Poll Projection: Left-Leaning Jagmeet Singh to Share Power with Trudeau in Canada

Late breaking momentum for the New Democratic Party (NDP) and its leader Jagmeet Singh according to Nanos, one of Canada’s most respected polling firms, spells further trouble for Prime Minister Justin Trudeau. In almost all events, Trudeau appears headed toward a second term as Prime Minister of Canada when polls close across Canada this evening. Voters, however, look set to chasten the previously sunny centrist after a year of corruption and racism scandals that have dominated Canadian, and sometimes international, headlines.

The end result, as projected by 10at10 and most other election models, will require Trudeau and his Liberal Party of Canada to share power with Singh. A best case scenario for Singh’s NDP (a 25% vote share in Ontario with twenty-two Ontario seats and almost certainly double that nationwide) would be among the worst for Trudeau. Under such an outcome, Trudeau might have to bargain with several parties to maintain a bulwark against the possibility of Andrew Scheer’s Conservative Party of Canada cobbling together a slim majority with Yves-François Blanchet’s Bloc Québécois.

While some reliable polling firms and credible poll aggregating forecasters are now making the case for a possible Liberal majority, the 10at10 forecast, focused solely on Ontario from which more than a third of parliamentary seats are determined, interprets the final set of polling as nearly eliminating that possibility and requiring a serious overperformance by Liberal candidates on election day. NDP overperformance is more likely given the way polls have moved and several other recent signals.

For much of the six week election period, Liberals seemed to suffer little by way of consequences in polls on account of the multiple photos and videos that emerged of Trudeau in costume in black or brown face. All of that changed, though, after three debates late in the campaign period. Polling has been somewhat volatile over the last two weeks since the English language debate on Monday, October 7 which Singh is widely regarded to have won handily.

Given differences among very good polling firms and forecasters, four scenarios, including our actual forecast, are reasonably possible.

In the Best Case for Andrew Scheer’s Conservatives, the NDP have surged in Ontario, just enough to hold back Liberal candidates in key ridings, but not enough to start taking seats from the few places in which NDP and Conservative candidates are competitive. In this scenario, the small Conservative advantage in advance polling carries over to election day with a small win for Conservatives in popular vote in Ontario and the Liberals shedding thirty seats or more. None of the best pollsters in Ontario, including Conservative leaning Campaign Research or Ipsos (led by Darrell Bricker in Canada), have numbers suggesting this scenario. It is the least likely of the four.

In the Best Case for Trudeau’s Liberals, recently or historically accurate phone-only polling firms Mainstreet and EKOS, led by Liberal data hounds, are correct. Rather than late momentum flowing to Singh and the NDP, undecided voters fearing a Scheer government break Liberal with Conservatives seriously underperforming their advance voting poll numbers and with NDP advances over the last two weeks having stalled or receded, particularly in Ontario. Another name for this scenario might be the 20-30-40 situation where NDP candidates garner approximately 20% of the total vote in Ontario, Conservative candidates 30%, and Liberal candidates 40%. A third Liberal, phone-only leaning polling firm (Forum), somewhat unorthodox in it methods and data presentation, also sees an outcome along these lines. While this case may have a stronger claim than the Best Case scenario for Conservatives and an equal or better chance than the best case for NDP, the 10at10 model sees it less likely given a raft of signals, including the minority report from Nanos (the fourth phone-only polling firm), the average of seven online only polling firms, and Ipsos’ (mixed phone and online) final results.

In the Best Case for Singh’s NDP, they reach 25% of the total vote share in Ontario, but still only capture 18% (22 of 121) of the parliamentary seats available in Ontario. The scenario sees Nanos data signalling two particular shifts — undecided voters breaking for the NDP fairly strongly and previously committed Green voters in Ontario shifting Orange. This set of circumstances would also be welcome by Conservatives. Simply plugging the Ontario numbers here into the nationwide final forecast from Eric Grenier at the Canadian Broadcasting Corporation’s (CBC) would give Scheer’s party exactly enough seats nationwide to combine with 39 Bloc Quebecois seats to form a slim 170-168 parliamentary majority. The odds of that possibility are incredibly low, however, and Trudeau would, in any event, get the first chance to persuade other parties to join him in an “anyone but Conservative” national unity government. A government so arranged, and dependent upon multiple other parties for its survival, would not be likely to last more than a few short years or even months.

While there are some good signs that the NDP surge may re-emerge in strong form on election day, a more holistic view of circumstances and key factors suggests a more modest uptick over the last few days of the campaign for the NDP. An evaluation of twenty key factors such as advance voting totals, engagement from young voters, online versus phone polling dynamics, top-quality polling, economic factors, strategic voting signals, leader approval polling, and more leads us to a 10at10 Final Forecast scenario in which Singh sees undecided voters break his way modestly over the strict average of all final polls, from the twelve firms in the field in the last week, along with some attrition from Green party voters to NDP. These factors would mean a small movement toward NDP over the strict polling average and the pick-up or retention of three seats (Brampton East, Sudbury, and Oshawa) that the NDP might otherwise lose.

While nowhere near enough to see Singh challenge Trudeau for top spot in Ontario or nationwide, NDP strength elsewhere and continuing Liberal challenges, especially in Quebec, mean the 10at10 projection model is almost certain that Trudeau and the Liberals will be required to share power with Singh and the NDP in some form or fashion when all the votes are in and the dust of parliamentary antic and upheaval has settled.

A full riding-by-riding projection for each of these four scenarios will be published here by 2pm eastern time today.

Doug Johnson Hatlem writes on polling, elections data, and politics. For questions, comments, or to inquire about syndicating this weekly column for the 2020 cycle in your outlet, he can be contacted on Twitter @djjohnso (DMs open) or at djjohnso@yahoo.com (subject line #10at10 Election Column).