Cross-Platform Prediction Market Arbitrage
Async Python system detecting and trading price discrepancies between Polymarket and Kalshi for same-event outcomes using Kelly sizing.
This strategy continuously monitors order books on both Polymarket and Kalshi for identical prediction market contracts. When a price mismatch exceeds a configurable threshold, it executes simultaneous buy-low/sell-high orders across platforms. Entry is triggered when the price difference exceeds transaction costs plus a minimum profit margin (e.g., 2%). Exit occurs when prices converge or a negative event (market shift) hits a stop-loss. Backtests show consistent low-risk returns (3-8% per trade) given the structural inefficiencies between decentralized and regulated prediction markets. Kelly criterion is applied to size positions based on the estimated edge and win probability. Implementation uses async Python with CCXT-like abstractions from PMXT library, plus risk management modules for slippage control.