in this campaign, and for several years in Quebec. The company did put out some numbers in 2008 and did well enough, however.

Angus Reid released only their estimation of likely voters, so we do not know what they got for all eligible voters.

Among likely voters, the Liberals managed 39%, where they have generally been in all polls conducted over the last two weeks. The PQ was at 27% (note that the four previous polls have had the PQ at between 26% and 29%), while the CAQ was at 25%. This is where Angus Reid stands out, as it is the best number the CAQ has recorded in any poll in over a year. But it also continues a rather consistent trend for the party. Since Forum pegged the CAQ at 13% on March 19, there has not been a poll where the CAQ's numbers have not increased.

Québec Solidaire was at just 7%, on the lower side compared to other recent polls.

Angus Reid reported a tight race among French-speakers, with the PQ at 31% to 30% for the Liberals and 28% for the CAQ. Note that Angus Reid appears to use a similar definition for language as Ipsos Reid does. Ipsos found similar numbers in their recent poll. It is inappropriate to compare the language results of Ipsos and Angus, however, to those of CROP and Léger.

Regionally, Angus Reid split up the province by postal codes. In Montreal and Laval, the Liberals were ahead with 48% to 28% for the PQ, 13% for the CAQ, and 9% for QS. This suggests higher PQ support, and lower QS support, than some other polls have recorded in the area.

In eastern Quebec, which encapsulates Trois-Rivières and everything east of it, the Liberals and CAQ were tied at 33%, with the PQ at 26%. As this region includes PQ strongholds like the Lac-Saint-Jean and the Gaspésie, this implies strong Liberal/CAQ numbers in Quebec City and west of it.

In western Quebec, which encapsulates everything west of Trois-Rivières, with the exception of Montreal and Laval, the Liberals led with 36% to 28% for the CAQ and 27% for the PQ. This implies rather strong numbers for the CAQ in the suburbs of Montreal.

But as these definitions are different from those of other firms, it is difficult to draw significant conclusions from them.

Angus Reid did have some other interesting numbers, however. Liberals and PQ supporters remain the most decided (80% and 75%, respectively), while little more than half of CAQ and QS voters have made their final choice.

On who would make the best premier, Philippe Couillard was ahead with 30% to 23% for François Legault and 20% for Pauline Marois. This corroborates the numbers published by Léger, which similarly showed a strong result for the CAQ leader.

Angus Reid also asked respondents whether their opinion of the leaders had worsened or improved during the campaign. For my part, I found the breakdown by 2012 vote the most revealing.

Liberals seem very pleased with Couillard, as 56% of the party's 2012 supporters said their opinion of Couillard improved during the campaign, against just 5% who said it worsened. The numbers are virtually identical for 2012 CAQ voters and Legault: 57% improved to 5% worsened. Françoise David also had similarly positive numbers among QS voters: 40% said their opinion improved against just 4% who said it worsened.

But Marois performed much less well. Only 21% of PQ voters from 2012 said their opinion of her improved during the campaign, while 20% said it worsened. And when we look at her target voters, we see why the PQ is on track for a defeat. Among people who voted CAQ in 2012, only 4% said their opinion of Marois improved during the campaign, while 63% said it worsened. That is enormous. Among QS voters, just 4% said their opinion of Marois improved, while 56% said it worsened. This has made it impossible for the PQ to gain new supporters during this campaign, and it also suggests how the PQ can be dropping from their 2012 performance.

Couillard, on the other hand, has done rather well among CAQ voters: 35% said their opinion of him improved, while 25% said it worsened. Legault had positive scores among Liberal voters (34% to 11%) and PQ voters (34% to 19%) as well, while David had a positive 27% to 10% split among PQ supporters. In other words, Couillard, Legault, and David all had success in pulling support from the kind of voters they needed to target to make gains. Marois had virtually none.

These numbers seem particularly relevatory, and hint at just how difficult it would be for the PQ to make the gains they need in these final days of the campaign. Marois's three main opponents, on the other hand, have good reason to believe they might be able to attract some waverers to their side. There is still much that could happen between this poll leaving the field on April 4 and voters casting their ballots on April 7. But a PQ comeback seems the least likely to occur.

EKOS, Léger show new CAQ strength

Billed as the poll everyone was waiting for, Léger's final poll of the campaign for Le Journal de Montréal this morning showed significant gains for the Coalition Avenir Québec, as the Parti Québécois takes a further step backwards. But an EKOS poll for iPolitics.ca the day before showed similarly strong numbers for the CAQ, while an Angus Reid poll just released in the last hour shows the CAQ at even higher levels of support. These final days of the campaign are turning into a whirlwind.

