The setup arrived. Four and a half hours after it was mapped, and the desk did not take it.
The morning plan was a long at 4378.93, the 0.764 of the overnight leg 4367.31 to 4416.53, with a stop at 4364.85 and a target at 4410.62. For most of the day it looked academic. London coiled between 4391.98 and 4411.40 for five hours and never came near it.
Then the overlap opened and the tape rolled:
- 13:30 to 14:45, price fell from 4404.95 to 4377.44. Three hundred pips in seventy five minutes.
- The entry at 4378.93 was crossed at 14:40. The low went 15 pips through it and turned.
- Equilibrium at 4391.92 was reclaimed at 15:35.
- 15:45 to 15:55 it ran again, high 4407.59, which is 32 pips short of the target.
On paper that is 287 pips in favour and a little over two R of open profit, unresolved. Nothing was ordered. The desk is flat and stays flat, on observation only, for the rest of August. Every setup still gets mapped, graded and written down, because the record is the point and a paper result counts the same as a real one for measurement.
Worth noting what the map got right and what it did not. The level was right and the patience was right, because at 12:25 a Trump headline threw a 373 pip candle that swept the morning low and closed near its high, and the desk called it a news flush rather than a displacement. It left no imbalance and was fully retraced inside five minutes. The real move came two hours later and looked nothing like it.
What the map did not get right is the shape of the day. A bullish read that spends six hours waiting and then gets its fill on a three hundred pip sell-off is correct on direction and lucky on timing, and it is worth saying that out loud rather than banking the level as skill.
The Iran story ran hot all afternoon and gold ignored it. Trump said the US will claim the Strait of Hormuz after defeating Iran, threatened Oman, called for a white flag, and confirmed a backchannel. Gold sold three hundred pips into that and then bought it all back. The Gaza de-escalation line in the same cluster is the more likely driver of both halves.
Levels
Trade decision log
Flat — nothing mapped
No call on the board. The desk is watching, not chasing — most sessions end here, and that is the method working rather than a gap in the record.
Before & after
The morning chart, left un-redrawn
The striking feature
Morning thesis vs market reality
End-of-day verdict
What we learned
- Driver that mattered most
- Two of them pulling opposite ways, and the smaller one won. Fed hike odds kept fading all day and the dollar sat at a two month low, which is the steady bid under gold. Against that, Trump escalated on Iran for four straight hours and gold sold 300 pips into it before buying every pip back. The likeliest explanation is the Gaza de-escalation line that landed in the same cluster. Whatever the cause, the day closed where the rate story said it should, not where the war story said it should.
- Did price respect the zones?
- Almost to the pip, and in both directions. The 0.764 at 4378.93 was tapped at 14:40 and the low went 15 pips through it before turning. The target at 4410.62 was taken at 15:55. The high of that push, 4416.30, stopped 2 pips under the 4416.53 that anchored the whole grid. Every number came from the overnight leg and none of them moved all day.
- Carry forward into tomorrow
- Three things. First, the grid was drawn at 10:00 and paid at 15:55 without a single adjustment, which is the argument for mapping early and then leaving it alone. Second, the 12:25 news candle was correctly refused: it swept the morning low, closed near its high, and left no imbalance, and the real move came two hours later and looked nothing like it. Third, and least comfortable, a bullish read that waits six hours and then fills on a 300 pip sell-off is right on direction and lucky on timing. Bank the level, not the skill.
The method
Two anchor prices on the displacement leg, and a fixed fib grid derives the whole order: entry floored at the 0.764 that names the method, stop at 1.05, target a fixed span below. R is a consequence of the grid, never a number chosen after the fact. Every read is timestamped and left un-redrawn.