The political forecasting landscape has a gap, and we're here to fill it.
Qualitative raters like Cook, Sabato, and Inside Elections produce expert judgments but refuse to attach probabilities to their ratings. Quantitative outlets like the late FiveThirtyEight and Split Ticket build models but don't systematically hold other forecasters accountable. Prediction markets offer real-time pricing but suffer from illiquidity and a lack of independent quality assessment.
The Heat Sheet sits at the intersection of all three.
What We're Building
Race Ratings — Every competitive House, Senate, and gubernatorial race rated on the standard scale, but with a critical difference: every rating comes with an explicit estimated margin range and implied win probability.
The Spread — When prediction markets, expert ratings, and fundamentals disagree on the same race, that's where the interesting analysis lives. We compare all three and tell you who we think is right.
Decision Desk Scorecards — Nobody systematically tracks how well AP, DDHQ, Fox, CNN, and NBC perform on election night. We will.
Prediction Market Health Grades — Not all prediction markets are created equal. A 62% on Kalshi doesn't mean the same thing as a 62% on Polymarket if one market has $500 in open interest and a 15-cent bid-ask spread.
Our Promise
Every claim we make will be backed by data. Every projection we publish will be scored after the fact. We grade ourselves with the same rigor we apply to everyone else.
Welcome to The Heat Sheet. We're just getting started.