Thesis
momentum-code runs a simple, legible idea: buy the top positive movers across a 24-name large-cap universe, capped per position. The universe spans mega-cap tech (AAPL, MSFT, NVDA, GOOGL), financials (JPM, BAC, V, MA), healthcare, staples, and energy — a diversified, liquid pool that keeps the momentum signal clean rather than chasing thin names. The status is live.
Backtest performance
Over 451 trading days the strategy returned 19.76%, growing a notional $10,000 to $11,976 for a 10.6% CAGR. That headline is respectable, but the risk profile deserves scrutiny: the Sharpe is only 0.65 and the max drawdown reached 20.48% — a drawdown almost as large as the total return. Turnover of 92.74% is moderate, and fees were negligible at $5 across just 5 trades. The reported 0% win rate against 5 trades looks like a bookkeeping quirk in how round-trips are counted rather than a literal all-loss record, given the positive equity curve; it should not be read at face value.
Validation
Here the picture gets more honest — and more cautionary. The walk-forward test ran 4 folds, all 4 positive, with out-of-sample returns of 15.38%, 3.27%, 12.18%, and 13.71%. Out-of-sample Sharpe on the final fold hit a strong 1.93. Yet the overall validation verdict is FAIL.
Why? The deflated metrics tell the story. The Probabilistic Sharpe Ratio is 0.811, but the Deflated Sharpe Ratio is just 0.338 after accounting for 6 trials of selection. In plain terms: once you penalise for how many variants were tested, the edge no longer clears the bar. Fold 2 is the tell — a 3.27% return with a 16.55% drawdown and a 0.46 Sharpe shows the strategy can stall badly in unfavourable regimes. Four-for-four positive folds are encouraging, but with only 5 trades per fold the sample is thin, and thin samples flatter momentum strategies.
Recent live activity
The live book is currently idle by circumstance. The last executed fills were a cluster of buys in late May and early June 2026 — MSFT, HON, BAC, then NVDA and XOM. Since then, every scheduled run from 21–28 August produced zero executions and one or two rejections, with cash pinned at $608.79 and total equity hovering between roughly $10,460 and $10,570. The strategy is effectively fully invested and cannot act on new signals without freeing capital — a practical liquidity constraint, not a signal failure.
Verdict
momentum-code is a clean, understandable strategy with a credible backtest and encouraging fold consistency. But the failed validation gate is the headline that matters: the deflated Sharpe says the apparent edge may be a product of trial selection, and the 20% drawdown means the ride is rougher than the return suggests. The stalled live book — starved of cash and rejecting orders for a week — also limits what we can learn from paper performance right now. Treat this one as promising but unproven: worth watching, not yet worth trusting.