AFOS-Analytics1/canada-2025-electoral-divergence

AFOS · Canada 2025 Electoral Divergence Dataset

🌐 English · Português · Español

Open dataset cross-referencing opinion polls × prediction markets for Canada's 2025 federal election (House of Commons, 28 April 2025), in the same spirit as the AFOS Brazil 2026 dataset: sources reported side by side with explicit divergence, not blended into one average.

Maintained by AFOS Analytics. No personal data, only public electoral information. Party-level: polls measure party vote share; the market prices the probability of winning the most seats (FPTP plurality), two different quantities, and the gap is the signal.

---

Press coverage layer

The qualitative third axis of the AFOS cross (market x polls x press) is now a structured file: news/canada-2025-press-coverage.csv, 7 dated headlines from 6 national outlets across the cycle (polls, election, result, analysis), in EN/FR. Headlines and links only (outlets retain copyright); dates are publication/coverage dates, best-effort. It complements the quantitative market-vs-poll divergence; it is not sentiment-scored.

---

English

Contents (start with the polls):

Path	Rows	Content
polls/canada-polls.csv	2,578	Party vote-share polling, long format (one row per party × poll), 6 parties, 437 polls, 2024→Apr 2025.
polls/canada-polls.json	n/a	Full structured polls (pollster, fieldwork, sample, per-party results).
data/canada-market-odds-timeseries.csv	684	Daily Polymarket "wins the most seats" probability per party (6 parties, Jan→Apr 2025) from the "Which Party wins most seats in Canadian election?" market.
data/canada-divergence-timeseries.csv	1,190	Market × poll divergence per party, each poll's party vote share joined to that party's market odds on its date.
data/canada-poly-raw.json	n/a	Raw Polymarket payload, kept for provenance.

_{Market fetched from Polymarket's gamma-api + clob via a US-resolving function. FPTP parliamentary system: one vote, no runoff; seats ≠ vote share.}

⚖️ Notable divergences (why divergence beats the average)

The market prices which party wins the most seats; the polls measure vote share. In a first-past-the-post system these come apart, and in 2025 they moved in one of the most dramatic reversals a prediction market has tracked.

• The great reversal (Conservative → Liberal). In late January 2025 the market gave the Conservatives about 85% to win the most seats; by late April it had flipped to the Liberals about 80%, while the two parties' vote shares stayed within a few points of each other the whole time. The trigger (Trudeau's resignation, Mark Carney's rise, and the tariff shock) repriced the winner long before vote-share polls told a clean story. The Liberals won the most seats (169 to 144), validating the market's final read.
• Seats vs votes. On election day the Liberals took about 43.8% of the vote to the Conservatives' about 41.3%, a roughly 2.5 pp gap, yet won a clear seat plurality. The market priced the plurality, not the vote margin; the divergence series shows that distinction day by day.
• The smaller parties (NDP, Bloc, Green): held meaningful vote share (NDP about 6-8%, Bloc about 6%) but about 0% probability of winning the most seats throughout, vote share and "winning" were never the same question for them.

The reading: vote share tells you how Canadians voted; the market tells you who was going to win the House, and in 2025 the two diverged violently before converging on the Liberals. A blended poll average would have shown a near-tie; only the market-versus-poll spread captures the swing.

Pollsters covered: Nanos, Léger, Abacus Data, Ekos, Ipsos, Angus Reid, Mainstreet, Research Co., and others.

Provenance & method: poll figures compiled deterministically (rowspan/colspan-aware HTML parser) from the public Wikipedia aggregation "Opinion polling for the 2025 Canadian federal election." Market odds from the public Polymarket market. Nothing imputed or smoothed; missing values left blank.

License (dual): data → CC BY 4.0 (LICENSE-CC-BY-4.0); code/scripts → Apache 2.0 (LICENSE-APACHE-2.0), matching the repo root and the Hugging Face mirror. Underlying poll numbers are facts released by the named pollsters; the Wikipedia aggregation is CC BY-SA. Please attribute AFOS Analytics and the original pollsters.

