Weekly Writ 3/26: What's the NDP's path back?
Should the NDP prioritize the voters they lost to the Liberals or to the Conservatives?
Welcome to the Weekly Writ, a round-up of the latest federal and provincial polls, election news and political history that lands in your inbox every Thursday morning.
Reading this in your email inbox and having trouble viewing any of the charts? To optimize them for your device, click/tap the charts or view them on thewrit.ca.
The New Democrats are choosing their next leader on Sunday and one of that leader’s top priorities will be to re-connect with the voters the NDP lost to the Conservatives in the last election.
Sure, there’s been a broad recognition that the New Democrats have lost a lot of progressive voters to the Liberals who also need to be recovered. But there’s been a great deal of focus on that lost connection with the blue-collar, working-class vote that Pierre Poilievre’s Conservatives have assiduously wooed over the last few years.
But just how key to the NDP’s current woes are those voters?
Whoever the next leader of the NDP will be (fundraising metrics point to Avi Lewis being the overwhelming favourite, but the answer will only be revealed this weekend), they will have a lot of work to do to get the party back to relevancy. With just seven seats and 6.3% of the vote in the April 2025 election, the New Democrats had their worst-ever result. Not even the Co-operative Commonwealth Federation’s first election in 1935 went as badly as Jagmeet Singh’s campaign in 2025.
So, any voters the NDP can get back into the fold will be welcome for the party. But results from the last election suggest that the pool of voters the NDP needs to recover is far more red-tinged than blue-hued.
By a margin of roughly four-to-one, the NDP lost more of its 2021 voters to the Liberals than it did to the Conservatives, according to polling conducted at the end of the 2025 campaign.
Three pollsters (Research Co., Abacus Data and the Angus Reid Institute) included voting intentions results by past vote in their final polls of the campaign. If we average those three polls together, we find that the NDP only retained 39% of their 2021 voter base, with 45% deciding to vote for the Liberals and just 11% voting for the Conservatives.
There was some variation between the three polls, which is understandable considering the smaller sample size of NDP voters and the problems poll respondents frequently have with recalling past voting behaviour. But the results from the three polls were broadly the same — the NDP retained between 32% and 47% of their 2021 vote, while between 37% and 56% was lost to the Liberals and just 7% to 17% went to the Conservatives.
New polling from the Angus Reid Institute points to the same phenomenon, even when we pull back a little further. Among Canadians who voted for the NDP at least once since the 2015 election, the Liberals currently have nearly five times as much support as do the Conservatives. The Liberals also have more support at the moment among these past NDP voters than does the actual NDP.
(I have more analysis on new polls delving into the NDP leadership race and the future for the party below.)
It’s a pretty clear indication of how, when it comes to the sheer number of raw votes at stake, the New Democrats have more to recover from the Liberals than they do from the Conservatives. A successful strategy that is aimed at getting votes back from the Liberals would pay greater dividends than one that prioritizes getting votes back from the Conservatives.
But when it comes to the seat math, things are a little more complicated.
The NDP dropped from 25 to seven seats on election night for a loss of 18. (One of those 18 is hard to place on the map because Northern Ontario lost one seat in redistribution, but for the purposes of this analysis we’ll consider Sudbury East–Manitoulin–Nickel Belt as the successor riding to Algoma–Manitoulin).
In eight of those 18 lost seats, the Conservatives gained a greater share of the vote (even after considering the losses suffered by the People’s Party, which we can presume largely went to the Conservatives) than did the Liberals. While we can’t assume a one-to-one transfer of vote between parties, it is reasonable to conclude that, in those eight seats, the NDP was defeated primarily because it lost votes directly to the Conservatives.
Those eight seats were Burnaby Central and Skeena–Bulkley Valley in British Columbia, Edmonton Griesbach in Alberta, Elmwood–Transcona in Manitoba and Kapuskasing–Timmins–Mushkegowuk, London–Fanshawe, Sudbury East–Manitoulin–Nickel Belt and Windsor West in Ontario.
With the exception of Burnaby Central, which went to the Liberals, the Conservatives won these seats.
In the 10 other seats lost by the NDP, it was the Liberals who gained more than the Conservatives (again, after accounting for the PPC vote). In these 10 seats, we can reasonably conclude that the main reason the NDP was defeated was because of votes lost to the Liberals.
These were Cowichan–Malahat–Langford, Esquimalt–Saanich–Sooke, Nanaimo–Ladysmith, New Westminster–Burnaby–Maillardville, North Island–Powell River, Port Moody–Coquitlam, Similkameen–South Okanagan–West Kootenay and Victoria in British Columbia, Churchill–Keewatinook Aski in Manitoba and Hamilton Centre in Ontario.
It’s notable that so many of the NDP losses caused by Liberal gains were on Vancouver Island, with three of the five seats going to the Conservatives rather than the Liberals. It was a vote split that worked against the NDP that helped elect Conservatives in these ridings because the Liberals were working from too small of a base.
And while the Conservatives won seven of the eight formerly-NDP seats where Conservative gains were greater than that of the Liberals, the Liberals only won six of the 10 seats where their gains were greater than that of the Conservatives. Liberal gains from the NDP in the party’s lost seats were more likely to elect Conservatives than Liberals.
It’s a bit of multi-dimensional chess for the NDP. The NDP lost 11 seats to the Conservatives and seven to the Liberals, but the Liberal gain was greater than the Conservative gain in 10 seats against eight where the Conservative gain was greater. This means it isn’t necessarily the party-banner of the incumbent that should decide the NDP’s approach — more often, the NDP has more votes to get back from the Liberals in their lost seats, even in ones where winning means beating a Conservative.
It’s not cut-and-dry, however, so this argues for a multi-pronged approach from the New Democrats. As has been argued throughout the leadership contest, the NDP’s path back goes through both the Liberals and the Conservatives. But, it’s clear that the two parties don’t loom equally as large in the New Democrats’ comeback plans — for the NDP to return to relevancy, they need to take the Liberals down a peg most of all.
Now, to what is in this week’s instalment of the Weekly Writ:
News on a change to the ballot in the Terrebonne byelection and a change of political allegiance in Quebec’s National Assembly.
An update on the federal polls, including some deeper looks at Manitoba and Atlantic Canada, and some insights on the challenges the NDP and its leadership contenders have ahead of them. Plus, provincial numbers out of Manitoba and Ontario.
#EveryElectionProject: The 1930 Alberta Liberal leadership race.