The Thesis
Momentum SMA20 Rider is a textbook trend-follower. It ranks a 24-name large-cap universe (AAPL, MSFT, NVDA, JPM, XOM and peers) by how far each trades above its SMA50, then buys the leaders that confirm strength by holding above their SMA20. Exits are rules-based and unsentimental: cut when RSI14 pushes past 75 (overbought exhaustion) or when price closes below the SMA20 (trend structure broken). When the book is full, capital rotates out of laggards and into stronger ranks. It is a clean expression of "ride winners, cut the broken," and it carries a live status.
Recent Activity
The last two weeks show the rotation engine at work. On 1 October the strategy sold AAPL (5 @ $329.83) and ABBV (8 @ $260.50); by 5 October it was buying ABBV back (7 @ $264.48) alongside fresh AAPL (5 @ $333.78) and CAT (2 @ $847.03) positions. A full PFE round-trip — bought 70 @ $28.42 on 24 September, sold 70 @ $27.60 on 2 October — closed at a small loss, a reminder that not every confirmed uptrend follows through.
Scheduled runs are frequent but often idle: several late-September sessions executed zero trades, while 2 October executed three and rejected one. The live paper book stands at $9,709.87 as of 5 October — modestly below its implied $10k start, so the live leg has yet to validate the backtest's optimism.
Backtest and Validation
On paper the record is strong: 81% total return, final equity of $18,100 over 1,233 days, a 12.89% CAGR, and a Sharpe of 0.98. Walk-forward validation passed, with 3 of 4 folds positive and a standout final fold (May 2025–Aug 2026) returning 43.48% at a 2.18 Sharpe and just 5.38% drawdown. The deflated metrics are encouraging — PSR of 0.986 across 24 trials, suggesting the headline Sharpe is unlikely to be pure luck.
Strengths and Risks
The strengths are real: robust out-of-sample behaviour, a disciplined exit framework, and validation that survives a multiple-testing haircut.
But the risks deserve equal billing. The win rate is only 37.13% — this strategy loses on nearly two trades in three and relies on a handful of large winners, which is emotionally and statistically demanding. Turnover is extreme at 26,971%, generating 1,427 trades and $1,427 in fees — a meaningful drag that compounds over time. Fold 3 (Feb 2024–May 2025) was essentially flat at −0.28% while enduring a 23% drawdown, and the overall max drawdown of 20.91% is not trivial. The DSR of 0.583, while positive, is more sober than the PSR.
Verdict: a credible momentum engine with validated edge, but one whose live performance and fee burden warrant close watching before conviction grows.