Thesis
momentum-code follows one of the oldest edges in systematic equity trading: buy what is already moving. The strategy screens its 24-ticker large-cap universe — spanning mega-cap tech, financials, healthcare, consumer staples, and energy — for the top positive movers on each scheduled run, then enters capped positions to prevent concentration risk. Simple, transparent, and battle-tested as a concept.
Backtest Performance
Over 451 days of simulated history, the strategy returned 19.76% on a starting book of roughly $10,000, compounding to $11,976 at a 10.6% CAGR. Transaction costs were minimal ($5 total fees, zero FX drag), reflecting the all-USD universe.
The headline number is encouraging, but the accompanying Sharpe of 0.65 and maximum drawdown of 20.48% deserve scrutiny. A Sharpe below 1.0 means the strategy is not generating a particularly smooth ride for the return it delivers, and a fifth of the book going underwater at peak stress is a meaningful tail risk for a momentum approach.
Walk-Forward Validation
The four-fold walk-forward is where the picture gets genuinely interesting — and where the strategy both shines and stumbles.
| Fold | Period | Return | Sharpe | Max DD |
|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +15.38% | 2.47 | 6.12% |
| 2 | Jan 2025 – Jul 2025 | +3.27% | 0.46 | 16.55% |
| 3 | Jul 2025 – Dec 2025 | +12.18% | 2.31 | 5.21% |
| 4 | Dec 2025 – May 2026 | +13.71% | 1.93 | 6.32% |
All four folds are positive — a meaningful consistency signal. Folds 1, 3, and 4 posted Sharpe ratios above 1.9, which is excellent for a simple momentum rule. The out-of-sample period (Fold 4) returned 13.71% with a 1.93 Sharpe, the cleanest real-world proxy in the set.
Fold 2 is the outlier: a 3.27% return with a 0.46 Sharpe and nearly 16.6% drawdown. That period (roughly the first half of 2025) was a regime test — choppy momentum, elevated volatility. The strategy survived, but barely earned its keep.
Validation Gate: Not Yet Passing
Despite the positive fold record, the strategy currently fails the Headmars validation gate. The culprit is the Deflated Sharpe Ratio (DSR) of 0.338 — well below the 0.5 threshold typically used to adjust for the number of trials tested (6 here). With a PSR of 0.811, there is a reasonable probability the true Sharpe exceeds zero, but the DSR correction for multiple testing is harder to dismiss. This is the framework working as intended: four positive folds can still be the product of selection bias across a small trial count.
Recent Live Activity
momentum-code went live with five executed buys at the end of May and early June — XOM, NVDA, BAC, HON, and MSFT — building a diversified position across energy, semiconductors, financials, industrials, and software.
Since then, the scheduled daily runs from June 15–22 have executed zero trades, with between one and three candidates rejected each session. Cash has sat at $608.79 throughout. Portfolio value has drifted from $9,712 to $9,562 over the period, reflecting mark-to-market movement in the existing holdings rather than new activity. The position-cap logic is doing its job — the strategy is not forcing trades into positions it already holds — but the pattern of persistent rejections suggests the current holdings are themselves the top movers, keeping the queue blocked.
Strengths and Risks
Strengths: All folds profitable; strong OOS Sharpe; low fee drag; clean, auditable logic; concentrated in liquid large-caps with tight spreads.
Risks: DSR failure flags overfitting risk given the small trial count; Fold 2 demonstrates vulnerability in low-momentum regimes; a 20%+ max drawdown on a simple rule warrants position sizing discipline; and the current live stall — while mechanically correct — means the strategy is not compounding while holding a partially deployed book.
The next test will be how momentum-code navigates the next regime shift. The OOS numbers are promising; the validation math demands more evidence before full deployment confidence is warranted.