Practical Spot–Futures Arbitrage: Stationarity, Execution, and Risk
Summary
The document explains how a crypto spot–futures arbitrage idea based on spread mean reversion becomes more complicated in live trading. The proposed approach enters when the spread’s Z-score moves beyond a threshold and exits as it reverts, while an augmented Dickey–Fuller test screens for stationarity. The author describes tracking spot and futures P&L using actual fill prices, checking live order-book data before entry, and closing residual futures positions before opening another trade.
Most of the discussion concerns execution risks: one leg may fill without the other, market orders can fail, spot market buys may require quote-currency sizing, and low liquidity can make rollback difficult. The strategy uses retries, fallback orders, cooldowns, and a final real-time signal check to address these issues. It also distinguishes slower ticker data for historical analysis from order-book depth for execution decisions. The author says the implementation remains immature and identifies latency, stop-loss design, slippage, and liquidity as unresolved limitations; no backtest or performance evidence is provided.
Key ideas
- Spread mean reversion is meaningful only when the spread behaves as a stationary series.
- Use actual spot and futures fill prices to calculate trade returns accurately.
- A failed leg creates directional exposure, so rollback and position checks matter.
- Recheck the spread using current order-book data before sending entry orders.
- Thin liquidity, latency, and slippage remain substantial risks for the described implementation.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.