The thesis
momentum-code runs a deliberately simple idea: buy the top positive movers across a 24-name universe of US large caps — think AAPL, MSFT, NVDA, JPM, XOM, UNH — with a cap on each position. No factor stacking, no regime overlay. Just chase strength, size it sensibly, repeat. Simplicity is a virtue here; it means fewer knobs to overfit and a story you can actually explain.
Backtest and validation
The headline backtest is respectable: 19.76% total return over 451 days, a 10.6% CAGR, and a final equity of $11,976 on a $10k base. Turnover of 92.74% and just $5 in total fees confirm this is a low-churn book, not a high-frequency grind.
The more interesting story is in validation. Across four walk-forward folds, all four were positive — 15.38%, 3.27%, 12.18%, and 13.71% — with out-of-sample return of 13.71% and a notably strong OOS Sharpe of 1.93. That consistency is the best thing this strategy has going for it.
And yet the validation gate reads passed: false. Why? The deflated Sharpe ratio (DSR 0.338) tells the tale. The probabilistic Sharpe (PSR 0.811) looks healthy in isolation, but with 6 trials behind this configuration, the deflation for multiple testing pulls confidence down below the bar. The full-sample Sharpe of 0.65 is mediocre, and the 20.48% max drawdown is a real gut-check for a $10k book. The gate is doing its job: four green folds are encouraging, but they aren't proof against selection bias.
Live activity
Here the picture cools considerably. The last executed trades were back on 31 May–1 June 2026 — buys in MSFT, HON, BAC, NVDA, and XOM. Since then, every scheduled run has printed the same pattern: 0 executed, 1–3 rejected, cash pinned at $608.79. Total equity has drifted between roughly $10,515 and $10,671 — up modestly, but on positions the strategy is holding, not adding to.
The repeated rejections suggest the book is effectively fully invested with too little cash to act on new signals. That's not a bug so much as a design consequence: a position-capped, mostly-deployed portfolio has little dry powder, so fresh momentum candidates get turned away at the door.
The balance sheet
Strengths: a clean, legible thesis; four-for-four positive folds; strong OOS Sharpe; negligible costs.
Risks: the validation gate failed on multiple-testing grounds, the full-sample Sharpe is thin, drawdowns approach 20%, and — most immediately — the live engine has been idle for weeks, unable to fund new entries. A reported win rate of 0 across only 5 recorded trades likely reflects open positions rather than realized losses, but it's a reminder how little live evidence exists so far.
The backtest earns a second look; the live deployment needs a cash-management fix before it can express the edge it claims.