momentum-code: Riding the Leaders — Strong Out-of-Sample Signal, but Validation Gate Still Open

Thesis and Universe

momentum-code follows one of the oldest edges in equity markets: buy what is already moving up. Each daily run scans a 24-name large-cap universe — tech heavyweights (AAPL, MSFT, GOOGL, NVDA), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), consumer staples and discretionary (PG, KO, WMT, COST, MCD, NKE, HD), and energy/industrials (XOM, CVX, CAT, HON, DIS) — and rotates into the top positive movers, capping each position to limit concentration risk.

The universe is deliberately blue-chip. This is not a small-cap momentum play chasing illiquid runners; it is a rotation strategy designed to stay liquid and fee-efficient. Total fees across the entire backtest period came to just $5, reflecting the low-churn nature of a five-trade book.

Backtest Performance

Over 451 days the strategy compounded to +19.76% total return, implying a 10.6% CAGR — comfortably ahead of cash but worth benchmarking against a passive S&P 500 allocation over the same window before drawing conclusions. The full-period Sharpe of 0.65 is modest, and a 20.48% max drawdown means the strategy is not immune to extended selling pressure.

Turnover came in at 92.74%, which sounds high but resolves to roughly five position changes over the observation period — consistent with a slow-rotation approach rather than daily churn.

One metric worth flagging: win rate is reported as 0%. This almost certainly reflects a data artifact (e.g., open positions counted as unrealised) rather than a strategy that has never booked a gain, given the positive cumulative return. It should be investigated before the figure is surfaced publicly.

Walk-Forward Validation

The four-fold walk-forward tells a more nuanced story:

All four folds are positive — a meaningful signal. Fold 2 stands out as the stress period: the drawdown nearly trebled and the Sharpe collapsed to 0.46. That roughly coincides with the early-2025 volatility window, suggesting the strategy struggles when momentum reverses sharply across large-caps simultaneously.

The out-of-sample return of +13.71% (Fold 4, which served as the true hold-out) with a Sharpe of 1.93 is encouraging. That is the cleanest read of live-conditions performance.

Why Validation Has Not Passed

Despite the positive fold record, the strategy currently sits at validation: false. The culprit is the Deflated Sharpe Ratio (DSR) of 0.338 — well below the 0.5 threshold the platform requires. DSR penalises a strategy for the number of parameter trials run during development (here, six), and with a full-period Sharpe of only 0.65 there is limited headroom to absorb that haircut. The Probabilistic Sharpe Ratio (PSR) of 0.811 is healthier, indicating an 81% confidence the true Sharpe exceeds a benchmark of zero, but DSR is the binding constraint.

In plain terms: the strategy shows real signal, but the evidence is not yet strong enough — relative to the search effort — to green-light capital escalation.

Recent Activity

The five most recent executed trades (late May / early June 2026) rotated into XOM, NVDA, BAC, HON, and MSFT — a cross-sector sweep that reflects the universe-wide scan working as designed. Since those buys, the daily scheduler has run without executing new trades, with one to two rejections per session. Portfolio value has oscillated between roughly $9,534 and $9,868 on a cash base of $608.79, suggesting the open book is experiencing normal mark-to-market drift.

Bottom Line

momentum-code demonstrates a legitimate momentum signal in a liquid large-cap universe. The walk-forward record is clean, and the out-of-sample fold is its best yet. The path to validation is straightforward: accumulate more live performance history to push the DSR above threshold without adding parameter trials. Resist the temptation to over-fit. The strategy earns its place in the paper-trading sandbox — it does not yet earn autonomous capital allocation.