The two systems

  • Turtle Trading (TURTLE-01): long when price breaks a 20-day high, out on a 10-day low — the faster original rule.
  • Donchian Breakout (TREND-02): long on a 55-day high, out on a 20-day low, ATR position sizing — the slower variant.

Unlike the original Turtle rules, both versions here trade long only and do not add to winning positions. Both run on daily data for US stocks, gold, silver, oil, natural gas, agriculture, broad commodities, EUR/USD, bitcoin and ether, with walk-forward optimization and no trading costs.

Returns: hard to beat holding

Over the last five years each system beat buy-and-hold on annual return on four of ten markets: EUR/USD, natural gas, bitcoin and ether. Bitcoin stands out — Turtle 20.3% a year and Donchian 12.7% vs 7.2% for holding — but on US stocks (8.5% and 4.5% vs 13.2%), gold, silver and oil, staying invested won.

Risk: where breakouts earn their keep

Drawdowns tell the other half. The 55/20 Donchian system had a smaller maximum drawdown than holding on nine of ten markets — for example −17.8% vs −24.5% on SPY, −38.0% vs −76.6% on bitcoin. The faster Turtle rule did so on seven, and on SPY its drawdown was only −8.5% with a Sharpe ratio of 0.92, above buy-and-hold's 0.81.

2022: the commodity trend

When stocks fell 24.5% between January and October 2022, the breakout systems found the trend elsewhere: Donchian gained 37.3% on oil and 28.9% on broad commodities, and Turtle gained 64.5% on natural gas and 13.1% on broad commodities. On bitcoin, both lost 26–30% in the same period — breakouts follow trends, they do not avoid every fall.

Takeaways

  • On a single market, breakout trend following rarely beats buy-and-hold on return; its job is cutting the drawdown.
  • Its strength is diversification across markets: in 2022 the commodity breakouts paid while equities fell.
  • Slower rules (55/20) were steadier than faster ones (20/10) on most markets.
  • Choppy, range-bound markets (EUR/USD, agriculture) are where breakouts struggle.