Thesis
momentum-code is one of the simplest strategies in the Headmars lab: at each scheduled run, it scans a 24-stock universe of U.S. large-caps — spanning tech (AAPL, MSFT, NVDA, GOOGL), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), consumer staples and discretionary, industrials, and energy — and buys the top positive movers, subject to a per-position cap. No short selling, no leverage. The strategy is intentionally readable: a hand-written rule, reviewed and approved at deployment.
Live Activity (May 31 – June 2, 2026)
The strategy deployed with $10,000 on May 31. Its opening three trades were MSFT (4 shares @ $450.24), HON (8 shares @ $237.86), and BAC (38 shares @ $51.60), leaving $4,335 in cash — consistent with the per-position cap doing its job.
On June 1, two more buys followed: NVDA (8 shares @ $224.20) and XOM (13 shares @ $148.69), with one order rejected — likely a momentum filter finding no qualifying mover at that slot. Portfolio value ticked up to $10,033.
June 2 was quiet: zero executions, three rejections. Total value slipped slightly to $9,993. That pattern — active on strong days, dormant when no stock clears the positive-mover threshold — is the strategy behaving as designed. Periods of inactivity are a feature, not a bug.
Backtest & Fold Validation
Over 451 days (roughly August 2024 – May 2026), the full backtest returned +19.76% (CAGR ~10.6%), with a Sharpe of 0.65 and a maximum drawdown of 20.48%. Fees were minimal at $5 total; no FX cost given the all-USD universe.
The four-fold walk-forward tells a more nuanced story:
| 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. The out-of-sample (OOS) return of +13.71% with a Sharpe of 1.93 is the strongest single-fold result, which is encouraging: the most recent, unseen period is where the strategy looks best.
What the Validation Gate Flagged
Despite four green folds, the Headmars validation system returned failed. The culprit is likely the Deflated Sharpe Ratio (DSR) of 0.338 — well below the 0.5 threshold commonly used to control for multiple-testing bias across the six trials run. The Probabilistic Sharpe Ratio (PSR) of 0.811 is acceptable but not decisive. With only five trades in the backtest, the sample is thin; the DSR penalty for repeated trials on sparse data is severe by design.
The zero win-rate metric warrants a note: with only five recorded trades and a winRate: 0 figure, this almost certainly reflects incomplete trade-close data rather than a strategy that never profits. Still, it should be resolved before drawing conclusions.
Strengths and Risks
Strengths: Simple, auditable rule. Broad blue-chip universe limits single-stock blow-ups. High-Sharpe recent fold (1.93) suggests the signal remains live. Turnover of ~93% keeps the book refreshed without excessive churn.
Risks: 20.5% maximum drawdown is meaningful for a paper-trading account and would test live investors during a sharp correction (Fold 2's 16.5% intra-fold drawdown is a preview). Low trade count means backtested statistics have wide confidence intervals. The failed DSR gate is the system correctly saying: we've seen this thesis before — prove it on more data.
Outlook
momentum-code is a solid baseline strategy in a class that historically earns a risk premium in large-cap equities. The live results so far are unremarkable but rational. The real test will come in the next volatile week — whether the rejection logic keeps the portfolio flat during broad selloffs is where this strategy either earns its Sharpe or gives it back.