Who really ate the NDP’s lunch in 2025?

        

by Maurice Y. Michaud (he/him)

On the lunch menu in 2025...

• Strawberries (Liberal Party of Canada)
• Saskatoon berries (Conservative Party of Canada)
• Blueberries (Bloc Québécois)
• Oranges (New Democratic Party of Canada)
• Grapes (Green Party of Canada)

If one were to ask any Canadian which party was responsible for the NDP’s demise in the 2025 federal election, most would not hesitate to blame the Liberals. However, that answer, which is based primarily on past experience, overlooks an important fact: the Conservatives actually won more seats from the NDP than the Liberals did. Also, the 2025 election was unusual in the sense that it was the first since the Diefenbaker sweep of 1958 that two parties shared 84% of all the votes between themselves, leaving only crumbs for the other parties. So, the question of who ate the NDP’s lunch in 2025 deserves a much deeper analysis.

To begin, let’s not be coy about it. Following the NDP’s spectacular rise to official opposition status in 2011, it’s been downhill for them ever since. As for the Bloc Québécois during that same time period, it only began recovering in 2019 from its 2011 near-elimination by the NDP, and never returned to its pre-2011 stature, when it held the most seats in Québec. For its part, the Green Party, while it had finally entered Parliament in 2011, remained a marginal fourth opposition group with no official status.

By Christmas 2024, with Parliament paralyzed and only months before a general election was to be called, polls indicated that the Conservatives were heading towards forming a massive majority government. Some of those polls also pointed to the possibility of the Bloc becoming the official oppo­si­tion for a second time, and the NDP winning a few more seats than the incumbent Liberals. Such was the fatigue of the electorate with Justin Trudeau. But the re-election of Mister Tariff Donald Trump to the presidency of the United States, the sudden resignation of Chrystia Freeland on the day she was to submit her economic update to Parliament, which in turn forced the resignation of Trudeau in the early days of 2025, and the arrival of Mark Carney as Liberal leader and prime minister, created the perfect storm. The Liberals were able to pull the carpet from under the feet of the Conservatives, and the smaller opposition parties got squeezed out in the most binary general election in Canadian history since 1958.

The situation of the small parties

The NDP’s situation
In 2015, with many Canadians fed up with a decade of those new Conservatives in power, they first took a look at the party claiming to be the government in waiting. Some polls at the start of the election campaign even suggested the country might be on the verge of electing its first NDP government, albeit an extremely weak minority one. But anyone who ever doubted the impact of an election campaign no longer have any doubts after 2015.

Indeed, NDP leader Thomas Mulcair was outflanked on his left by Liberal leader Justin Trudeau. Convinced that his party should not be perceived as wasteful, Mulcair repositioned it toward the centre on economic issues, while Trudeau asserted that he wasn’t afraid of a budget deficit if it meant giving Canadians a boost. But above all, in its categorical desire to oust the Conservatives from power, the electorate turned to a familiar entity — the Liberal Party — rather than one that had never held the reins of power federally. From 103 seats following the 2011 general election, the NDP found itself as the second opposition group with only 44 seats, which led to Mulcair being shown the door as leader.

The party chose Jagmeet Singh as its new leader, the first visible minority person to lead a major federal party, as well as the first of the Sikh faith. The four core focuses of his leaderhip campaign were inequality, climate change, reconciliation with indigenous peoples, and electoral reform. He also supported decriminalizing personal possession of all narcotics and instead promoting harm reduction for users.

Whether it’s because of these positions or his knack for proposing ideas that encroached on provincial jurisdictions (or a combination of both), Singh brought his party further down than Mulcair.

• In 2019, the NDP went down to third opposition group, with 24 seats and just shy of 16% of the popular vote.
• In 2021, the party remained third opposition with one more seat and 17.7% of the popular vote, and participated from March 2022 to September 2024 in a confidence and supply agreement with the Liberal minority government.

Because of either this agreement or the way Singh ended it, combined with the existential threat posed by the re-election of Donald Trump as president of the United States, the NDP exited the 2025 general election with only 6.3% of the popular vote, 7 seats, and no official party status. Failing to win his own seat, Singh resigned on election night.

Although the 2025 general election was far from ordinary, the NDP knew the scenario all too well.

• Holding the balance of power in 1972 with 31 seats, it had entered into such an agreement with the Liberal minority government of the time. After the 1974 general election, it found itself with only 16 seats against a Liberal majority government.
• Due to disgust with the outgoing Progressive Conservative government in 1993, it saw its votes shift to the Liberals and the new Reform Party, and went from 43 to 9 seats, losing its official party status.
• Facing a Liberal minority government in 2004 and a Conservative minority government in 2006, each time it was only one or two seats shy of holding the balance of power.

The disaster for the NDP in 2025 was that, even though the electoral system in Canada is such that the prime minister is not directly elected, voters were looking for one who would be capable to stand up to Donald Trump. And neither the Conservative Pierre Poilievre nor (especially?) Singh was seen as measuring up to the former head of the Bank of Canada AND the United Kingdom: the new Liberal prime minister, Mark Carney.

