Assessing Supertrend Parameter Robustness with a Sensitivity Grid
Summary
The document describes an indicator that evaluates a grid of Supertrend settings across different ATR lengths and multipliers. Each configuration is treated as an always-in-market system: trend changes close the current position and open one in the opposite direction. Users can examine several performance measures, such as net profit, win rate, and profit factor, as a three-dimensional surface and a matrix. The display distinguishes the single best-performing setting from a neighborhood selected for a high average result and relatively low variation.
The central lesson is to inspect the shape of results rather than trust an isolated optimum. A broad, smooth region suggests nearby settings perform similarly, while a sharp peak signals parameter sensitivity. The indicator can also apply the selected stable-area setting to the plotted Supertrend. Its trade record is explicitly a simplified in-sample calculation without transaction costs or slippage, and it assumes continuous market exposure. It is a diagnostic for comparing settings, not a complete backtest or evidence of future performance.
Key ideas
- The indicator compares Supertrend results across a grid of ATR lengths and multipliers.
- It plots selectable performance measures to reveal how results vary across settings.
- A local stability score rewards a strong neighborhood and penalizes variation among nearby configurations.
- A broad performance plateau is generally a more robust candidate than an isolated best result.
- The simplified in-sample trade record omits costs and slippage and assumes continuous exposure.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.