Writing So the Next Researcher Can Pick It Up

A research note that only shows the final chart and the conclusion is close to useless to anyone else on the desk, because it hides the actual information: what data was used, what was tried and discarded, and why the final choice was made instead of the alternatives. A note that lets the next researcher pick up the thread needs to record the path, not just the destination.

Concretely, this means writing down the exact data window and universe used (so results can be reproduced rather than approximately guessed at), the specific things that were tried and rejected and why (so nobody re-wastes a week testing an idea that was already ruled out), and any known limitations or caveats discovered along the way (a data quirk, a period that had to be excluded, a parameter that was sensitive). A note that just says "signal X works, Sharpe 1.2" forces the next person to redo all the exploratory work from scratch to find out what "works" actually depended on.

The habit that makes this cheap rather than a burden is writing the note as you go, immediately after each meaningful step, rather than reconstructing it from memory at the end — by the time a result is final, the reasons for the choices made along the way are usually already forgotten.