The Bloc’s situation
Still reeling from its humiliation four years earlier at the hands of the NDP, the Bloc needed to win 7 seats in 2015 to regain official party status. The good news: it won back 8 from the NDP; the bad news: the Liberals took away 2 and the Conservatives one, still leaving them without party status. But after a rough patch under a controversial leader, it gained 22 seats in 2019, taking back all but one of the Québec seats still held by the NDP, as well as 8 from the Liberals and 3 from the Conservatives, and holding onto those 32 seats in 2021.

But in late fall of 2024, just as some polls suggested that the Bloc could form the official opposition to a massive Conservative majority, it was also hit by that Trump-Freeland-Trudeau-Carney perfect storm, and many worried Québécois, who like other Canadians were shopping for a prime minister capable of facing Trump, considered the Bloc too much of a gamble in 2025. While not bruised as badly as in 2011, it lost 9 seats to the Liberals and one to the Conservatives, leaving them with a caucus of 22.

The Greens’ situation
After winning their first seat in 2011 and coming into the 2019 election with a second seat it had won in a by-election a few months earlier, the Greens surprised everyone by winning their first seat outside of British Columbia — in New Brunswick, of all places! Perhaps feeling she had brought her party to a good place, Elizabeth May decided to pass on the leadership to somebody else, but that somebody else proved unpopular within the party, causing infighting and the defection of Jenica Atwin to the Liberals at the eve of the 2021 election. While May agreed to return as leader with a co-leader, the public’s perception of the party soured. With environmental issues no longer at the forefront of voters’ mind in 2025, the Greens also faced the headwind against any party unlikely to form government, losing the seat in Ontario they had won in 2021 and leaving May as the sole Green in Parliament.

The net effect of binary polarization in 2025
Soft NDP voters have long been suspected of shifting to the Liberals when they sense a significant rise of the Conservatives. It’s often said jokingly that soft NDPers so despise the Conservatives that, when they think that party stands a good chance of forming government, they pee their pants and vote Liberal. But while that definitely happened in 2025, the Conservatives were the net beneficiary in terms of seats, just like in 2011. In fact, the Conservatives devoured 30 lunches belonging to other parties compared to only 26 such steals by the Liberals!

As usual, the Liberals and Conservatives swapped some seats, but the Conservatives ate eight more strawberries than the Liberals ate Saskatoon berries. So the way the Liberals strengthen their minority to only three seats shy of a majority was by eating more BQ blueberries than NPD oranges, with the Conservatives surprisingly managing to steal 10 oranges, bringing Chantal Hébert’s to observe on election night that the Liberals owed their achievment in good part to Québec. In the end, the Liberals won 16 seats from oppositions other than the Conservatives thanks to those bloquistes who acted like the traditional soft NDPers, while the Conservatives got 12 from oppositions other than the Liberals.

So, in terms of seats, the Conservatives ate the NDP’s lunch, while the Liberals feasted on the Bloc’s.

In terms of votes, however, that’s another story. It’s easy to count the number of seats that have shifted from one party to another from one election to the next, but the same cannot be said about votes. To begin, voter turnout in the 2021 election was 62.6% and a little more than 17 million votes were cast nationwide. In 2025, voter turnout jumped to 69.5% and more than 19.6 million votes were cast.

Therefore, in order to compare apples with apples, we must look at the percentage of votes each party received riding by riding, specifically the percentage gained or lost by each party between 2021 and 2025. By identifying the party that lost the most in terms of vote percentage and the one that gained the most, we can conclude that the latter party benefitted the most of the shift, even though that does not necessarily mean that the latter party won the seat. In fact, the split may have caused another party to slip from behind and actually win the seat, as happened during the Orange Crush federal election of 2011, among others.

Let’s be clear: we are NOT assuming that votes fit into a net zero equation. For example, in 2025, the Liberals flipped Calgary Confederation from the Conservatives. In that riding, the vote share for the NDP dropped 13% between 2021 and 2025, falling from 17.1% to 4.1%. In fact, the drop for the NDP and the Greens combined was 16.1%, yet the Liberals’ increase from 2021 was 19.6% and the Conservatives’ was 0.2%, which, of course, exceeds 16.1%. That would suggest that the Liberals gained about 3.5% between elections entirely on their own merit, most likely aided by the higher voter turnout in 2025.

This caveat aside, the following table clearly shows that, in terms of votes, the Liberals nibbled the most on the NDP’s lunch, but the Conservatives’ appetite for oranges in 2025 was far from negligeable, which explains how they managed to flip more orange seats to their side than the Liberals did. Federally, this significant shift from the NDP to the Conservatives is a new phenomenon in the 2020s, and it flies in the face of the received orthodoxy that soft NDP voters always wish to block the Conservatives and flock to the Liberals to achieve this objective. It remains to be seen if this phenomenon was specific to the unusual circumstances in 2025, or if it is indicative of a further rightward drifting of the Canadian electorate that began in 1993.

