Angus Reid, who conducts his polls online, also added a "certain to vote" question, and it makes a difference in the poll that could be indicative of what to expect on Monday.
Here are the final polls based just on the ballot question:
|Strategic Counsel||Feb. 27-28||50||25||8||10||8|
|Angus Reid||Feb. 27-28||43||28||13||10||7|
Leger does qualify its findings by noting there is a large number of undecideds, an unusual statement for a pollster to make at the end of a campaign. Strategic Counsel said their research "found no evidence of this".
When Reid applies his "certain to vote" filter it pushes the PC total below 40%.
|Reid Certain Voters||Feb. 27-28||39||30||13||10||8|
The Reid numbers seem to TC to correspond better to the flavour of news coverage coming out of the election, and makes him wonder if even these numbers have captured the full extent of what appears likely to be a low turnout prompted in no small measure by normally PC voters who don't want to support the almost 37 year old government yet again.
TC is inclined to think Reid's numbers will be closer to reality. They still produce a comfortable Conservative majority of 53 seats with 23 for the Liberals, 5 for the NDP and 2 for the Wildrose Alliance. But TC's model should not be expected to work as reliably in this context so the seat numbers need to be taken with more than the usual amount of caution.
As a final word, consider what Alberta's Members of Parliament were reported to have said to Stephen Harper:
...the confided consensus of MPs was Premier Ed Stelmach is about to lose a bunch of seats in Monday's provincial election and, if the large undecided vote shifts to the opposition or stays home, perhaps lurch into the nightmare scenario of forming Alberta's first ever minority government.