About Northern Vibe

Northern Vibe is an independent electoral platform for Canadian federal politics built on three pillars: a riding-level simulation that projects seats from current polling, an economic forecast that projects party support months into the future, and a vibe tracker that reads daily Canadian news to measure the political mood in real time.

The goal is simple: make it easy for anyone — not just political scientists — to explore how changes in public opinion translate into actual seats in the House of Commons. Instead of reading a static poll number, you can drag a slider and watch the map shift.

Every riding starts with its historical voting pattern — how people voted in 2021 and 2025 — adjusted for who actually lives there. A riding full of renters in their 30s behaves differently than a rural riding with a large retiree population.

Polling data provides the national and regional mood. If the Liberals drop three points nationally, the model distributes that drop across ridings in a way that reflects each riding's sensitivity to that kind of shift.

To turn all of that into seat counts, the model runs thousands of simulated elections. Each simulation adds a small amount of randomness — because real elections are never perfectly predictable — and then counts up who wins each riding. After thousands of runs, you get a probability distribution: not just “LPC wins 170 seats” but “LPC wins between 155 and 185 seats 80% of the time.”

The Scenario Builder lets you apply your own assumptions on top of all of this. Shift the NDP up five points. Collapse Liberal support in Ontario. The simulation reruns immediately and the map updates.

On top of the simulation, a trajectory forecast projects where the parties are likely to stand 3, 6, and 12 months from now. The forecast is powered by a Bayesian model trained on economic time series — inflation, unemployment, housing starts — and blended with current polling to produce forward-looking seat estimates with uncertainty bands.

The Vibe layer adds real-time signal. Every day, the platform fetches thousands of Canadian political news headlines from MediaCloud, runs each article through a political relevance classifier, and scores the framing for each party. Those scores are aggregated into a 28-day exponentially weighted attribution modifier that nudges party support up or down in both the simulation and the forecast.