Using MAE Scatter Plots to Evaluate Stop-Loss Levels

Summary

The document asks whether maximum adverse excursion (MAE) can help select a stop-loss level when a scatter plot of long-trade outcomes against MAE shows winning and losing trades mixed across the range. The proposed visual inspection finds no obvious MAE region dominated by losing trades, raising doubt that a simple cutoff would improve outcomes. It supplies no underlying observations, plotted values, sample size, or quantitative analysis, so the apparent overlap cannot establish whether a stop is useful.

A useful evaluation would compare trade results under candidate stop rules with the original strategy, accounting for trades stopped out before later gains, execution costs, and the possibility that the observed pattern is sample-specific. MAE describes how far a trade moved against the position; it is not by itself proof that a particular excursion should trigger an exit. The document poses the question but does not provide an answer or demonstrate a validated threshold, so any decision would require out-of-sample testing and risk-aware analysis.

Key ideas

• MAE measures the largest adverse move experienced during a trade.
• Overlapping winning and losing trades may make a simple MAE cutoff hard to justify visually.
• The described plot alone does not establish whether a stop-loss improves strategy performance.
• Candidate stops should be assessed against the original strategy with costs and later trade outcomes considered.
• A threshold inferred from one sample may not generalize and needs out-of-sample evaluation.

Tags

# Is this an Example where Maximum Adverse Excursion (MAE) is not useful for a Stop-Loss?


# Is this an Example where Maximum Adverse Excursion (MAE) is not useful for a Stop-Loss?












Below is an attached screenshot of a scatter plot of a long position Percentage Return of a Asset Security on the Y-axis, and the Maximum Adverse Excursion (MAE) Percentage on the X-axis. Green dots are winning trades, red dots are losing trades. While looking and analyzing the scatter plot to see where it might be good to place the stop loss, it seems this might not be a good use case of using the Maximum Adverse Excurusion (MAE), because it seems there are about the same amount of winning trades or more (green dots) where there groups of loosing trades (red dots) are. Not exactly , but visually it seems almost evenly distributed or more heavy on the green dots (winning trades). In other words, there is not an area on the plot where the red dots (losing trades)out number heavily on one side than compared to the green dots (winning trades). Because of this, would trying to use the Maximum Adverse Excursion (MAE) be not useful at all to find a good Stop Loss Level in this Example ?

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

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