News process: after a losing streak
Posted: Mon Oct 05, 2026 7:21 pm
A losing streak does not change the calendar. That is the line I write at the top of the sheet before I touch size.
Last month I took three flat-to-red London mornings in a row on EURUSD, all of them after I had already decided the news map was "manageable". What actually happened was softer: I kept the blackout windows but let the post-release first pullback rule stretch by a couple of minutes because I wanted a winner back. The calendar process was fine. The streak process was not.
So the rule now is mechanical. After two red sessions that tagged a news-adjacent trade, the next Tier-1 window on my pairs is stand-aside only, even if the setup looks clean. I still fill the night-before sheet: consensus, priced move in bp or %, and the time I am allowed to look again. I just do not click.
What I ignore during a streak: any impulse to "make back" the loss into a CPI or NFP print, any size bump "because the level is obvious", and any rewrite of blackout length the morning of. The sheet from the night before stays as written.
What I still do: log the skipped window with a one-line reason. That log is what stops the streak story from turning into a mythology about bad brokers or bad luck. The news process after a losing streak is mostly about refusing to let P&L edit the calendar.
Last month I took three flat-to-red London mornings in a row on EURUSD, all of them after I had already decided the news map was "manageable". What actually happened was softer: I kept the blackout windows but let the post-release first pullback rule stretch by a couple of minutes because I wanted a winner back. The calendar process was fine. The streak process was not.
So the rule now is mechanical. After two red sessions that tagged a news-adjacent trade, the next Tier-1 window on my pairs is stand-aside only, even if the setup looks clean. I still fill the night-before sheet: consensus, priced move in bp or %, and the time I am allowed to look again. I just do not click.
What I ignore during a streak: any impulse to "make back" the loss into a CPI or NFP print, any size bump "because the level is obvious", and any rewrite of blackout length the morning of. The sheet from the night before stays as written.
What I still do: log the skipped window with a one-line reason. That log is what stops the streak story from turning into a mythology about bad brokers or bad luck. The news process after a losing streak is mostly about refusing to let P&L edit the calendar.