How we research and write

The byline

Every article is published under the byline NovaLens Research. The byline stands for the method on this page, not for one person.

Method

Each article covers one idea and starts from the questions readers ask about it. The research comes before the writing: published papers, original data and documented practice from working quants.

Every worked example and every chart is computed by a script with a fixed seed, so the same inputs give the same numbers each time. Simulated data is labelled as simulated, on the chart and in its caption, with its assumptions.

Sources

Every number is either cited or labelled. A cited number names its source, links to it and gives the date it was read. Papers, original data and official documentation come first, in open versions where they exist.

A claim about what professional or systematic quants do carries a source of its own. An article never presents a NovaLens result as a track record.

Review

Before an article is published, an independent review checks its facts, numbers, sources and tone against the rules below. The final wording is approved before it goes live.

The rules: describe how things work and never tell anyone what to trade; no trade calls, no forecasts and no performance claims; name only published NovaLens tools; quote a published backtest only with its basis and the label "past results, not a forecast".

Corrections

When a fact, a number or a source turns out to be wrong, the article is corrected and its Updated date changes. Every article is checked again within 12 months of its last update, or sooner when a cited source changes.

To report an error, write to hello@novalens.io.