Common Pitfalls When Specifying and Automating Trading Systems

Summary

This article offers practical advice for customers commissioning trading software, with lessons about defining and testing strategies before automation. It warns that vague requirements, untested indicator combinations, and systems based on visual impressions can create mismatches between what a trader expects and what a developer can implement. In particular, indicators with different scales can be misleading when plotted together, while some indicators revise historical values as new bars arrive. The suggested check is to observe signals dynamically, including in a visual tester, rather than relying only on a historical chart.

For traders whose decisions depend on intuition, the article recommends dividing automation into manageable stages or starting with a semi-automated tool that handles routine operations while the trader chooses entries. It also argues for limiting an Expert Advisor to functions that support trading, since extra monitoring and notification features add cost and possible failure points. These are experience-based recommendations rather than measured results. The article is a broad service guide, and its examples do not establish that any particular indicator or automation approach is profitable.

Key ideas

• Write precise requirements and manually assess a trading idea over an extended period before commissioning automation.
• Check indicator behavior as new bars form because some indicators can revise historical signals.
• Use visual testing to detect signal movement, disappearance, or effects caused by chart scaling.
• Break a discretionary process into stages or begin with semi-automation when full rules are difficult to formalize.
• Keep an Expert Advisor focused on necessary trading tasks to reduce complexity and potential failure points.

Tags

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