Cite: AFOS Analytics. Canada 2025 Electoral Divergence Dataset. Hugging Face, 2026. CC BY 4.0. (see CITATION.cff)

Disclaimer: observational research. Not investment advice, not voting guidance.

---

Português

Dataset aberto cruzando pesquisas × mercados de previsão para a eleição federal canadense de 2025 (Câmara dos Comuns, 28/abr/2025). Nível partido: as pesquisas medem voto por partido; o mercado precifica a probabilidade de vencer mais cadeiras (pluralidade FPTP).

• polls/canada-polls.csv, voto por partido, formato largo, 6 partidos, 437 pesquisas (2024→abr 2025).
• data/canada-market-odds-timeseries.csv / canada-divergence-timeseries.csv, probabilidade Polymarket de "vencer mais cadeiras" por partido e divergência mercado × pesquisa.

⚖️ Divergências em destaque

• A grande virada (Conservadores → Liberais). Fim de janeiro/2025: o mercado dava cerca de 85% aos Conservadores de vencer mais cadeiras; fim de abril virou cerca de 80% aos Liberais, enquanto o voto dos dois ficou a poucos pontos um do outro o tempo todo. O gatilho (renúncia de Trudeau, ascensão de Mark Carney, choque tarifário) reprecificou o vencedor muito antes de o voto contar uma história limpa. Os Liberais venceram mais cadeiras (169 a 144), validando a leitura final do mercado.
• Cadeiras × votos. No dia, Liberais cerca de 43,8% do voto contra cerca de 41,3% dos Conservadores (cerca de 2,5pp), mas pluralidade de cadeiras clara. O mercado precificou a pluralidade, não a margem de voto.
• Partidos menores (NDP, Bloc, Verdes): voto relevante (NDP cerca de 6-8%, Bloc cerca de 6%) mas cerca de 0% de chance de vencer mais cadeiras, voto e "vencer" nunca foram a mesma pergunta.

A leitura: o voto diz como os canadenses votaram; o mercado diz quem ia vencer a Câmara, e em 2025 os dois divergiram violentamente antes de convergir nos Liberais. Uma média esconde a virada.

---

Español

Dataset abierto que cruza encuestas × mercados de predicción para la elección federal canadiense de 2025 (Cámara de los Comunes, 28 abr 2025). Nivel partido: las encuestas miden voto por partido; el mercado valora la probabilidad de ganar más escaños (pluralidad FPTP).

⚖️ Divergencias destacadas

• El gran vuelco (Conservadores → Liberales). Fines de enero: el mercado daba cerca de 85% a los Conservadores de ganar más escaños; fines de abril viró a cerca de 80% a los Liberales, mientras el voto de ambos quedó a pocos puntos todo el tiempo. Los Liberales ganaron más escaños (169 a 144).
• Escaños vs votos. Liberales cerca de 43,8% del voto contra cerca de 41,3% (cerca de 2,5pp) pero pluralidad clara de escaños. El mercado valoró la pluralidad, no el margen de voto.
• Partidos menores: voto relevante pero cerca de 0% de ganar más escaños.

Fuente: agregación pública de Wikipedia; Polymarket. Licencia: CC BY 4.0 (atribuir a AFOS Analytics y a las encuestadoras). Investigación observacional; no es asesoría de inversión.

---

Sources / Fontes / Fuentes: Pollsters (Nanos, Léger, Abacus, Ekos, …) · Wikipedia aggregation · Polymarket. Column definitions in DATA_DICTIONARY.md.

Structural context (World Bank)

Beyond the divergence data, this dataset ships data/canada-structural-context.csv: official, open World Bank indicators that frame the country, governance (Worldwide Governance Indicators, 0-100 scale) plus economy & education (World Development Indicators: population, GDP, GDP per capita, inflation, public education spending, expected years of schooling). These are annual structural indicators that contextualize the country; they do not predict the electoral outcome. Columns are documented in DATA_DICTIONARY.md.