An update to this post was made on April 15, for two reasons. Firstly, the original post failed to include a poll by the Innovative Research Group that had been conducted in March. Secondly, there was an error made in how the maximum seat projection ranges have been calculated since the election. Edits have been made throughout the text and the graphics have been updated as well.
---
Federal polling picked up a little in March, with five national and one Quebec poll being conducted and published throughout the month. In total, over 10,000 Canadians were sampled, and the numbers continue to show robust support for the Liberals.
The Liberals led in March with an average of 45 per cent support, down four points from
February. But they are down only 0.2 points from
January and 1.2 points from
December.
As February had only two national polls, it would perhaps be unwise to draw too many conclusions from a comparison to that month. It might be more accurate to say the Liberals are down slightly from where they were in December and January. It is also their lowest result since the election.
The Conservatives averaged 30.5 per cent support, up one point from February and 2.1 points from January. This was their best score since the election.
The New Democrats were at 13.1 per cent, up 0.6 points from February but down 3.2 points from January — and 6.6 points since the October vote.
The Greens were at 5.5 per cent and the Bloc Québécois at 4.4 per cent, steady numbers since the election. Another 1.5 per cent, on average, said they would support another party or independent candidate.
I won't make any direct comparisons to the regional results in February, as the two polls from that month would have still had very small combined regional sample sizes.
The Liberals led in British Columbia with 47.4 per cent support, a second consecutive month of increase putting the party back where it was in the aftermath of the election. The Conservatives have been wobbling back and forth, and averaged 25.6 per cent in the province. The New Democrats were down again, falling to 15.8 per cent. The Greens were at 9.9 per cent.
This would likely deliver between 29 and 38 seats to the Liberals, with the Conservatives winning between two and 11 and the New Democrats and Greens only one apiece. That is a decrease from last month for both the Conservatives and NDP, and a gain for the Liberals.
In Alberta, the Conservatives continued to lead with 58.7 per cent support, followed by the Liberals at 28 per cent, the NDP at 6.6 per cent, and the Greens at 4 per cent. This would likely deliver 29 to 31 seats to the Conservatives and three to five seats to the Liberals.
The close race in the Prairies continued, with the Liberals and Conservatives tied at 40.8 per cent, enough to give the Conservatives 17 to 19 seats and the Liberals between nine and 11. The NDP was at 11.3 per cent and the Greens at 6 per cent.
The Liberals dropped to a post-election low in Ontario to 46.7 per cent, dropping them to 75 to 93 seats in the projection. The Conservatives were up to one of their highest level of support since the election with 35.4 per cent, enough to give them 27 to 43 seats. The New Democrats were at 12.6 per cent (one to five seats), and the Greens were at 4.5 per cent.
The Liberals were down 4.1 points in Quebec from February (there were three polls in the province that month) to 46.2 per cent, but that would still give them almost all of the province's 78 seats with 63 to 73. The NDP was up 0.6 points to 17.9 per cent, but that would likely only win them one seat. The Bloc Québécois was up 2.8 points to 17.5 per cent (zero to five seats), while the Conservatives were down 0.2 points to 13.5 per cent. That would likely give them five to nine seats. The Greens were at 3.8 per cent.
And in Atlantic Canada, the Liberal voted oscillated back down to 59.9 per cent, followed by the Conservatives at 19.9 per cent, the NDP at 11.3 per cent, and the Greens at 6.5 per cent. This would likely give the Liberals 29 to 32 seats and the Conservatives zero to three seats.
Altogether, the Liberals would likely have won between 211 and 255 seats in an election held in March, well above the 184 seats they won in the election.
The Conservatives would have won between 80 and 116 seats, straddling the 99 seats they won in the October vote.
The New Democrats would win between two and seven seats, well down from the 44 they currently have.
The Greens would have retained their one seat, while the Bloc Québécois would have won between zero and five seats, an improvement over the projected shutout in February.
The maximum ranges take into account big polling and projection misses. But they might also be a good proxy for the impact of a campaign.
In 2015, when the one-election model was in use, the Conservatives began the campaign on August 2 with a projected maximum range of between 83 and 189 seats — so it did envision their eventual outcome.
For the Liberals, NDP, and Bloc, however, it took until October for the maximum ranges to extend to where the parties eventually wound up. So I think it is fair to say the maximum ranges give a window of what two to three weeks of campaigning could do to the polls. In that sense, they give an indication of what outcomes we might expect if we were in the early stages of a campaign.
The maximum ranges currently give the Liberals anything between a huge majority and a very slim ones. The Conservatives would almost certainly finish second.
The New Democrats could best hope to win 16 seats while the Bloc still could not achieve official party status (9 seats), or be shut out (the NDP too).
I had made an error with the earlier projections, as after the election I had forgotten to re-classify the parties. That is why the Liberal lower end was so low — they were being treated like a third party, not like the governing party.
With the chart now corrected, you can see that the Liberals have not been in a position since the election that would put their majority government in doubt. And only in January did the NDP have an outside chance of finishing in second place.
Provincial polling averages
It was a busy month at the provincial level, with new polls in every province but British Columbia.
In Alberta, Wildrose led with 34 per cent, followed by the New Democrats at 27 per cent, the Progressive Conservatives at 25 per cent, the Liberals at 8 per cent, and the Alberta Party at 4 per cent. Though Rachel Notley's governing NDP is back in second, they have been on a pretty consistent slide since the summer.
The March polling in Saskatchewan averaged 57.1 per cent for Brad Wall's Saskatchewan Party, 32 per cent for the NDP, and 6.4 per cent for the Liberals. The result of the election on April 4, however, was 62.6 per cent for Wall's party and 30.4 per cent for Cam Broten's NDP.
The campaign continues in Manitoba and we have already seen some polls conducted in April. But in March, the Progressive Conservatives averaged 44.8 per cent, followed by the NDP at 24.8 per cent and the Liberals at 23 per cent. It was the first time the New Democrats were in second since last summer.
One poll in Ontario showed continued stability in the province, with Patrick Brown's PCs ahead with 40 per cent to 30 per cent for the Liberals and 24 per cent for the NDP.
In Quebec, the Liberals fell to 32.5 per cent, giving new support to the Coalition Avenir Québec, which was up to 23 per cent. The Parti Québécois was steady at 30 per cent, while Québec Solidaire stood at 10.5 per cent.
The Liberals dropped in New Brunswick to 45 per cent support, followed by the PCs at 27 per cent, the NDP at 18 per cent, and the Greens at 8 per cent.
In Nova Scotia, the Liberals were down to 56 per cent, with the PCs up to 23 per cent and the NDP falling to third place to 16 per cent support.
It was steady sailing in Prince Edward Island, wit the Liberals at 61 per cent, the Progressive Conservatives at 19 per cent, and the Greens at 11 per cent.
And in Newfoundland and Labrador, the post-election honeymoon is on with Dwight Ball's Liberals, who were up to 66 per cent. The PCs were at 23 per cent and the NDP at 11 per cent.