polymarket-weather-edge
Automated paper-trading bot betting the NO side of overpriced Polymarket temperature markets using METAR data and Gaussian modeling
This strategy identifies overpriced NO sides on Polymarket temperature markets by comparing market odds with a Gaussian probability model built from live METAR station observations and Open-Meteo forecasts. The bot automatically places paper trades when the implied probability of a temperature outcome is significantly higher than the model's estimate. Entry conditions require a minimum edge threshold between market price and modeled probability. Exit is at market close or when the edge dissipates. Implemented in Python with a full METAR ingestion pipeline and the CLOB API. Core risk is weather forecast error and model miscalibration; slippage is minimal in prediction markets.