Practical Risks in Spot–Delivery Futures Spread Arbitrage
Summary
The article describes building a spot and delivery-futures spread strategy around the idea that the spread tends to fluctuate near a mean. Its initial signal uses a spread deviation from that mean, while an augmented Dickey-Fuller test screens for stationarity; repeated failed checks disable trading temporarily. The development notes also cover trade accounting from actual fill prices, managing residual positions, avoiding one-sided exposure when only one leg fills, and rechecking the signal just before entry.
A central implementation lesson is to distinguish last-trade ticker data from order-book depth. The author uses ticker observations to maintain a historical series for statistical analysis, then uses depth-derived midpoint prices to verify current opportunities, set orders, and estimate open P&L. The examples describe fallback orders, retries, and cooldowns as responses to thin liquidity and exchange failures. This is an account of an unfinished strategy, not validated evidence of profitability; the author explicitly warns that the code is immature and excludes important improvements such as stronger risk controls, slippage modeling, and lower-latency data.
Key ideas
- A spot–delivery spread signal can be screened for stationarity before mean-reversion trades are considered.
- Ticker prices may lag in thinly traded contracts, while order-book depth better reflects current executable prices.
- Using actual average fill prices improves trade-level P&L accounting across both legs.
- Rollback procedures, retries, and position checks address execution failures and leftover exposure.
- The strategy remains experimental and lacks demonstrated live profitability and complete risk controls.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.