A news blackout that ignores live spread is theatre. A spread filter that ignores the calendar is how you get clipped on a “quiet” Tuesday that was not quiet. The useful desk work is defining how the two rules interact when they disagree.
How I link them in practice:
Calendar blackout is the floor — I am flat for the written window regardless of how calm the chart looks.
Spread filter can extend the blackout, never shorten it. If spread is still blown after the timer, I stay flat.
Journal tags both: calendar-flat vs spread-extended-flat so I see which rule saved the week.
The interaction matters more than either rule alone. Humans love exceptions; linked rules reduce them.
Do your spread filters ever override a news timer (I argue they should only lengthen it) — how have you coded that on a human desk?
When the two rules conflict in the heat of the moment, the written hierarchy removes the debate: calendar wins on duration; spread wins on extension. I do not allow a “looks tradeable” override from either side.
Agree that spread can only lengthen the blackout. The detail I'd add is what "normalised" means. If it means the first tick below the cap, a single tight quote in the middle of a still-wild market reopens the book. I require the spread to stay under the cap for a full minute before the extension ends. On a human setup that's easy to do by eye with a spread indicator that colours the reading.
The case where I let spread act on its own is unscheduled news. A headline on a quiet Tuesday has no calendar window to extend, so the spread filter is the only rule that fires at all.
For that I use the same hierarchy in reverse. If spread blows out with nothing on the calendar, I treat it as an instant 15 minute blackout and then apply PropScalpDesk's full minute under the cap before reopening.
The cap also needs to depend on the hour. EURUSD at 0.2 pips during London and 0.6 at 22:30 are both normal, so a single fixed cap is either too loose for the morning or keeps me out all evening.
I set it at roughly twice the median spread for that hour, taken from a couple of weeks of my own logged quotes. It is crude, but it stopped the evening false alarms.
One gap neither rule covers is depth. Spread can look normal two minutes after a release while the book behind the quote is still thin, and that shows up as slippage on the stop rather than on the entry.
LondonScalper, do you log the spread at the moment of the exit as well as the entry? That was what showed me where my real news cost was.
One weakness in the hourly cap from my last reply showed up with releases outside London hours. Because the evening median is wider, twice that median leaves a lot of room, and an Australian number after midnight London can move AUDUSD hard without the spread ever crossing the cap.
So for pairs with their own late releases I set the cap from the quietest London hours instead of the release hour. It keeps the filter honest on the one pair that actually reacts, and the hourly cap stays in place for the rest.
The other edge case is the rollover around 22:00 London. A calendar blackout that ends at 21:55 can hand over directly to several minutes of rollover spread, and the extension logic then reads it as news.
I now treat rollover as its own fixed window rather than letting the spread rule discover it every night. That keeps the journal tags clean, since a spread extended day should mean news and not just the end of the New York session.
On my exit spread question, I log the spread at entry, at exit and at the stop price whenever a stop is hit. The third number is the one that changed my sizing on release days.
Spread is more than a cost. It reflects how comfortable liquidity providers feel.
A steady, narrow spread means providers are confident and competing. A spread that widens slowly suggests growing uncertainty or thinning participation. A sudden widening signals news, a gap in the order book or a technical issue.
Watching spread alongside price offers early warning. Before many news releases, spreads begin to widen a minute or two ahead. Before large moves in thin markets, spreads sometimes jump while price stands still.
A practical habit is to keep a small spread display on the chart and note its normal range for each pair in each session. Deviations then stand out.
After a spike, spreads normalise gradually. Entering while they are still elevated pays a premium. Waiting for normal levels costs a few seconds and often saves a pip.
Is spread watched as a signal here or purely as a cost, and has it ever warned of a move?