Home › Gold forecasts › Thursday 15 October 2026
Gold closed Wednesday at $4,105, inside the scenario we ranked second. Core inflation came in at 0.3%, exactly as that scenario required against a 0.2% forecast, and the 10-year yield did spike to 5.33% within minutes of the print. But the yield we said had to close at 5.30% or above closed at 5.20%, because equities sold off through the afternoon and money ran into Treasuries. By the rule: right by chance. And at 4 PM yesterday we had written, in as many words, “we are not calling this now — the verdict is the New York close.” At 4 PM both conditions were met. By 11 PM one of them was gone. This morning the ounce is back at $4,146 with the yield at a three-week low.
Spot now $4,146 · Wednesday’s close $4,105 · 21K in Suez EGP 6,185 ·
We published at 1 PM, two and a half hours before the inflation print: 40% core in line ($4,130–$4,200), 35% core at 0.3% or above ($4,060–$4,130), 25% core cool ($4,200–$4,270). The close was $4,105.57, inside the second. Right range, wrong order.
And the mechanism ran exactly as written: core CPI came in at 0.3% on the month against a 0.2% forecast, core at 2.6% on the year against 2.5%, headline at 3.7% against 3.6% — while headline on the month landed at 0.6%, precisely as forecast. The 10-year went from 5.19% to 5.33% within minutes, and gold broke $4,100 to its lowest level since early August.
But the number we wrote was: “a close at $4,130 or below with the 10-year at 5.30% or above.” The yield closed at 5.20%, because US equities — technology and AI names in particular — sold off through the afternoon and money moved into Treasuries as a haven, pulling the yield back down. The condition broke at the close for a reason that had nothing to do with inflation. By the rule: right by chance.
And this is the part that matters most: at 4 PM yesterday we wrote, word for word, “we are not calling this now… the verdict is the New York close, not the price while I am typing.” At that moment both conditions were met — core at 0.3%, the yield at 5.33%, gold at $4,097. We could have closed the book and claimed the win. We waited until 11 PM, and the close took the number away from us. That is the whole difference between a page that publishes its call before the event and one that writes it afterwards.
Correction number four: on a day with a scheduled economic release, the confirming number has to be the release value itself plus the price closing inside the range. That is all. The driver’s level at the close cannot be a third condition, because the driver can reverse hours later for an entirely unrelated reason — which is what happened. The scenario said “if core comes in at 0.3% or above, gold goes to $4,060–$4,130”, so those two things are what we should be graded on. On ordinary days with no scheduled release, the driver’s level stays a condition exactly as before. The rule applies from today, not retroactively — yesterday’s call stays on the record as right by chance.
Record: 41 published · 38 graded · 35 correct · 92% (the weekly call and today’s are still open). Wednesday’s report → · How grading works →
At $4,146 the ounce is still inside the second scenario of Saturday’s call (35% hot inflation, $4,060–$4,200). The inflation did come in hot — but that box’s confirming number (the yield at 5.28% or above, or Brent at $108 or above) was touched and not held: the yield reached 5.33% during Wednesday’s session and is 5.09% now, and Brent reached $106 and is $104 now. The first scenario (40%, $4,200–$4,340) is only $54 away. So the week can still turn on a single session. The verdict is at midnight on Friday — and correction number four does not touch the weekly call, because the weekly call is not a release-day scenario.
| Now | Reference | |
|---|---|---|
| Spot gold, per ounce | $4,146 | +$40 on Wednesday’s close |
| 24K, per gram, we sell | EGP 7,068 | +6 |
| 22K, per gram, we sell | EGP 6,479 | +5 |
| 21K, per gram, we sell | EGP 6,185 | +5 |
| 18K, per gram, we sell | EGP 5,301 | +4 |
➡️ 21K up just 5 pounds to EGP 6,185 (buy 6,120, unchanged), from our counter reset today. If it is reset again, the live number is at today’s prices. 📍 And here is something worth seeing. Yesterday we said the gram was sitting above its value because the ounce had fallen and the gram had not followed: the implied dollar rate reached 53.51 at Wednesday’s close. Today the ounce came back $40 and the gram moved 5 pounds — so the implied rate is back at 53.03. The gap closed without the gram having to fall; the ounce rose to meet it. 💡 That is the rule governing the local market: when the gram sits above its value, it waits for the ounce to catch up rather than coming down. Which is why local falls are always slower than local rises. Not a trick — just the shape of the market. Since last Friday’s close ($4,194) the ounce is down 1.15% and the gram down only 0.4%.
