Thesis
Dual-momentum is a classic trend-follower: it holds the strongest-trending names by 60-day return and exits on a trend break. The universe is a 24-name large-cap basket spanning tech (AAPL, MSFT, NVDA), financials (JPM, V, MA), healthcare (JNJ, UNH), and staples (PG, KO, WMT). The logic is intuitive and well-documented in the literature — ride winners, cut losers — but it lives or dies on how cleanly trends persist versus whip back.
Recent Activity
The live sleeve has been quiet. The last six scheduled runs (2026-08-19 through 2026-08-26) each report 0 executed, 0 rejected, with cash at $10,000 and total equity at $10,000. In other words, the strategy is currently parked entirely in cash — no name in the universe is clearing its entry threshold, so it is sitting out. That is by design for a trend model in a choppy tape, but it also means there is no live P&L yet to corroborate the backtest.
Backtest Performance
Over 451 days the strategy returned 23.5% (CAGR 12.52%), ending at $12,349.63 from $10,000. Sharpe was 0.95 with a max drawdown of 15.67%. Two numbers deserve attention. First, the win rate is just 28.79% across 136 trades — this is a fat-tailed profile where a handful of large winners carry a majority of small losers, entirely characteristic of momentum. Second, turnover is 2,638%, extremely high; the model churns the book many times over, and while fees here were modest ($136 total), that level of trading is fragile to slippage and cost assumptions in live markets.
Validation — The Red Flag
Walk-forward validation did not pass. The mechanics are worth reading honestly:
- 3 of 4 folds were positive, but fold 2 (Jan–Jul 2025) lost 7.31% with a 17.15% drawdown and a -1.05 Sharpe across 60 trades — the strategy's worst regime.
- Folds 3 and 4 were excellent (+25.59% and +13.34%), producing an out-of-sample return of 13.34% at a 2.15 Sharpe.
- Probabilistic Sharpe (PSR) is a healthy 0.893, but the Deflated Sharpe Ratio is only 0.476 across 6 trials.
The DSR is the tell. Once you deflate for the number of configurations tried, the edge is not statistically convincing — which is why the gate blocked it. The strong aggregate return leans heavily on two favourable folds, and fold 2 shows how badly the model behaves when trends reverse.
Verdict
The strengths are real: a coherent thesis, respectable risk-adjusted returns, and disciplined cash-raising when no trend qualifies. But the risks are equally real — a sub-30% win rate, punishing turnover, a demonstrated ugly drawdown regime, and a failed deflated-Sharpe test. The current all-cash posture is prudent, not a malfunction. This strategy earns continued paper monitoring, but the validation result argues firmly against sizing it up until live trades and a cleaner robustness profile back the backtest.