Back to a Horserace

The following seat projection drawn from polls conducted in November indicates a substantial tightening of Canadian public opinion reflecting an erosion of the Liberal lead in recent months. While the Liberals still maintain an approximate two percentage-point lead in popular support nationally, their decline in Ontario from a 10- to a two-point margin over the Conservatives since the previous LISPOP projection largely accounts for a modest Conservative plurality in seats. Data presented here are derived from a blended and weighted aggregation of polls released between Nov. 4 and Dec. 1 conducted by Ipsos, Ekos, Leger and Abacus among others, including some 13,000 respondents. It should be noted however that the large Ipsos sample represented well over half the respondents, and has been weighted accordingly. The most obvious conclusion to be drawn from this is that we might well be heading toward a minority government. A heartening observation of this set of polls is that they were basically in regional alignment with each other, meaning there wasn't the unsettling level of variance noted in the previous projection of Oct. 24. 

Projected distribution of seats by party and region compared with actual election results (in brackets), released December 9, 2014

  liberal conservative ndp bq
Other
Canada
123(34)
132(166)
80(103)
2(4)
1(1)
Atlantic provinces
24(12)
5(14)
3(6)
--
--
Quebec
28(7)
8(5)
40(59)
2(4)
--
Ontario
52(11)
52(73)
17(22)
--
--
Prairies & North
7(2)
17(26)
7(3)
--
--
Alberta
1(0)
32(27)
1(1)
--
--
British Columbia
11(2)
18(21)
12(12)
--
1(1)

Note: The "regional swing model" is more fully explained in a paper originally prepared and presented by Dr. Barry Kay to the 1990 annual meeting of the Canadian Political Science Association, entitled "Improving Upon the Cube Law: A Regional Swing Model for Converting Canadian Popular Vote into Parliamentary Seats". It should be noted that the application of the model above does not make use of the "incumbency effect" described in that paper. In tests for past elections, using late campaign polls to project electoral outcomes, the model has proved to be accurate within an average of four seats per party since 1963. Readers interested in post-dictions for past federal elections dating back to 1963, for projections using pre-election polls dating back to the 1980 federal election and for three Ontario provincial elections, may contact me at bkay@wlu.ca.

Level of Government: