The thesis
Dual-momentum keeps things simple: hold the strongest trending names by 60-day return across a 24-stock universe of US large caps — the mega-cap tech, financials, staples and healthcare you would expect — and exit when the trend breaks. It is a classic trend-following design, and like most trend-following it is built to be wrong often but right big. The headline numbers bear that out.
Backtest performance
Over 451 days the strategy returned 23.5%, growing a $10,000 book to $12,349.63 — a 12.52% CAGR. The Sharpe ratio is a middling 0.95, with a max drawdown of 15.67%. Two numbers stand out. The win rate is just 28.79% across 136 trades: fewer than one trade in three is a winner, which is normal for trend-following but demands that the winners run. And turnover is a punchy 2,638%, so the strategy churns hard. Mercifully, total fees came to only $136 and FX cost was zero, so trading friction did not meaningfully dent the result here.
Validation: the red flag
This is where the shine comes off. The walk-forward validation did not pass. Across four folds, three were positive:
- Fold 1 (Aug 2024–Jan 2025): +5.85%, Sharpe 1.23
- Fold 2 (Jan–Jul 2025): −7.31%, Sharpe −1.05, 17.15% drawdown
- Fold 3 (Jul–Dec 2025): +25.59%, Sharpe 3.32
- Fold 4 (Dec 2025–May 2026): +13.34%, Sharpe 2.15
The out-of-sample return of 13.34% and OOS Sharpe of 2.15 look respectable, but the deflated Sharpe ratio tells the real story: DSR of 0.476. After adjusting for the 6 trials run against the data, the strategy clears only about half the bar we would want. The probabilistic Sharpe ratio (0.893) is healthier, but DSR is the number that accounts for selection, and it is weak. Fold 2's momentum crash — a 7% loss into a 17% drawdown — shows the regime dependence plainly: when trends whipsaw, this strategy bleeds.
Recent live activity
Perhaps the most telling signal is what the live book is not doing. Across every scheduled run from October 1 through October 8, the record reads the same: 0 executed, 0 rejected, cash $10,000, total $10,000. The strategy is fully in cash and has placed no trades. Its trend-break logic is finding nothing strong enough to hold, so it is sitting out — defensible discipline, but a reminder that the 23.5% figure is backtested, not live earnings.
Verdict
Dual-momentum has a coherent thesis and a flattering backtest, but it fails the test that matters most: out-of-sample validation after accounting for trial count. The low win rate and heavy turnover are tolerable features of the style; the regime fragility and sub-threshold DSR are not. For now it is prudently idle. We would want to see it survive a tighter validation gate — and actually deploy capital — before treating that 23.5% as anything more than a hopeful backtest.