Thesis and Approach
momentum-code applies one of the most durable edges in equity markets: short-term price momentum. Each scheduled run scans a curated 24-stock universe spanning mega-cap tech (AAPL, MSFT, NVDA, GOOGL), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), consumer staples and discretionary (PG, KO, WMT, COST, MCD, NKE, HD), energy (XOM, CVX), industrials (CAT, HON), and media (DIS). It buys the top positive movers from that pool and caps each position size — a sensible guardrail that prevents the signal from piling into a single crowded name.
Live Deployment: First Week
The strategy went live on May 31, 2026, with a $10,000 allocation. The opening run placed three buys: BAC (38 shares @ $51.60), HON (8 shares @ $237.86), and MSFT (4 shares @ $450.24). A follow-on run on June 1 added XOM (13 shares @ $148.69) and NVDA (8 shares @ $224.20), nudging the portfolio to a short-lived peak of $10,033.
The picture since then has been quieter. Daily runs on June 2–5 executed zero new trades, with one to three candidates rejected each session — suggesting the position caps are doing their job and no new entrants cleared the momentum threshold. Portfolio value drifted to $9,627 by June 5, a roughly 4% pullback from the post-deployment high.
Backtest Performance
Over 451 calendar days the full-period backtest returned +19.76%, with a CAGR of 10.6% and a Sharpe of 0.65. The max drawdown reached 20.48% — meaningful for a blue-chip universe with position caps. Turnover was high at 92.74%, though total fees amounted to just $5, reflecting the paper-trading environment.
The walk-forward validation across four folds is the more informative lens:
| 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 result that is not trivial to achieve across distinct, non-overlapping market regimes. The most recent fold (Fold 4) also serves as the out-of-sample benchmark, posting a 13.71% return and a 1.93 Sharpe, the best risk-adjusted reading in the dataset. Fold 2 is the visible weak point: a thin 0.46 Sharpe paired with a 16.55% drawdown during the turbulent early-to-mid 2025 period.
Where Validation Stalled
Despite consistent directional performance, momentum-code did not pass the platform's validation gate. The culprit is the Deflated Sharpe Ratio (DSR) of 0.338: after correcting for six optimization trials and non-normal return distributions, the statistical probability that the observed Sharpe reflects a genuine edge — rather than selection noise — sits at only 34%. The Probabilistic Sharpe Ratio (PSR) of 0.811 is more encouraging on its own, but the DSR penalty exists precisely to keep strategies with limited trial history honest.
The reported win rate of 0% is a metric to watch rather than act on: with just five executed trades in the backtest, the sample is too thin to draw conclusions.
Assessment
momentum-code shows a credible and consistent directional edge. Four consecutive positive walk-forward folds, a strengthening OOS Sharpe, and a clean cost structure are genuine positives. The risks to watch are the 20.48% historical max drawdown — large relative to the modest CAGR — Fold 2's sensitivity to volatile regimes, and the DSR flag that appropriately restrains confidence until live trade volume grows. Re-running validation after another quarter of live data should meaningfully sharpen the picture.