Thesis
Dual-momentum is a classic trend-following idea: hold the strongest trending names by 60-day return, and exit on a trend break. It runs over a 24-name universe of large-cap US equities spanning tech (AAPL, MSFT, GOOGL, NVDA), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), and staples/industrials (PG, KO, WMT, COST, CAT, HON). The strategy is currently marked live.
Backtest Performance
Over 451 days the strategy returned 23.5%, growing a notional $10,000 to $12,349.63 — a CAGR of 12.52%. The Sharpe ratio of 0.95 is respectable but not spectacular, and the max drawdown of 15.67% is meaningful for a large-cap book.
The texture of the returns is telling. Win rate is just 28.79% across 136 trades — the signature of a momentum system that loses small and often while relying on a handful of large winners. Turnover is very high at 2,638%, meaning the book churns many times over the test window. With fees of $136 (one per trade) the frictional drag is modest here, but real-world slippage on this much turnover deserves scrutiny.
Validation
This is where the picture turns cautious. The walk-forward validation did not pass. Across four folds, three were positive and one — the January-to-July 2025 window — lost 7.31% with a punishing 17.15% drawdown and a -1.05 Sharpe. The other three folds were strong (up to 25.59% with a 3.32 Sharpe in H2 2025).
The out-of-sample block returned 13.34% with a 2.15 Sharpe, which looks encouraging. But the deflated Sharpe ratio (DSR) is only 0.476, below a passing threshold, even as the probabilistic Sharpe ratio (PSR) reads 0.893. With six trials logged, the deflation is doing exactly its job: discounting an in-sample Sharpe that may not survive multiple-testing bias. The single deeply negative fold is the clearest warning — the edge is regime-dependent, not all-weather.
Recent Live Activity
Here the story is quiet to the point of dormant. The last six scheduled runs (Aug 3–10, 2026) each executed 0 trades and rejected 0, leaving the book fully in cash at $10,000 the entire time. No 60-day trend has met the entry bar recently, so the strategy is sitting out — a defensible, disciplined stance, but one that means it is currently earning nothing and contributing no fresh live evidence of its edge.
Verdict
Dual-momentum has a coherent thesis and a genuinely good headline backtest. The strengths are real: strong H2-2025 and early-2026 folds and a healthy OOS Sharpe. The risks are equally real: a low win rate that depends on tail winners, one badly negative fold, very high turnover, and — most importantly — a failed validation gate driven by a weak deflated Sharpe. Combined with a live book that has done nothing for a week, the honest read is promising but unproven. I would keep it on paper and demand more out-of-sample runs before trusting it with real conviction.