Donchian Channel

Start with the classic 20-period for entries and 10-period for exits. This is the original Turtle Trading setup and works well across most markets. Only adjust after gaining experience.

No. Many breakouts fail, especially in ranging markets. Add filters like ADX > 25, volume confirmation, or higher timeframe alignment to improve your win rate.

Common methods: 1) At the opposite band (lower for longs), 2) At 2× ATR from entry, or 3) Use the 10-period exit channel. The ATR method is often preferred as it's tighter but still gives room.

Donchian is a trend-following system. Many breakouts fail (whipsaws), but winning trades are typically much larger than losers. One big trend win (3-10R) covers many small losses (1R each).

Donchian works best on liquid, trending stocks. Avoid illiquid penny stocks or stocks that tend to range. In Singapore, it works well on DBS, OCBC, UOB, STI ETF, and other blue chips.

The Turtle filter skips a breakout if the previous 20-period breakout was profitable. It prevents chasing extended trends. However, modern traders often take all signals but reduce position size after winners.

Only take Donchian breakouts when ADX > 25, indicating a trending market. When ADX < 20, the market is ranging and breakouts are more likely to fail. This filter significantly reduces whipsaws.

In Singapore, shorting individual stocks can be challenging. Many traders focus on long signals (upper band breakouts) or use inverse ETFs/options for bearish plays. Use lower band for exit signals.

Turtle System 1 uses 20-period entry/10-period exit with a filter. System 2 uses 55-period entry/20-period exit with no filter. System 2 catches longer trends but gives back more profit before exit.

Whipsaws are inevitable. Use filters (ADX, volume), proper position sizing to limit damage, and accept that low win rate is normal. The key is keeping losses small (1R) while letting winners run (3-10R).

Turtles added units at every 0.5× ATR in profit. Max 4 units per market. After 2nd unit, stop moved to breakeven for first unit. This aggressive pyramiding amplified winning trades significantly.

Turtles limited to 4 units per market, 6 units in closely correlated markets, 10 units in loosely correlated, and 12 units in total direction (long/short). This prevented over-concentration and managed correlation risk.

Use standard parameters (20/10), test across multiple instruments and time periods, use walk-forward analysis, and avoid optimizing parameters excessively. Expect live performance to be 20-30% worse than backtests.

REITs have lower volatility. Use longer periods (30/15) and lower ADX threshold (20). Expect smaller but more consistent trends. Position sizing should account for lower volatility and dividend impact.

The edge comes from asymmetric payoffs: small fixed losses (stopped at ATR) vs large variable gains (riding trends). Win rate ~40%, but winners average 3-5× losers. Expected value = (0.4 × 3R) - (0.6 × 1R) = +0.6R per trade.