The thesis
momentum-code runs one of the oldest ideas in the book: buy the strongest positive movers in its universe and cap each position. The universe is a 24-name slate of U.S. large caps — AAPL, MSFT, NVDA, JPM, XOM and the usual mega-cap suspects. There is no shorting, no sector tilt, no mean-reversion safety valve. It simply leans into whatever is already winning. Clean to reason about, and honest about what it is.
Backtest and validation
Over 451 days the strategy turned $10,000 into $11,976.21 — a 19.76% total return, or roughly 10.6% CAGR. Turnover was 92.74% and total fees just $5, so costs are not the story here. The story is the risk-adjusted numbers: a Sharpe of 0.65 and a maximum drawdown of 20.48%. That is a lot of pain for the reward, and the Sharpe is firmly in mediocre territory.
The walk-forward validation is more flattering at first glance: all four folds were positive (+15.38%, +3.27%, +12.18%, +13.71%), and out-of-sample Sharpe on the latest fold reached 1.93. So why does the gate report passed: false? Because across 6 trials the deflated Sharpe ratio (DSR) lands at just 0.338 — the probabilistic Sharpe (PSR 0.811) looks fine on its own, but once you penalise for the number of attempts, the edge no longer clears the bar. Four green folds built on a 0.65 full-sample Sharpe is exactly the kind of result that evaporates under multiple-testing scrutiny. The gate is doing its job.
One oddity worth flagging: winRate reads 0 across 5 recorded trades. That is consistent with a book of open buys and no closed round-trips yet, rather than five outright losses — but it means we have no realised win/loss evidence to lean on.
Recent activity
Here is the uncomfortable part. The last executed trades were a cluster of buys back on 31 May–1 June 2026 (MSFT, HON, BAC, then NVDA and XOM). Since then the live book has done nothing. Every scheduled run from 14 to 21 October reports 0 executed, 1–3 rejected, with cash pinned at $608.79 and total equity drifting between roughly $10,395 and $10,474. The strategy wants to trade and can't — $608.79 isn't enough to open the positions it's signalling, so it has quietly become buy-and-hold since June.
Verdict
Strengths: a transparent thesis, trivial costs, and a backtest that at least stayed positive in every window. Risks: a weak Sharpe, a 20%+ drawdown, a validation gate it genuinely failed, and — most pressing — a live book that has been functionally frozen for weeks on insufficient cash. The headline return is real but modest, and it is currently being produced by inertia, not strategy. This one needs a hard look at position sizing and cash management before anyone reads the green numbers as a green light.