Ridings ranked by forecast gap between the favourite and the runner-up. Selecting a row or a map cell updates the riding panel.
Official Elections Canada results by riding (2015, 2019, 2021 and 2025), the official transposition of the 2021 results onto the 2023 map, national polls since 2015 and their regional breakdowns since December 2019 (Atlantic, Quebec, Ontario, Manitoba and Saskatchewan, Alberta, British Columbia). Riding boundaries of the 2023 Representation Order (Elections Canada), simplified for display.
For each region, the breakdowns published since the 2025 election are pooled with weight n·exp(−age/τ), age being measured on a clock where one campaign day counts as D days: τ = 120 days and D = 16 are estimated by one-step-ahead forecasting of the region-national gap of each of the 717 broken-down polls. The targets are corrected for the mean regional bias of the eve polls of 2021 and 2025, multiplied party by party by a shrinkage factor m²/(m² + s²/k), where m and s² are the mean and variance of the national gap between eve polls and results over the last k = 3 elections.
Within each region, the 2025 results are carried to the target by transfer: a declining party loses the same proportion of its vote everywhere, modulated by (s/s̄)^(−β) with β = 0.5 (stronger retention where it is strong), and the released votes go to the gaining parties in proportion to their gains. Evaluated at known regional shares on the 2015→2019, 2019→2021 and 2021→2025 transitions, this transfer reduces the absolute seat error from 106 to 90 compared with a uniform proportional swing.
Each draw combines a national shock and a regional shock in centred log-ratios (per-party standard deviations estimated on the errors of eve polls) and, for each riding, a local deviation drawn from the errors of the three transitions, centred by region. On the eve of the 2021 and 2025 elections, each evaluated with parameters estimated without it, 11 of the 12 per-party 90% seat intervals contain the result; the absolute seat error of the median scenario (medoid) is 36 and 14.
Each draw is assigned to a single government scenario: majority when the first party reaches 172 seats, minority otherwise. For a minority government, the parties whose seats, added to those of the first party, reach a majority are listed, excluding the Official Opposition (second party in seats): these parties hold the balance of power. In the certainty chart, each riding is assigned to its favourite party and classed by that party's probability of winning: safe (95% or more), likely (75 to 95%), lean (60 to 75%), toss-up (under 60%).