Brad Wall continues to lead in new Saskatchewan polls

Two new polls continue to show Brad Wall's Saskatchewan Party with a healthy lead in the province's election campaign, but one of those surveys suggests the New Democrats may be closing the yawning gap.

The Saskatchewan Poll Tracker puts the Saskatchewan Party ahead with a projected 53.5 per cent support, followed by the New Democrats at 34.5 per cent.

• Saskatchewan Poll Tracker: Check out the latest numbers
• Analysis: The challenges parties face as Saskatchewan campaign begins
• Use Vote Compass to compare the party platforms ahead of Saskatchewan's election

The Liberals and Greens trail at length with 8.4 and 3.2 per cent support, respectively.

With these levels of support, the Saskatchewan Party would likely win between 42 and 49 of the province's 61 seats, with the NDP taking between 12 and 19 seats if the election were held today.

This latest update comes following the release of two polls published this week:

• A poll by Forum Research pegged the Saskatchewan Party at 57 per cent support among decided and leaning voters, followed by the NDP at 33 per cent, the Liberals at seven per cent, and the Greens at three per cent.
• A poll by Mainstreet Research for Postmedia showed the Saskatchewan Party at 51 per cent among decided and leaning voters, with the NDP at 37 per cent, the Liberals at 8 per cent, and the Greens at three per cent. Compared to Mainstreet's previous survey conducted on March 1, this represents a drop of four points for the Saskatchewan Party and an increase of four points for the NDP.

The difference in results between Forum and Mainstreet are at the margins, with both surveys pointing to the Saskatchewan Party being on track to win a big majority government. 

But Mainstreet's numbers suggest that support for Brad Wall's party may be shrinking to the benefit of the New Democrats.

This is not the first time Mainstreet has shown the margin narrowing. Polls taken in February recorded a four point decline for the Saskatchewan Party to 52 from 56 per cent support, before the March 1 survey showed a rebound to 55 per cent.

So this improving situation for the New Democrats may prove to be ephemeral. The next set of polls will provide a clue as to whether or not that will be the case. Nevertheless, the NDP at 37 per cent is the highest it has been in Saskatchewan polling since long before the 2011 provincial election.

NDP's Broten ceiling

The polling by Forum, however, indicates that the New Democrats may have a bit of a ceiling.

Cam Broten, leader of the NDP, scored an approval rating of just 28 per cent in Forum's survey, compared to a disapproval rating of 39 per cent. That left 34 per cent of Saskatchewanians unsure — and there is Broten's opportunity. He needs voters in the province to get to know him better, but that is no guarantee that those who are currently unfamiliar with him will like what they see.

He still compares unfavourably to Wall, who had an approval rating of 58 per cent and a disapproval rating of 33 per cent. On who people in Saskatchewan prefer to be premier, Wall beat Broten by a margin of 55 to 19 per cent. That has the potential to put a cap on the NDP's growth prospects.

And growth the NDP desperately needs. Even in Mainstreet's more competitive poll, the NDP was only tied with the Saskatchewan Party in Regina. The New Democrats trailed by 12 points in Saskatoon and 20 points in the rest of the province. 

If the NDP is to put the Saskatchewan Party's majority into a doubt, they have a long way to go yet.

• The Pollcast: Subscribe to Éric Grenier's podcast
• ​Analysis: Marco Rubio's hopes for Republican nomination could be over

The poll by Forum Research was conducted on March 7, 2016, interviewing 904 eligible voters in Saskatchewan via interactive voice response. The margin of error associated with the survey is +/- 3 per cent, 19 times out of 20.

The poll by Mainstreet Research was conducted on March 8, 2016, interviewing 1,536 eligible voters in Saskatchewan via interactive voice response. The margin of error associated with the survey is +/- 2..6 per cent, 19 times out of 20.