Spain 2026 General Election: electoral projection model

Congress of Deputies · 29 November 2026 · polling through 25 Oct · 2023 results by province · national vote inputs drive a swing in every province, then D’Hondt
350 seats · 52 constituencies · 3% threshold in each
National vote (% of valid votes, blank ballots included)
Type a share or use the arrows. Each change is swung into every province from its 2023 result, by the method picked below. In each province, lists on 3% or more of the valid vote share its seats by D’Hondt; Ceuta and Melilla elect one deputy each, first past the post. Whatever the boxes leave over is Other, which wins no seats.
Swing:
Polls used and weights
Half-life 14 days: how close the election is and how often Spain is polled. Each firm’s latest poll counts; its weight halves for every half-life it is older than the newest poll.
How the average is made. The latest poll from each polling firm (by its fieldwork end date), weighted down by age: weights halve every 14 days before the newest poll, by the site-wide rule, the longer of 1.3 × √(days left until the election), at least 7, and 1.5 × the typical gap between the latest six polls, capped at 60 days. Not every firm reports every list (several leave SALF or the smaller nationalists in “others”), so each list is averaged over the polls that do report it. A poll whose weight comes out under 1% is dropped and the rest renormalised, so an old poll can’t carry a list no current poll reports. The polls’ lists already match the 2026 ballot: PP includes UPN, Broad Front includes the Valencian Front (Compromís, with Podemos there) and Canarian People’s Unity, ERC includes Podem in Catalonia. Existe and Soria ¡Ya! aren’t polled and keep their 2023 vote. Parties’ own trackings (PSOE’s, 25 Oct) are not used: they publish three figures only. The CIS is left out of the average, as Faktor Plus is on our Serbia model and Channel 14 on Israel: since 2018, under José Félix Tezanos, a former member of the PSOE’s executive, the state pollster’s vote estimates have run well ahead of other firms for the PSOE. Its polls are still listed, and the “include CIS” scenario counts them. The average is applied to 2 decimals and adds up to exactly 100 with Other.
Labels:
One hex = one seat. Thick white lines are autonomous communities, thin ones provinces. In each province the seats run left → right in the same order every time.last seat filled in that province
* A province’s 2023 seats are notional: the 2023 votes run through this model on the 2026 lists, since Sumar’s seats can’t be split between the Broad Front and Podemos province by province. Split that way, Sumar’s vote wins 7 fewer seats (Podemos clears 3% nowhere), so the notional 2023 has PP 141, PSOE 123, ERC 8, PNV 6, Broad Front 24 and Podemos 0. The national seat changes elsewhere on the page are against the seats each list’s parties actually won. The Broad Front’s and Podemos’s 2023 vote shares are starred too: notional parts of Sumar’s. “New” = not on the 2023 ballot in that province: SALF was founded in 2024, and Adelante Andalucía stood only in Cádiz.
Show:
Seats run left → right by the site-wide rule: an invisible 45° line through Spain’s average list on the compass quiz (every list with a 2023 vote counting once, whatever its size), each list placed by how far it sits from that line. The dashed line marks half-way; 176 is a majority. Shape = stance on the territorial model (triangle: independence; circle: keep or deepen self-government; square: recentralise), pattern = stance on the monarchy (lines: republican; checks: backs the Crown), as on the compass.
What happened to every list from the July 2023 election to the 2026 ballot: who governed, who backed Sánchez’s investiture, who split and who arrived. Hover the dots and the government bands for details.
Where each list sits on the economy (left → right) and on social values (progressive → conservative). Bubble size = national vote; faded = no seats. Click a bubble to add it to the coalition.
ListQuizecon. / socialExperts (CHES 2024)econ. / socialTerritoryHead of state
Where these positions come from, how the quiz compares with the expert survey, and the territory and monarchy stances: Methodology tab. Treat every position as approximate (±1 point).