Skip to content
All library documents

How Stop-Loss Rules Change Returns and Act as Trend-Following Signals

Article QuantInsti blog

Summary

The article examines how fixed and trailing stop-loss rules affect a strategy’s return distribution. Its central point is that stopped trades remain part of the results: a stop can cut large losses while also closing positions that might have recovered or become small winners. In simulations of a positively drifting geometric Brownian motion, a fixed stop lowered the mean return from 10% to 8.8%; a trailing stop in a separate setup lowered it to 4.7%. The article also describes a simple SPY intraday test in which adding a daily volatility-based stop improved reported profit and Sharpe ratio while lowering the win rate, producing statistics associated with trend following.

The author argues that a price-based stop expresses a view that adverse price movement will continue, so it is better understood as an exit signal with trend-following logic than as stand-alone risk control. This distinction matters for mean-reversion strategies, where exiting after a decline can remove potential rebounds. The evidence is illustrative, not universal: simulated returns depend on assumed processes, and the market test uses one instrument, sample, and rule. Real returns may have changing distributions, skew, fat tails, costs, and slippage; the article says risk preferences can still justify stops.

Key ideas

  • Stops reshape the whole outcome distribution because stopped positions may otherwise have recovered or become winners.
  • In the stated simulation, a fixed stop reduced average return while removing the largest losses.
  • Trailing stops can further reduce expected returns by closing trades before larger gains develop.
  • A price-based stop assumes adverse movement predicts further movement, giving it trend-following logic.
  • Stop rules may fit momentum strategies better than mean-reversion strategies, depending on the source of the strategy’s edge.

Tags

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