The thesis
momentum-code runs a deliberately simple idea: buy the top positive movers across a 24-name large-cap universe — the mega-cap tech, financials, healthcare, staples and energy names you'd expect — with a cap on any single position. There's no clever timing or regime filter here; it's a naive momentum tilt, and simplicity is both its appeal and its ceiling.
Backtest and validation
Over a 451-day backtest the strategy returned 19.76%, ending at $11,976 on $10k, for a 10.6% CAGR. The headline Sharpe is a modest 0.65 and the max drawdown is a hefty 20.48% — a meaningful pain point for a two-figure annual return. Turnover was 92.74% on just five trades, and total fees came to a trivial $5.
The validation picture is genuinely encouraging in places. All four walk-forward folds were positive (15.38%, 3.27%, 12.18%, 13.71%), and the most recent out-of-sample fold delivered 13.71% at a Sharpe of 1.93. The Probabilistic Sharpe Ratio of 0.811 is respectable. Notably, individual fold Sharpes (2.47, 0.46, 2.31, 1.93) run far above the full-period 0.65 — the concatenated equity curve carries drawdown across fold boundaries that the segmented view smooths over.
Why it failed the gate
Despite four positive folds, validation was marked failed. The tell is the Deflated Sharpe Ratio of 0.338 against six trials — once you penalise for the number of variants tried, confidence in the edge collapses. With only five trades per fold, the sample is thin enough that a couple of lucky winners can carry the whole result. The reported win rate of 0 almost certainly reflects unclosed positions rather than five losing bets, but it underlines how little round-trip evidence exists. This is a textbook overfitting risk: a clean-looking curve built on very few decisions.
Live activity: stalled
The live account tells a quieter story. The last executed trades were a cluster of buys — XOM, NVDA, BAC, HON, MSFT — on 31 May and 1 June 2026. Since then, every scheduled run from 3 September through 10 September logged "0 executed, 1–2 rejected" with cash pinned at $608.79. Total equity has drifted between roughly $10,458 and $10,671, so the book is effectively buy-and-hold while new orders bounce — most likely blocked by the per-position cap and thin cash.
Verdict
momentum-code is a promising sketch, not a proven system. The consistent positive folds and strong recent OOS Sharpe are real strengths; the failed DSR gate, 20%+ drawdown, and five-trade sample are real reasons for caution. Until it can deploy capital rather than reject orders — and accumulate a longer trade record — treat the backtest as a hypothesis, not a track record.