Preventing Stop-Lossed Stocks from Being Reopened by a Model

Summary

The document addresses a conflict between stop-loss logic and a machine-learning strategy's next-day buy list. A stock may be sold when it reaches a trailing stop, then selected again by the model on the following trading day. The proposed fix is to track which stocks were stopped out and exclude them from the buy candidates during the entry stage. This makes the stop-loss and purchase-selection steps work together instead of allowing the buy logic to immediately undo an exit.

The advice describes a practical control-flow safeguard, not a change to the model's predictions. It does not specify how long a stopped-out stock should remain excluded, when eligibility should resume, or how to handle a fresh signal after a substantial price change. The post includes no code, backtest, or performance evidence. A strategy implementation would need to define the exclusion period and ensure that stop records and purchase filters use consistent timing and data, then evaluate how the rule affects turnover, risk, and returns.

Key ideas

• A model can select a stock again after a stop-loss sells it.
• The proposed safeguard is to filter recently stopped-out stocks from the buy list.
• Stop handling and model-based purchases are distinct steps that need coordinated rules.
• The document does not specify how long the exclusion should last.
• No implementation details or performance results are provided.

Tags

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