Quadratic Curve Breakouts with Multi-Period Signal Checks

Summary

This breakout strategy fits quadratic curves to rolling bar highs and lows, treating the resulting upper and lower curves as potential resistance and support. It identifies candidate signals by comparing the fitted curves across nearby bars: a qualifying high-side pivot produces a buy signal, while a qualifying low-side pivot produces a sell signal. The described method uses a configurable fitting window and a multi-period check intended to screen out isolated pivots. The document also characterizes the approach as trading closes that break fitted bands, but the supplied source logic centers on pivot comparisons and does not clearly implement that close-based rule.

The source includes fixed-size entries and allows pyramiding, while no explicit exit rule is described. Published settings show a short BTC/USDT futures backtest on hourly bars, but no performance metrics are given. The document flags lag, parameter overfitting, lack of volume confirmation, and the need for dynamic sizing and stops. Its claims about robustness and statistical arbitrage remain proposals for further testing rather than established findings.

Key ideas

• Quadratic regressions of rolling highs and lows define candidate price boundaries.
• A multi-period pivot comparison is used to generate buy and sell signals.
• The text describes breakout confirmation, but the supplied source logic does not clearly match that account.
• The example uses fixed-size entries and pyramiding without a specified exit rule.
• The short backtest configuration includes no reported performance evidence, so robustness is unproven.

Tags

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