Archive·July 22, 2026·ARCHIVE edition

10 names that earned the most return per unit of risk — 2026-07-22

A point-in-time ranking of S&P 500 as of 2026-07-22, return measured against the volatility it took to get there, computed from the price history available on that date.

This is an archive edition. It was computed retrospectively from the price history that existed on 2026-07-22 — every series is truncated to that date before anything is ranked, so there is no lookahead in the result — but it was not published on the day. Archive editions carry no market commentary, because what the coverage was about that week cannot be reconstructed after the fact.

The ranking measures return earned per unit of risk, favouring shallow drawdowns, weighting return-to-risk (34%), 1-year max drawdown (26%), 1-year volatility (22%), 1-year return (18%). 500 names cleared the liquidity and history gates on that date.

The list

At #3, WDC had returned 712.0% over the prior year with 75.1% volatility and a worst drawdown of -37.4%.

Reading an archive edition

The value of these is not the picks — it is seeing how a fixed rule behaved across different markets. A screen that only ever looks good in hindsight is a screen that was fitted after the fact. Ours is published in full and applied identically every time, so the archive is a record you can check the method against rather than a highlight reel.

Nothing here is tailored to your circumstances, and nothing here is a recommendation to buy or sell. Past performance does not predict future results.

Free newsletter

Get the next edition before anyone asks you to pay

Two editions a weekday, free — the same rankings, the same published method. Unsubscribe in one click.

Screenline publishes impersonal, rules-based rankings on a fixed schedule. The same content goes to every reader. Nothing here is tailored to your circumstances and nothing here is a recommendation to buy or sell any security. We are not a registered investment adviser. Rankings are computed from historical price data, which may contain errors. Past performance does not predict future results.