An edge that worked can quietly stop working. The lifetime average hides it — for a while.
No edge is permanent. Markets change regime, a setup gets crowded, your own execution drifts — and expectancy starts sliding. The trouble is that a lifetime average moves slowly, so by the time your all-time number turns, the recent damage is already deep. The decay shows up first in a recent window measured against your own baseline, not against zero.
Your book’s overall numbers can stay green while a single setup quietly rots underneath them. One decaying setup gets averaged out by the ones still working, so the whole-system view looks fine right up until it doesn’t. The same is true of profit factor slipping trade by trade — gross wins shrinking or losses growing while the headline holds. You have to look per-setup and per-window to see it coming.
Do not panic-rewrite everything on a losing streak — that is optimizing for noise. Audit the recent losers for a common cause: one setup, one session, one regime. If a single setup is dragging, pause it and confirm the rest still pays. Then re-check after another 10–20 trades: recovering, or a genuine shift? Kill what stopped working; keep what still does.
Reading about R is the easy part. The journal makes you keep the stats — grade the setup before you know how it ends, and hear the one thing to fix next.
Start alpha test