Guides / Quant Modeling

Detecting Volatility Regime Transitions in Prediction Markets

A volatility model fit to one regime — calm, range-bound trading ahead of a known event, say — will misprice risk once the market shifts into a fast-moving regime around resolution. Detecting the transition point (rather than assuming a single stationary volatility) is usually the harder half of the modeling problem.

Where the data comes from

ProbSights' historical data gives you the BTC/ETH/SOL Up/Down price series at 5m–24h resolution across Kalshi and Polymarket to fit and validate a regime-detection model against.

Related

• Volatility modeling in prediction markets
• Time-series analysis of prediction market data

Related guides

• Anomaly Detection in Prediction Market Order Books
• Bid-Ask Bounce and Microstructure Noise in Prediction Markets
• Feature Engineering From Prediction Market Order Book Data
• Iceberg Order & Hidden Liquidity Detection in Prediction Markets

Data from Kalshi and Polymarket. Search live · API docs · Pricing