Trading Chande Oscillator Extremes with Rolling Percentile Thresholds

Summary

This strategy calculates the Chande Momentum Oscillator (CMO) from recent price changes and compares it with rolling percentile thresholds. Using a stated lookback of 425 periods, the document describes entering long when CMO reaches its low extreme percentile and short at its high extreme. It also describes exits at less extreme percentile levels, intended to avoid repeatedly entering while the oscillator remains extreme.

The prose says the approach targets Bitcoin and other liquid crypto pairs on short timeframes, though the published backtest settings show BTC/USDT futures on a one-hour period with a 15-minute base period. No performance results are provided, and the code comments out the described exit rules; it also contains no stop-loss or take-profit logic. The document warns that noisy momentum readings can create false signals, extreme thresholds may mean long waits between trades, and parameter optimization can overfit. It proposes adding risk controls, trend filters, and multi-timeframe checks for further study.

Key ideas

• The strategy compares the CMO with rolling percentile levels to identify unusually low or high readings.
• The documented entry thresholds are the 1st and 99th percentiles over a 425-period lookback.
• The prose describes exits at less extreme percentile levels, but those exit rules are disabled in the supplied code.
• The strategy has no implemented stop-loss or take-profit, creating unbounded trade risk under adverse moves.
• Short-timeframe CMO signals may be noisy, and tuned thresholds can overfit historical data.

Tags

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