So how should we interpret this table?
Let’s read down the Liberal (strawberries) column under the Advantaged heading.

• The Liberal against Liberal cell is obviously blanked out.
• There were 4 times that the Liberals’ vote percentage had increased the most while the Conservatives’ had decreased the most.
• There were 26 times that the Liberals’ vote percentage had increased the most while the BQ’s had decreased the most.
• There were 169 times that the Liberals’ vote percentage had increased the most while the NDP’s had decreased the most.
• Finally, there were 5 times that the Liberals’ vote percentage had increased the most while the Greens’ had decreased the most.

Again, to be clear: This table is NOT saying that the Liberals took 169 NPD seats, for the NDP never had 169 seats to be taken! We are only speculating that the NDP’s decrease from 2021 to 2025 likely transferred mostly to the Liberals in 169 ridings nationwide in 2025, all the while recognizing that, in some ridings, it may have gone to the second most advantaged. But it’s the value of 113 in the cell next to the Liberals’ 169 value (in the Saskatoon berries versus oranges cell) is how we can point to the new phenomenon of NDP votes drifting to the Conservatives more than before 2025. And by coincidence, it happens to be true in this case that the 1 at the intersection of the Saskatoon berries and the grapes indeed represents a seat won by those berries.

So, the fact that the Liberals and Conservatives combined ate 17 whole NDP oranges (or seats) explains how the NDP found itself with only 7 left and no official party status in the 45th Parliament. But the fact those same Liberals and Conservatives combined heartily squeezed out so much NDP orange juice (or votes) in an astounding 282 of 334 ridings explains how the NDP went from getting 17.74% of the votes nationwide in 2021 to only 6.29% in 2025 — incidentally, the same vote share as the BQ, which only ran candidates in Québec’s 78 ridings.

Stepping away for a moment from trying to figure out who exactly ate the NDP’s lunch, the next table reveals the extent of the NDP famine. The vote share differential is the subtraction of the vote share a party received in 2025 in a given riding from its vote share in the same riding in 2021. In other words, if a party had a 10% share in 2025 but a 20% share in 2021, its differential in that riding would be −10%. Thus the riding is the point of comparison, not the party. If the party were the point of comparison, then it could be argued that the party lost 50% of its votes compared to the previous election — a far more brutal assessment!

Still, even if we look at the much gentler vote share differential, we find that the NDP had a differential of −10% or worse in 184 of the 334 ridings evaluated, and one from −9.99% to −3.00% in 134 ridings. Put differently, in a little more than 95% of the ridings, the NDP took either a moderate or big step backwards. In contrast, we find the Liberals’ differentials in the moderate or significant gain ranges (+3.00% or greater) in slightly more than 85% of the ridings, while the Conservatives’ were in those same ranges in 80% of the ridings, further demonstrating that they were BOTH responsible for starving the NDP.

The following table, however, is not for the faint of heart, but it is what made the above analyses possible. Indeed, it is the detailed riding by riding analysis that compares the ranks and, more critically, calculates the vote share differentials of each of the five parties. Using a heat mapping principle, the third row for each riding presents each party’s rank change and, moreover, the intensity (either positive or negative) of its vote share differential. Additionally, the most disavantaged and most advantaged parties are shown for each riding. Clicking on the name of a riding opens a popup window showing the details of that year’s election, but clicking on Hist. opens a different popup window giving you the history of that riding, thus allowing you to gain more perspective on the significance (or lack thereof) of this analysis for that particular riding.

For example, the Liberals may well have achieved a large positive vote share differential of +11.26% in Airdrie—Cochrane in 2025, but it’s safe to assume that it did not lead the Con­ser­vatives to break into a sweat in fear of losing their seat. In fact, they simply consolated their hold onto the seat by siphoning off more votes they didn’t really need from both the NDP and the Greens who hadn’t even bothered running a candidate in 2025. There­fore, that +11.26% differential isn’t overly significant. How­ever, with an even larger positive differential of +19.61%, the Liberals flipped Calgary Confederation, which never before voted Liberal and, in fact, was the last seat occupied by former Conservative prime minister Stephen Harper. Therefore, that +19.61% differential is quite significant.

Slightly complicating matters is the fact that the electoral map changed between 2021 and 2025. There were 338 ridings in 2021 but 343 in 2025, meaning that 5 ridings from 2025 cannot be compared to 2021. What’s more, the territory or name (or both) of some ridings was changed, but they could still be held together thanks to PoliCan’s “follow the territory” master ridings scheme. But while the redistribution added 5 seats, there were 3 master ridings that came to an end after 2021, which is why only 335 ridings could be compared.

So who really ate the NDP’s lunch in 2025? The answer is that both the Liberals and Conservatives did. But the Conservatives stole more whole oranges (or seats), while the Liberals squeezed out more juice from those oranges (or votes).

---