The thesis
momentum-code runs a simple, legible premise: buy the top positive movers in a 24-name large-cap universe, capped per position. The universe spans megacap tech (AAPL, MSFT, NVDA, GOOGL), financials (JPM, BAC, V, MA), healthcare, staples, and energy — a broad, liquid pool that keeps the strategy from over-concentrating in any single sector by construction. It is currently flagged live.
Backtest performance
Over 451 days the strategy returned 19.76%, ending at $11,976 in equity for a 10.6% CAGR. That headline number is respectable, but the risk profile deserves scrutiny: the Sharpe ratio is a modest 0.65, and the maximum drawdown reached 20.48% — meaning an investor had to stomach a peak-to-trough loss roughly equal to a full year's gain. Turnover of 92.74% on just 5 trades and $5 in total fees suggests a low-frequency, high-conviction approach rather than active churn.
One caveat: the reported win rate is 0% despite the positive return. With only 5 trades recorded — all recent buys — this almost certainly reflects open, unrealized positions rather than a genuine string of losers. Treat the win-rate metric as uninformative here, not as a red flag.
Walk-forward validation
The validation story is genuinely encouraging on robustness: across 4 folds, all 4 were positive (15.38%, 3.27%, 12.18%, 13.71%), with out-of-sample return of 13.71% and a strong OOS Sharpe of 1.93. The Probabilistic Sharpe Ratio sits at a healthy 0.811.
And yet the gate reports passed: false. The likely culprit is the Deflated Sharpe Ratio of 0.338 — once performance is discounted for the 6 trials tested, the edge no longer clears the confidence bar. This is the validation framework doing exactly its job: four positive folds are reassuring, but they are not the same as statistically deflated proof of skill. The strategy looks promising, not proven.
Live activity: a stalled book
The most pressing concern is operational. The last executed trades were a batch of buys at the turn of June — XOM, NVDA, BAC, HON, and MSFT. Since then, every scheduled run from July 13 through July 20 executed zero trades and rejected one to two candidates each time. Cash has been frozen at $608.79 throughout, and account total has drifted between roughly $9,720 and $9,921 — below the backtest's final equity and effectively flat-to-down in live paper trading.
The pattern points to a fully-deployed book with too little cash to act on new signals: the strategy sees movers it wants to buy but cannot fund the positions. That is a portfolio-construction gap, not a thesis failure.
Verdict
momentum-code has a clean thesis and robust walk-forward behaviour, but it is held back by a modest deflated edge and, more urgently, a cash-starved live book that has stopped trading. Resolving position sizing so signals can actually execute is the near-term priority.