Field Notes · 3 July 2026
Building a rejection column that you will actually use
Signal filtering fails when rejections live only in your head. Here is the journal column we force in the lab.
Entry columns fill themselves. Rejection columns starve. In the Signal Filtering Lab we treat that imbalance as the main problem, not a lack of indicators.
The minimum viable rejection line
Each rejected alert needs four fragments:
- Time — when the alert appeared
- Claim — what the indicator claimed (for example “MACD cross up on 1H”)
- Rule — the filter that stopped you
- Temptation — one honest word for why you almost ignored the rule
Without the temptation word, students rewrite history later and the filter never gets stress-tested.
Filters that survive a noisy week
We prefer filters tied to measurable chart facts: average volume, distance from a moving average, or a prior swing. Personality-based filters (“I felt rushed”) matter for coaching, but they are hard to grade in a group lab.
A mild warning
Some traders build twenty rejection rules and then freeze. Start with three. Retire any rule that never fires for two weeks — it is decoration.
Practice the habit in our Signal Filtering Lab, or bring a stubborn alert stack to a private chart review.