Short write-ups on rules-based strategies, how they are tested, and where the tests go wrong.
Every parameter you tune on the full history is a small leak from the future. Walk-forward analysis limits the damage: optimise on one window, trade the next, roll forward, and only ever report the stitched out-of-sample results.
Classic distance-based pairs trading earned steady returns in the 1990s and has been fading since. The spread still mean-reverts; it just does so faster than a daily strategy can capture.
Give two traders the same signals and different sizing rules and their equity curves will look unrelated within a year. Entry rules decide which trades you take; sizing decides whether you survive them.
The 50/200 crossover is the first system most people code, and the first one they abandon. On daily index data since 1990 it trades about once every two years, which means any single decade tells you very little about its edge.