Covering the Thursday 15 October session — New York close · the first day under the new rule: the release value plus the price close, with no third condition on the driver
🇪🇬 21K in Suez tomorrow morning: EGP 6,027 – 6,296, most likely EGP 6,206 – 6,296, assuming the implied dollar rate holds at 53.03.
✏️ Note that none of today’s three scenarios carries a condition on the yield’s level at the close. That is correction number four applied directly: today has a scheduled release, so the grading runs on the release and on the price. The yield stays in the explanation, not in the condition.
The scenario we wrote yesterday said one thing: if core inflation comes in at 0.3% or above, gold goes to $4,060–$4,130. That is a scenario that resolves in minutes — the number lands at 3:30 PM and the market moves immediately.
But we attached a third condition to it that is measured seven hours later: the yield’s level at the close. And in those seven hours something with no connection to inflation — an equity selloff — pulled the yield back down. So the condition broke even though the scenario itself worked.
The rule that comes out of it: a condition has to be measured on the same clock as the scenario. A scenario that rests on a scheduled release is confirmed by the release and the price. A scenario that rests on the day’s whole direction is confirmed by the driver’s level at the close. A “quiet” scenario is confirmed by the size of the move — that was correction number three.
And it applies to you too: next time someone ties a move in gold to a cause, ask one question — when did the cause move, and when did gold move? If there is no answer with a clock on it, that is hindsight, not an explanation.
| If the ounce is | 21K per gram works out near |
|---|---|
| $4,040 (floor of today’s third scenario) | EGP 6,027 |
| $4,060 (floor of the weekly call) | EGP 6,057 |
| $4,100 (support) | EGP 6,117 |
| $4,105 (Wednesday’s close) | EGP 6,125 |
| $4,146 (now) | EGP 6,185 |
| $4,160 (today’s resistance) | EGP 6,206 |
| $4,200 (the line) | EGP 6,266 |
| $4,220 (ceiling of today’s first scenario) | EGP 6,296 |
| $4,340 (boundary of the weekly call’s third scenario) | EGP 6,475 |
| $4,460 (ceiling of the weekly call) | EGP 6,654 |
📍 Raw metal at today’s local conversion factor (1.4918), before making charge, hallmarking and tax.
| $4,340 | Boundary of the weekly call’s third scenario |
| $4,230 | The neckline |
| $4,220 | Ceiling of today’s first scenario |
| $4,200 | The line that grades the weekly call |
| $4,160 | Today’s resistance and the floor of the first scenario |
| $4,146 | The price now |
| $4,130 | First support |
| $4,105 | Wednesday’s close |
| $4,100 | Floor of today’s second scenario |
| $4,066 | Last week’s low |
| $4,040 | Floor of today’s third scenario |
Today at 3:30 PM: retail sales at 0.3% or below with core PPI at 0.3% or below opens the road to $4,200. Core PPI at 0.4% or above puts gold below $4,100. And tomorrow’s weekly close is what grades the weekly call — the line is $4,200.
Thursday 3:30 PM — September PPI (0.5% expected on the month, core 0.3%) · jobless claims
(200K expected) · retail sales (0.3% expected, after 1.2%) · 8:30 PM
Barkin · 9:00 PM Beige Book · 11:30 PM Hammack.
Friday: Fed Chair Kevin Warsh speaks — the last word before the blackout that
begins on Saturday ahead of the October meeting.
Friday midnight: the weekly close on which the weekly call is graded.
Spot price, Wednesday’s close ($4,105.57), the yield (session high 5.33%, close near 5.20%), oil (close $106 and $104 today), the dollar (close 102.55), the September inflation figures, the Goolsbee and Daly remarks, hike odds, the Thursday and Friday calendar and the PPI and retail-sales consensus from Trading Economics; the detail on the yield reversing after the equity selloff from Investing.com and BabyPips (14 October); technical levels ($4,230, $4,066) from Kitco News (9 October); local prices from our own published price list, reset on 15 October.
This report is the English edition. The Arabic edition is primary and was published first, at the same date’s Arabic page. Every figure here appears there.