Research
How the forecast works, in full
The 11 papers behind our Canada forecast (22,706 words): how far the polls have missed, how we correct them, the seat model and the tests.
These papers set out every part of our method: what we measure, how we correct the polls, how votes become seats, how uncertain the result is, and what we have tested and rejected. Figures marked Live are computed from the database each time the page is built, so they always match the forecast. En français
Also: France · Germany · Spain · Italy · Netherlands · Poland · Australia · United States · Great Britain
11 papers · about 23k words · Our method, in short · How the elections work · Data sources
Canada
- Forecasting the House of Commons from imperfect polls: an overview The Canada section forecasts the next federal election with the rule used for every country on the site: take half the average past miss off today's polls, but only where that would have improved past forecasts made without knowing the result. Canadian final polls pass that test: since 2008 they have understated the Conservatives every time, and since 2000 overstated the NDP every time. The seats come from 343 separate first-past-the-post contests, so the forecast reads seven regions and moves every riding's 2025 result with its region. This paper maps the other ten Canada papers.
- How far Canadian polls have missed, 1988–2025 92 final polls from 27 pollsters at twelve general elections, 82 of them at the nine from 2000 that the model uses, against the official result. The average missed each modelled party by 1.21 points and each main party by 1.32, and the misses have a direction: the Conservatives were understated at all six elections from 2008, the NDP overstated at all nine from 2000, the Greens and the People's Party every time. We define the error, work through 2025, split shared from pollster-specific error, and give the gap between the leaders and every pollster's record.
- How we read Canadian polls: the average, the correction and its test The core Canada methods paper. The average takes each pollster's latest national poll within 21 days of the newest, one per pollster. The correction takes off half of each party's average past miss, and is kept only because it passes a leave-one-out test: across the nine general elections from 2000 to 2025 it cut the average miss per party from 1.21 to 0.94 points and was closer at seven of nine. Today it adds about half a point to the Liberals and Conservatives and takes three quarters of a point from the NDP.
- From national polls to 343 ridings: regions, Quebec and the swing model How the Canada forecast turns national shares into riding shares. Seven regions are each read two ways, by national swing from 2025 and from the pollsters' own regional figures measured against the same release's national figure, blended by how many pollsters report the region and scaled to add back up to the nation. Each riding's 2025 result then moves by its region's change by uniform swing. The paper covers the 151 regional readings, Quebec and the Bloc, a worked riding, seats by province, and the Alberta and Manitoba provincial forecasts.
- Testing the seat engine: the plurality check, 2025 and the swing replays Three tests of the Canada seat engine. The plurality rule on the official riding votes gives the declared seats at all twelve general elections from 1988 to 2025. Fed the real 2025 regional votes, the engine gives back the 2025 House exactly. And replaying seven pairs of elections from 2004→2006 to 2021→2025 with the true regional change, uniform swing got 268 of 2,281 ridings wrong, against 278 mixed and 288 proportional, so it is used. The same replays size the regional and riding noise in the simulations.
- Simulating Canadian elections: where the chances come from The Canada seat ranges, majority chances, official-party-status chances and riding chances come from 2,000 simulated elections with error at three levels: a national error per party the size of its past final-poll miss, scaled to its size now; a regional error fitted on past departures from national swing, smaller where regional polls exist; and riding noise fitted so past replays would have missed as many ridings as they did. National errors are shared by every riding and regional errors by every riding in a region, so ridings fall together.
- Polling day: how far Canadian polls move before an election Polls at ten distances from a month to three years before the nine general elections from 2000 to 2025, compared with the result and the previous result. Pulling each party part of the way back towards its previous result beat the polls as they stood on average, narrowly, so the polling-day forecast does it: three years out, today's polls count for 43% of the gap to 2025. The typical miss grows from about 3.9 points per party a month out to 6.6 at the fixed date, 15 October 2029, and the ranges widen to match.
- If we're wrong: today's polls with each past miss replayed Beside the projection, the Canada pages show the House if today's polls miss exactly as they did at each general election from 2000 to 2025: each election's final-poll errors are taken off today's average and the seat engine run as usual. Every lens keeps a Liberal majority today, from 228 seats (a repeat of 2011) to 266 (2006). The post-merger misses of 2004 and 2006 favour the Liberals; those since 2008 favour the Conservatives. 48 ridings change hands under at least one lens.
- Rating Canadian pollsters: accuracy, the lead and openness Every pollster that has polled Canada is rated out of 100: 50 points for its final polls since 2000 and 20 for its miss on the gap between the leaders, both ranked against Canada's own 82 final polls and shrunk towards a typical one, and 30 for openness, including CRIC accreditation. Léger and Mainstreet score highest (A). The ratings do not weight the average. Seven pollsters' releases, including Léger's regional crosstabs and Nanos's weekly PDF, are now read automatically; the rest are entered by hand.
- The Canada data: sources, grades and checks Where every figure in the Canada section comes from: Elections Canada's results for 2011 to 2025 and its transpositions onto new ridings, the Library of Parliament's record for earlier elections, Statistics Canada's census and boundaries, the House of Commons member lists, and 481 polls, 302 from the pollsters' own releases and 179 from Wikipedia's tables, including all 92 final polls. The paper sets out the checks each file passes, the law behind the forecast, the polling-day switch, the provincial data and the licences.
- What we don't know yet: limitations of the Canada forecast The Canada forecast rests on nine elections whose final polls are all second-hand, and its Conservative correction on a pattern six elections long. Its swing model has no candidates and moves every riding in a region alike; its regional readings rest on one to three pollsters; the Bloc's whole vote is in Quebec; its simulation draws party errors independently; and the fixed date is three years off and may move. For each limitation we say how much it matters and what evidence would change the method.
