Tonight's leaders' debate, the only one of the Ontario provincial election campaign, comes at a significant point in the race. Kathleen Wynne's Liberals and Tim Hudak's Progressive Conservatives remain effectively tied in the polls, while Andrea Horwath's New Democrats have yet to build any momentum. The debate could serve to unlock the status quo, or further solidify it.
The latest vote projection from ThreeHundredEight.com, encompassing all of the polls released to date in the campaign, pegs the Ontario Liberals to have 37 per cent support, one point up on their standing last week and virtually unchanged from their 2011 vote haul. The Progressive Conservatives sit at 36 per cent, a gain of two points, while the New Democrats have dropped three points to 20 per cent support. The Greens are projected to have 6 per cent support.
There is no significant trend in one direction or another for any of the parties. On average, the Liberals and PCs have been wobbling back and forth around the 35 per cent mark since the beginning of the campaign while the NDP has not been able to budge itself out of the low-20s.
The seat result could thus look very similar to the legislature at dissolution. The Liberals would likely win between 41 and 54 seats on current polling levels, a slight worsening of their position over the last week. The Tories' prospects have improved marginally, with a likely haul of between 37 and 48 seats. This means that either the Liberals or the PCs could emerge out of an election held today with the most seats. The debate tonight could go a long way toward breaking the deadlock.
The New Democrats would likely win between 14 and 21 seats based on their current support levels, putting them in range of a repeat of 2011's performance.
The polls converge, but not among likely voters
The great disparity recorded in the polls at the beginning of the campaign has lessened considerably, with the last set of polls released by the four polling firms active in this campaign pointing to either a Liberal lead or a statistical tie between them and the Tories. The Liberals have been estimated to have between 34 and 39 per cent support in the latest polls. That range of five points is half of what it was in the last set of four polls.
Consensus support for the Greens, among eligible voters at least, has been around 7 per cent.
Support for the Tories and New Democrats has proved harder to nail down, however. Four of the last five polls put the Tories at between 34 and 36 per cent, but one poll has them as low as 30 per cent. Three of the last five polls have put the NDP at between 23 and 25 per cent, but two have had them at either 17 or 20 per cent support.
While this makes it more difficult to estimate the precise state of the race, these polls do paint a general picture: the Liberals slightly, but not definitively, ahead of the Tories with the NDP well back.
Among likely voters, however, the picture is far less clear. In an election where a minority of voters may actually cast a ballot, estimating who is going to vote is absolutely essential for pollsters. Each polling firm is using a different method to make that estimation, which introduces a greater element of variation from one survey to the next than the mode of contact (online or via telephone).
Ipsos Reid's latest survey, for example, showed a two-point edge among eligible voters for the Tories but a 12-point advantage for the party among those most likely to vote. Abacus Data, on the other hand, gave the Liberals a two-point lead among likely voters. What the two surveys have in common is that the PCs seem to improve their standing considerably when only those most likely to vote are taken into account. This could prove to be a major factor on election night.
The 2013 provincial election campaign in British Columbia demonstrated the importance of differentiating between voters and non-voters. But it also showed how polling data outside of the horserace can be important. Despite the sustained (and ultimately misleading) lead for the B.C. NDP throughout the campaign, Christy Clark, leader of the B.C. Liberals, polled much closer to NDP leader Adrian Dix at a personal level.
In this campaign, leadership numbers point to an advantage for Kathleen Wynne. In the 11 polls so far released with questions related to who would make the best premier, the Liberal Leader has led or tied for the lead in nine of them (or 82 per cent). By contrast, her party has led or been tied for the lead in 61 per cent of polls conducted since the start of the campaign.
Ms. Wynne has received an average of 31 per cent support on the question of who would make the best premier, far ahead the 25 per cent of Tim Hudak and the 21 per cent of Andrea Horwath. Her approval and favourability ratings have averaged about 10 points higher than Mr. Hudak's in this campaign as well.
In the personal contest of the debate, this would seem to give Ms. Wynne a head start over her two rivals, particularly Mr. Hudak. But tonight may be the first time a lot of Ontarians will give serious thought to the election campaign. For those just starting to tune in, that first impression could be significant.
ThreeHundredEight.com's vote and seat projection model aggregates all publicly released polls, weighing them by sample size, date, and the polling firm's accuracy record. The seat projection model makes individual projections for all ridings in the province, based on the regional shifts in support since the 2011 election. Projections are subject to the margins of error of the opinion polls included in the model, as well as the unpredictable nature of politics at the riding level. Full methodology can be found here.
Éric Grenier writes about politics and polls at ThreeHundredEight.com.