Skip to content
All library documents

Heuristic Futures Position Selection for Small Accounts

Article Systematic trading blog (Rob Carver)

Summary

This article examines how a retail-sized account can trade a broad futures universe when positions must be whole contracts. A diversified portfolio performs well in a fractional-position backtest, but integer rounding prevents the smaller account from holding many markets. The resulting portfolio becomes concentrated in instruments with affordable contract sizes and tends to use less risk than intended. The author also considers choosing a fixed subset of markets, while noting that this gives up some diversification and can involve look-ahead in how the subset is selected.

The proposed alternative is a daily heuristic: rank markets by absolute forecast strength, retain existing positions where appropriate, and allocate scarce position capacity toward markets with stronger signals, subject to tradeability and forecast thresholds. This turns constrained allocation into a simple selection process that can be explained without a complex daily optimizer. The excerpt includes partial implementation and signal filters, but cuts off before the full algorithm, detailed results, and conclusion. Its reported tests use a particular futures universe and strategy setup, so the approach’s performance and suitability may depend on contract costs, available capital, and portfolio design.

Key ideas

  • Integer contract sizes can prevent small accounts from holding a well-diversified futures portfolio.
  • Rounding can concentrate exposure in markets with smaller contract sizes and leave portfolio risk below target.
  • Selecting a fixed subset of markets may help with tradability but sacrifices some diversification.
  • A daily heuristic can rank tradeable markets by absolute forecast strength to guide position selection.
  • The excerpt does not provide the full algorithm or enough results to assess its general performance.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.