The thesis
momentum-code runs a simple, legible idea: buy the top positive movers across a 24-name large-cap universe — spanning tech (AAPL, MSFT, NVDA), financials (JPM, BAC, V, MA), staples (PG, KO, WMT) and energy (XOM, CVX) — with a per-position cap so no single winner can dominate. It is a classic cross-sectional momentum play, and its virtue is that you can explain it in one sentence.
Recent activity
The live book has gone quiet. The last executed orders date to late May and 1 June 2026 — buys in MSFT, HON, BAC, NVDA and XOM. Since then, every scheduled run from 10–18 September logged zero executions and one to three rejections apiece, with cash pinned at $608.79 and total equity drifting between roughly $10,267 and $10,509. In practice the strategy is now holding its June positions and turning away new candidates, most likely because the leftover cash cannot clear the per-position sizing. That is momentum in name but buy-and-hold in behaviour right now.
Backtest and validation
Over 451 days the backtest returned 19.76% (final equity $11,976, about 10.6% CAGR) on just five trades, with fees of only $5 and no FX cost. Encouragingly, all four walk-forward folds were positive: 15.38%, 3.27%, 12.18% and 13.71%, with the most recent fold showing a 13.71% out-of-sample return and a 1.93 Sharpe. Individual fold Sharpes ran as high as 2.47.
And yet validation is marked failed. The reason sits in the deflation stats: across six trials the deflated Sharpe ratio is just 0.338, and the full-period Sharpe only 0.65 — well below the crisp per-fold numbers. The probabilistic Sharpe of 0.811 is respectable but not conclusive. In plain terms, once you account for how many variants were tried, the apparent edge shrinks toward the ordinary.
Strengths and risks
The strengths are genuine: consistent positive folds, very low trading costs, and a transparent rule set. Four-for-four positive out-of-sample windows suggest the signal has not simply overfit a single lucky period.
The risks are equally real. The reported win rate is 0% — but with five buys and no closed round-trips, there is nothing to score yet, so treat that headline as uninformative rather than damning. Max drawdown of 20.48% on the full run is heavier than any single fold (worst was 16.55%), a reminder that stitched-together windows hide the fuller pain. And the failed deflated-Sharpe gate is the honest brake here: promising, not proven. Before upgrading conviction, I would want to see the live book actually rotate — clear cash and take a sell — rather than quietly sit on its June entries.