Philippe J. Fournier: It is not a sure victory for the Liberals, but the final projection of 338 votes in Canada puts them at 146 seats (11 less than in 2019) and the Conservatives at 127.
Since the dissolution of the House of Commons on August 15, pollsters and media organizations across the country have published more than 150 polls. Although each survey has its share of uncertainty and noise, taken together, we were able to spot a few different trends.
On August 15, the Liberal Party (LPC) walked out the door with a clear advantage in voting intentions, but the gap between the LPC and the Conservative Party (PCC) began to narrow from the second day of the campaign. While the Liberals appeared to be on cruise control, conservative Erin O’Toole’s personal numbers began to improve (see this Mid-August Abacus Data Survey). This trend continued until the first debate, held in Montreal for the TVA network. Then the numbers came to a standstill.
After the Federal Commission debates, the most notable movement was detected in Quebec, where the Bloc Québécois, which had been running a tepid campaign up to that point, saw their numbers increase to finally reach the level of party support in the province in 2019, dashing hopes for significant CPC and LPC gains in Quebec.
Meanwhile, in the west, Liberals appeared to slide to third place in British Columbia, although there is no clear consensus on this in final polls.
So this is where we find ourselves with less than 24 hours to go on voting day. As of this writing, on Sunday night, it appears that the Liberal Party and Justin Trudeau are the favorites to win the most seats in Canada, although the gap between the two main parties remains narrow. Several important variables remain unknown, and while there appears to be a consensus on national voting intentions, regional figures, which are crucial for seat projections, differ from company to company. (See the full list of surveys here).
Here are the final polls from polling companies that published at least one poll during the campaign:
With few exceptions, most polls show that liberals and conservatives are two points apart, a scenario very similar to the 2019 campaign when the CCP won the popular vote by one point nationally, increasing the score by Alberta and Saskatchewan, but failing to make a profit in central Canada.
Unsurprisingly, the final 338-vote projection in Canada has Liberals and Conservatives tied at 32 percent each. The NDP ranks third with an average of 19 percent. If the NDP gets your vote tomorrow and translates these favorable voting intentions into actual votes, it would represent a three point increase from 2019, and Jagmeet Singh could definitely hope to increase the size of his group.
Using the aforementioned surveys with the federal 338Canada model, we calculated the following seat projections:
The Liberal Party wins an average of 146 seats, 11 seats less than in 2019, but still within ranges similar to those held by the party two years ago. However, when looking at the regional breakdown of vote projection, this number of seats for Liberals may appear low. Why? Because the LPC is actually projected to lead on 155 seats, but since many of them are pitches or too close to call, it lowers the party’s seat average.
Therefore, the probability density of seats is asymmetric: while the average is 146 seats, the most likely scenario is close to 155-160 seats. To reach that total, the Liberals would have to win virtually every projected cast district in their own way. However, if the polls overestimated the Liberals even at one point (especially in Ontario), this optimistic result for the LPC would likely not materialize:
The Conservative Party wins an average of 127 constituencies across the country, slightly above its total of 121 seats under Andrew Scheer. Although the confidence intervals between the two parties overlap, conservatives have a much lower ceiling than liberals and they too would need high voting efficiency to overcome this. However, a modest survey error against the CPC could skew those figures, especially in Ontario.
Canada’s federal model gives the LPC at least a plurality of seats in 68 percent of the simulations, compared to 31 percent for the Conservatives. We are far from a clear result here. In fact, we could use the analogy of a roll of the dice: one to three, the liberals win a minority; Four, the Liberals reach the majority; Five or six, and it is the Conservatives who win the most seats.
So while one prefers to be the frontrunner heading into Election Day, a two out of three chance of winning is far from a sure win for Liberals. For comparison, according to some sports betting sites, the Las Vegas Golden Knights had a two out of three chance of beating the Montreal Canadiens in the Stanley Cup semifinals last June. AND we know how that turned out.
In closing, let us remind readers of this column that polls do not vote. The support expressed by respondents in the surveys is entirely theoretical. Polls take the pulse of the electorate, but no more. In 2019, federal model 338Canada correctly identified 299 of the 338 winning districts, or nearly 90 percent of the districts. If the polls released in the last few days prove to be accurate on Monday night and throughout the week (don’t forget the mail-in ballots), I’m sure the election results will be generally in line with the numbers in this projection. . However, humans are complex entities, much more so than electrons or galaxies, so any form of prediction about their behavior will always contain a significant amount of uncertainty.
The Liberals are expected to win Monday night. The data tells us that this is the most likely outcome. Also, if polls underestimate liberal support, a majority of LPC would even be a plausible scenario (15 percent probability). And if the polls underestimate conservatives, especially in Ontario, Erin O’Toole’s CPC could tie or even exceed the seat count, but that’s a big “yes.”
Go vote, Canada. Many of you may not have wanted this choice, but it is here. Don’t let others speak for you.
* * *