Gold did the most constructive thing it has done in a week: it bought its own washout. The December contract opened near 4,408, sold off to 4,365 in the first part of the session, then reversed and rallied all the way to 4,454 before settling at 4,437, up 0.38 percent and near the top of its range. A morning flush that gets fully reclaimed and then extended is the signature of dip-buying, not distribution, and the settle sitting near the day's high tells you buyers controlled the afternoon and were willing to pay up into the close.
The driver was a soft run of US data. Retail sales contracted, sentiment collapsed, and the dollar eased, which is the familiar tailwind for the metal. Working against it, the ten-year yield actually rose as traders leaned on the inflation side of the reports, and some of the geopolitical premium leaked out as the Iran situation was described as static. That cross-current is why gold gained modestly rather than surging. The bigger picture is a two-timeframe conflict: price is above its five, twenty and fifty-day averages, and the composite jumped from 32 to 48 percent buy in a single day, but the metal is still below its hundred and two-hundred-day averages and roughly 23 percent under its high. This is a mid-range bounce inside a larger correction, not a fresh leg to new highs, and that distinction is exactly how aggressively to press it.
The 100-day is the whole question
Everything about gold's medium term comes down to one level. The hundred-day average sits at 4,487, and it lines up almost exactly with the dense resistance shelf just overhead, where the pivot resistance and a band of deviation levels converge. Gold has now failed beneath that hundred-day twice in three sessions. Until it closes back above it, the read stays what it is: a recovering market that keeps running into the ceiling the prior correction built. A daily close above 4,487 would be the first genuine signal that the medium-term correction is turning, and it's worth waiting for rather than anticipating. Below it, the rising twenty and fifty-day averages keep the short-term uptrend intact. That is what makes dips buyable, and it is the entire basis of the plan.
A tailwind that isn't one-directional
The macro case for gold has genuinely improved, and it's worth being precise about why it still isn't a green light. Payrolls printed negative, the consumer is visibly weakening, and a central bank that's forced to hold rather than tighten is structurally supportive for a metal that rewards falling real rates. But the one-year inflation expectation ticked up to 4.3 percent, and the ten-year yield rose into the weak data, so the real-yield calculation that anchors the metal isn't moving cleanly in gold's favour. And the safe-haven premium is leaking: a market that won't rally hard on tanker attacks in the Strait of Hormuz has much of its fear bid already in the price, which means a concrete de-escalation headline over the weekend is a real downside risk. The tailwind is there. It just isn't blowing in only one direction.
Buy the dip, respect the low
The plan is a buy of a controlled pullback into the 4,400 to 4,420 base rather than a chase into the resistance shelf, because the short-term oscillators are stretched and first tests of a dense ceiling usually fail. The stop sits below 4,362, beneath Friday's 4,365 washout low, because a decisive break of that low neutralizes the recovery and shifts the structure back to neutral. The targets run to the 4,443 pivot, then the 4,454 high, then the 4,486 band where the pivot resistance and the hundred-day converge. A weekend de-escalation headline or a sharp dollar rebound negates the thesis, so step aside on a disorderly gap rather than trading into it. How we grade these afterward is in our performance methodology.
A morning washout to a fresh low was bought back completely and then some, and a market that pays up into its own close after being handed weak data has decided the dip is the opportunity.
Gold was handed weak data, flushed to a new low, and closed near its high anyway. A market that pays up into its own close after news like that has already decided the dip is the opportunity.
This is the read our members get every session, before the bell, with the levels drawn and the setup defined. See how the same dealer-positioning work turns into systematic signals.
View pricingThe complete data picture
Every number behind Friday’s plan, charted first; the full numeric reference follows underneath.
Full numeric reference — every remaining figure from the review
| Average | Value | Price vs 4,437 |
|---|---|---|
| 5-day | 4,435.7 | +1.6 (holding it) |
| 20-day | 4,234.1 | +203 (rising) |
| 50-day | 4,228.7 | +208 (rising) |
| 100-day | 4,487.3 | -50 (the first wall) |
| 200-day | 4,620.3 | -183 (overhead) |
| Level | Reference |
|---|---|
| 4,400.6 | one-SD support (top of the buy) |
| 4,391-4,392 | retracement grouping |
| 4,377.2 | first pivot support |
| 4,365.5 | Friday low, invalidation |
| 4,334.1 | second pivot support |
| 4,268.1 | third pivot support |
| Metric | Reading |
|---|---|
| Proxy close | 401.42, +0.6% |
| Inflection zone | ~396, just beneath spot |
| IV rank / skew rank | 27% / ~60% |
| Implied move | ~$5.46 (1.3 to 1.4%) |
| Gamma configuration | room to travel, not pinned |
| Concentration expiry | mid-September |
| Input | |
|---|---|
| Retail sales / sentiment | -0.6% / 51 (soft consumer) |
| Dollar index | 99.636, -0.31% |
| 10-year yield | 4.695, rose (mild headwind) |
| Sept hike odds | ~32% from 35% |
| 1-yr inflation exp | 4.3% (the dovish wrinkle) |
| Hormuz | static; premium leaking |
| When | Event |
|---|---|
| Sun 19:50 | Japan GDP |
| Sun 22:00 | China activity data |
| Mon 08:30 | Canada CPI + NY Fed survey (second-tier) |
| Tue | US housing starts + production |
| Mid-week | FOMC minutes (the real gold catalyst) |
| Aug 27-29 | central-bank symposium |
