Thesis
Momentum SMA20 Rider is a classic trend-following engine dressed in disciplined risk rules. It ranks a 24-name large-cap universe — mega-cap tech (AAPL, MSFT, NVDA), financials (JPM, BAC, V, MA), healthcare, and staples — by distance above the SMA50, then buys only names whose price is also holding above the SMA20. The logic is coherent: the SMA50 gap identifies leadership, and the SMA20 filter demands the short-term structure still confirms it. Exits are rule-bound rather than discretionary — positions are cut when RSI14 pushes above 75 (overbought) or price closes below the SMA20 (trend broken), with capital rotating into higher-ranked names when the book is full.
Backtest & Validation
Over 1,233 days the strategy returned 81% (final equity $18,100 from a $10k base, ~12.9% CAGR) at a Sharpe of 0.98 and a max drawdown of 20.9%. What sets it apart from the usual overfit momentum system is the validation. Across 4 walk-forward folds, 3 were positive, out-of-sample return was 43.5% at a Sharpe of 2.18, and the Probabilistic Sharpe Ratio landed at 0.986 even after deflating for 24 trials (Deflated Sharpe 0.583). That is a genuinely robust profile — the edge survives being tested on data the model never saw.
The fold detail is where balance matters. Fold 4 (mid-2025 to Aug 2026) carried the record with a 43.5% return and a shallow 5.4% drawdown, while Fold 3 (2024–2025) was flat-to-negative (−0.28%) with the deepest 23% drawdown. In other words, most of the reported strength is concentrated in favorable trending regimes; in the 2024–25 chop, the strategy essentially treaded water and gave back capital.
Risks
Two numbers deserve scrutiny. The win rate is just 37%, which is normal for cut-losers-let-winners-run momentum but means the equity curve depends on a minority of large winners — an unforgiving profile psychologically and one vulnerable to whipsaw markets. And turnover of ~27,000% across 1,427 trades signals heavy churn; in a live account with real spreads and fees, that friction compounds quickly.
Recent Live Activity
The live paper book tells a more sober story. On 2026-08-06 it deployed into financials — buying BAC, V, MA, JPM, and ABBV — but every scheduled run since (Aug 10 through Aug 17) executed zero trades, with total equity drifting between roughly $9,840 and $9,982 and cash parked at $626.50. The book is essentially fully invested and holding, currently sitting just below its starting stake. That is not a red flag — a trend-rider is supposed to sit tight while positions run — but it is a reminder that impressive backtest math has yet to translate into live gains here.
Verdict
Strong validation, honest drawdowns, and a clear rule set make this one of the more defensible momentum strategies in the lab. The caveats — regime dependence, low hit rate, and high turnover — are inherent to the style, not defects. Watch whether the live book can convert its financials tilt into the kind of run Fold 4 delivered.