One fixed rule, re-run at every month-end since 2016.
Quality names the engine is long, ranked by 12-month momentum at the last session of each month, the top 20 held at inverse-volatility weights times the engine's exposure, no leverage, after costs. This page is the published record of that rule. The same code builds the live portfolio for subscribers.
Growth of $10,000, monthly, log scale
Model portfolio at full exposure against equal-weight holding of the same names and the Nasdaq 100.
Every year
Losing years against holding are dimmed. The last row is the running year.
| Year | Model portfolio | Holding all names | Nasdaq 100 |
|---|---|---|---|
| 2016 | +19.8% | +21.1% | +10.2% |
| 2017 | +37.8% | +35.3% | +32.5% |
| 2018 | +6.7% | +1.6% | -2.7% |
| 2019 | +38.2% | +45.0% | +43.2% |
| 2020 | +69.5% | +49.1% | +47.8% |
| 2021 | +16.8% | +35.5% | +27.2% |
| 2022 | -15.4% | -19.7% | -32.2% |
| 2023 | +54.9% | +43.7% | +52.1% |
| 2024 | +123.7% | +38.4% | +27.5% |
| 2025 | +131.3% | +37.7% | +21.2% |
| 2026 | +50.4% | +22.9% | +14.4% |
| Per year since 2016 | +44.1% | +27.4% | +20.0% |
| Worst drawdown | -37% | -31% | -37% |
The same book at each risk profile
Subscribers pick a volatility target; the book is scaled into cash to meet it, never levered. This is what each setting would have done.
| Profile | Return /yr | Worst drawdown | Sharpe | Volatility | Time invested |
|---|---|---|---|---|---|
| Defensive · 12% vol target | +18.1% | -15% | 1.15 | 13% | 49% |
| Balanced · 18% | +25.0% | -22% | 1.15 | 19% | 69% |
| Aggressive · 28% | +33.0% | -28% | 1.17 | 25% | 84% |
| Full send · always invested | +44.1% | -37% | 1.21 | 33% | 96% |
| Holding all names | +27.4% | -31% | 1.20 | 20% | 100% |
| Nasdaq 100 | +20.0% | -37% | 0.84 | 22% | 100% |
Past monthly picks
The top 20 at each month-end, most recent first. The current book is a subscriber feature; the history is public so the record can be checked.
The honest caveat
The universe is today's tracked names, assembled in 2026 knowing which ones won, so equal-weight holding of it beat the Nasdaq 100 by itself; read the gap between the first two columns, not the level. On a broad set of 445 US stocks that were never selected for anything, the same rule did +43.2% a year against +21.3% holding; on sector ETFs, with no survivorship at all, it earns the few points a year the academic literature reports, at equal Sharpe and higher volatility.
Momentum strategies suffer sharp reversals; this window holds the 2018, 2020 and 2022 drawdowns but not 2000 or 2008. Costs are modelled at 5 basis points per unit of turnover with next-open fills. Backtested, not live. The research note explains why the per-name signals became a sizing layer and the portfolio became the product, and the JSON artifact holds every row, the frozen-input hash and the monthly picks. An agent can pull the same artifact through the MCP server.
Hesper Atlas is an educational publication, not investment advice. The portfolio is one rules-based construction, identical for everyone, and knows nothing about your situation. Past performance, historical or backtested, does not predict future results. Data: end-of-day. Read the terms.