The thesis
Dual-momentum is about as plainly stated as a trend strategy gets: hold the strongest trending names by 60-day return, and exit when the trend breaks. It fishes from a 24-name large-cap pond spanning tech (AAPL, MSFT, NVDA), financials (JPM, V, MA), staples (PG, KO, COST), healthcare (JNJ, UNH) and energy (XOM, CVX). The logic is sound and well-trodden: let winners run, cut laggards, and lean on relative strength rather than forecasts.
Recent activity
Here the story is quieter than the thesis suggests. Across the last six scheduled runs — 21 through 28 October — the strategy executed zero trades every single day, holding a flat $10,000 entirely in cash. No positions, no rejections, no churn.
That is not a bug; it is the exit rule doing its job. When no name in the universe clears the trend filter, dual-momentum declines to force a bet and parks in cash. The upside is discipline. The downside is that a fully-cash book earns nothing and can sit out early-stage rallies until a trend is confirmed — the classic lag of any momentum system.
Backtest and validation
The full-sample backtest is the strategy's best face: a 23.5% total return over 451 days (12.52% CAGR), a Sharpe of 0.95, and a max drawdown of 15.67%, ending at $12,349. Two numbers deserve scrutiny. First, the win rate is only 28.79% — typical for trend-following, where a minority of large winners must pay for a majority of small losers, but it leaves little margin if the big winners stop showing up. Second, turnover is a heavy 2,638%, so execution costs and slippage matter more than the modest $136 in modeled fees implies.
More important, the strategy failed walk-forward validation. Three of four out-of-sample folds were positive, but the spread is wide:
- 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
Fold 2 is the warning: during that stretch the trend signal whipsawed badly, trading 60 times for a loss. The out-of-sample composite (+13.34%, Sharpe 2.15) actually looks strong, and the Probabilistic Sharpe Ratio of 0.893 is encouraging. But the Deflated Sharpe Ratio of 0.476 — below the 0.5 threshold, after accounting for six trials — is why the gate says no. Translation: once we penalize for the number of configurations tried, the edge is not statistically convincing.
Verdict
Dual-momentum is a coherent, disciplined strategy that behaves exactly as designed — including its current refusal to trade. The strengths are real: strong fold-3 and fold-4 performance and a solid headline Sharpe. The risks are equally real: a failed validation gate, a brutal fold-2 drawdown, a low win rate, and heavy turnover. For now it stays live but unproven — worth watching when it finally redeploys, but not yet a strategy I would size up with conviction.