Unfortunately, the Angus Reid poll emerged too late for it to be added to the projection, so it will need to wait until tomorrow morning's update. But the EKOS and Léger polls do give us an indication of where things stood up to April 3. The Liberals have held mostly steady in the projection, though they have slipped slightly to 39.1% support (or between 38% and 43%). The PQ is also down a little to 28.1% (or between 27% and 31%), while the CAQ made another big jump to 21.2% (or between 20% and 23%). Québec Solidaire dropped to 9.3% (or between 9% and 10%).

In terms of seats, the Liberals fell from 72 to 69, while the PQ increased from 46 to 47 and the CAQ from five to seven. The ranges have changed more significantly, however, with the Liberals now projected to win between 58 and 75 seats and the PQ between 41 and 58. Thus, a Liberal minority is more likely than it was in the last update, while a tie can now be envisioned. The CAQ increased to a range of between 5 and 11 seats. The maximum and minimum ranges for the CAQ and QS now extend to 18 and four seats, respectively.

Léger was last in the field on March 21-23. Since then, the Liberals have dropped two points to 38%, while the PQ has fallen four points to 29%. The CAQ picked up eight points to reach 23%, while QS was unchanged at 9%.

That surge for the CAQ is enormous, and most definitely outside the margin of error (of a probabilistic sample of this size). The PQ's drop is similarly notable.

The shift came among francophones, as the PQ fell five points to 35% while the CAQ gained 10 points to reach 27%. The Liberals were mostly unchanged at 29%, while among non-francophones the party had 71% support to 11% for the CAQ.

Regionally, the Liberals suffered losses in the Montreal region (down six points to 39%) and Quebec City (down eight points to 35%), but were steady in the rest of Quebec with 37%. The PQ was down slightly in each of the three regions, to 28% in Montreal, 20% in Quebec City, and 32% in the rest of the province.

The CAQ made large jumps in the Montreal region (up 11 points to 22%) and Quebec City (up 15 points to 36%), where they narrowly edged out the Liberals for the first time in this campaign. The party was up more marginally in the rest of the province, to 22%.

However, the CAQ's supporters remain the softest. While 83% of Liberals and 85% of PQ voters said their choice was definitive, only 67% said the same of the CAQ (and 56% of QS). So, there is still the potential for movement towards the two main parties.

But François Legault has certainly made gains. He went from 16% to 24% on who would be the best premier, putting him ahead of Pauline Marois (who fell to 23%). Philippe Couillard was also down, by four points to 27%. Nevertheless, a majority of Quebecers think the Liberals will form the government.

The EKOS poll certainly suggests as much. EKOS shows the Liberals leading with 40% support, followed by the PQ at 26.3%, the CAQ at 21%, and QS at 9.6%. While it is the firm's first foray into Quebec in this campaign, it does jive with other surveys.

Among likely voters, the numbers change only marginally: 39.8% for the PLQ, 27% for the PQ, 21.1% for the CAQ, and 9.4% for QS. These were the numbers used for the model.

EKOS did not report a breakdown by language, but I did manage to obtain a regional breakdown. As EKOS polls by telephone, their breakdowns were by area code. While the 514 area code corresponds with the island of Montreal, the others do not fit in so well with the regions used by Léger and CROP. The 450 area code encompasses the suburbs to the north and south of Montreal, but also much of Montérégie. The 418 area code includes Quebec City, but also the rest of eastern Quebec north and south of the St. Lawrence. And the 819 area code stretches from Abitibi to Trois-Rivieres and Sherbrooke.

The Liberals led in all four regions, though the 450 was far closer: 30.3% for the Liberals against 28.8% for the PQ and 27.4% for the CAQ. QS had its best showing in Montreal with 15%, echoing the high numbers recorded by Ipsos Reid's most recent poll. The Léger poll also implies about that much support for QS on the island, which puts them in range of three or four seats.

EKOS did have an interesting breakdown by place of birth. Among Quebecers born in Canada, the Liberals were ahead with 37% to 28% for the PQ and 22% for the CAQ. Among immigrant Quebecers, however, the Liberals had 70% support to 10% for the PQ and 7% for the CAQ.

The poll also included an interesting snapshot of the federal voting intentions of Quebecers by support for each provincial party. Provincial Liberals were largely federal Liberals (68%) or Conservatives (18%), while PQ supporters were primarily from the Bloc (59%) or the NDP (22%). The most interesting result, however, was the breakdown of CAQ voters: 42% NDP and 29% Conservative. It is difficult to fathom how such diametrically opposed federal parties could attract the same provincial supporters. It may speak to Quebecers appetite for a 'third way', rather than anything left-right or sovereigntist-federalist.

Clearly, the Liberals are now the overwhelming favourites to win on Monday. The PQ's campaign has faltered and there looks to be little likelihood of a significant rebound. But the CAQ is making up some serious ground, which Angus Reid also seems to be suggesting. If the CAQ continues to make inroads, the result on Monday could indeed be a bit of a surprise.