Ahmed Aaref JewelrySuez · since 1986🇪🇬 العربية

Home › Gold forecasts › Friday 9 October 2026

Day five of five · the weekly close

Overnight, both legs our weekly call stood on gave way. Tonight’s close decides it.

Gold closed Thursday at $4,133, inside the scenario we ranked second, with the yield at 5.24% against a band we wrote as 5.25%–5.33%. One hundredth short, for the third time in a week. Then the picture flipped overnight: Trump ruled out a strike on Iran before the elections, oil fell, the 30-year auction was strong, and the yield came off its 24-year high. Gold jumped to $4,184 this morning, above the $4,150 line that 70% of our weekly weight sits below. The week closes at midnight; this is the last day of grading the weekly call. The map is below.

Spot now $4,184 · Thursday’s close $4,133 · 21K in Suez EGP 6,160 ·

✅ Forecast correct · number 0.01 below the band

Grading Thursday — inside the second scenario, yield 5.24% against a 5.25% floor

We published: 40% yields keep going ($4,060–$4,120), 35% holding between the week’s low and the line ($4,120–$4,165), 25% Iran returns to the picture ($4,165–$4,215). The close was $4,133.12, inside the second.

The confirming number was “the 10-year between 5.25% and 5.33%”. The 30-year auction was strong and the yield closed at 5.24%. One hundredth below the band. By the rule: correct by chance, and we write it that way.

But three times in one week the number has landed on or beside the edge (Monday, Wednesday, Thursday), and that is not chance; it is a flaw in our method. The yield bands we write are too narrow (0.08 points) in a market where the yield moves 0.10 a day. From today, any yield band is at least 0.12 points wide, and thresholds are “at or above”. The second correction this week, published in the open like the first.

Record: 35 published · 30 correct · 91% (30 of 33 graded; the weekly call and today’s are graded tonight). Thursday’s report → · How grading works →

🗺️ The map for tonight — the weekly close grades the weekly call

Saturday’s call, unedited: 40% the decline continues slowly ($4,030–$4,150) · 30% $4,000 breaks ($3,900–$4,030) · 30% a bounce ($4,150–$4,300). So that nobody grades it by mood afterwards, here is the map ten hours before the close:

If the close isThe weekly callConfirming number
Below $3,900❌—
$3,900–$4,030✅ (second scenario)30-year above 5.7% or Brent above $108
$4,030–$4,150✅ (first · 40%)10-year above 5.2%: met now (5.26%)
$4,150–$4,300 ← gold is here now ($4,184)✅ (third · 30%)2-year below 4.7%: not met (4.80%)
Above $4,300❌—

If it closes here: the weekly call is correct by the rule, but in the scenario we ranked third and without its confirming number, so “correct by chance”, with 70% of our weight on the wrong side. If it falls back below $4,150: correct, in order, for the reason. The difference between the two is $34, with ten hours and the Michigan survey at 5:00 PM still to come. The verdict is at midnight, published tomorrow with next week’s call.

💰 Where things stand

 NowReference
Spot gold, per ounce$4,184+$51 on Thursday’s close
24K, per gram, we sellEGP 7,040up from 6,988
22K, per gram, we sellEGP 6,453up from 6,406
21K, per gram, we sellEGP 6,160up EGP 45
18K, per gram, we sellEGP 5,280up from 5,241

📈 21K at EGP 6,160 (buy 6,100), from our counter reset at 1:40 PM today, up EGP 45 on yesterday. If it is reset, the live number is at today’s prices. The dollar rate implied in our counter price is 52.33 (52.68 yesterday): the ounce rose 1.4% and the gram 0.7%; the local market did not catch all of the move and the gap is stored in the implied rate, as usual. For the week: since last Friday’s close ($4,140) the ounce is up 1.1% and the gram 0.5% (from 6,130 to 6,160).

What is moving the market today

🎯 Our forecast for Friday’s close

Covering the Friday 9 October session — New York close = the weekly close · yield bands under the new rule (at least 0.12 points)

40%The bounce holds above the line → $4,160–$4,230
The two drivers that pressed gold all week (oil and yields) reversed in one night, and nothing on today’s calendar reverses them back. The market closes the week above $4,150 and below the $4,230 neckline. Confirming number: a close at $4,160 or above with the 10-year at 5.28% or below
35%Profit-taking before the weekend, back to the line → $4,120–$4,160
Trump’s words have been said twice before and broken, the Iranians themselves do not believe them, and nobody wants to hold a position over a weekend with strike options ready. Gold gives back part of the $50 jump and closes around the line. Confirming number: a close between $4,120 and $4,160 with the 10-year between 5.28% and 5.40%
25%A jump to the neckline → $4,230–$4,280
Heavy short-covering after two weeks of decline, or official confirmation from Tehran of talks that pushes oil below $100, takes gold to $4,230, last seen on 29 September. Confirming number: a close at $4,230 or above with the 10-year below 5.22% or Brent below $100

🇪🇬 21K in Suez on Saturday morning: EGP 6,036 – 6,301, most likely EGP 6,125 – 6,228, assuming the implied dollar rate holds at 52.33.

✏️ Why is “continuation” today the continuation of the bounce rather than of the decline? Rule three says the first scenario is the continuation of the trend, and the week’s trend is down. But that trend stood on two legs (oil above $100 and the yield above 5.3%), and both broke in one night. When the cause changes the trend changes, and the rule follows the cause, not the line on the chart. If we are wrong, we will write that we misread the night.

The lesson today — a forecast fails when the cause changes, not when the number moves

Our weekly call stood on two causes: oil above $100 and the yield above 5.2%. For four days both were present and gold was in the box we named. In one night Trump pulled oil down and an auction pulled the yield down, and gold left the box.

That does not mean the call was wrong the day it was written. It means the event that breaks it happened, and we had given that event a 30% chance. What we take from it: when someone tells you “gold will do this this week”, ask what it stands on, and what they will say if that changes. Whoever has no answer has no forecast.

Converting the ounce to a gram in Suez

If the ounce is21K per gram works out near
$3,900 (floor of the weekly call)EGP 5,742
$4,030EGP 5,933
$4,100 (the week’s low)EGP 6,036
$4,133 (Thursday’s close)EGP 6,085
$4,150 (the dividing line)EGP 6,110
$4,184 (now)EGP 6,160
$4,200EGP 6,184
$4,230 (the neckline)EGP 6,228
$4,250EGP 6,257
$4,300 (ceiling of the weekly call)EGP 6,331

📍 Raw metal at today’s local conversion factor (1.4723), before making charge, hallmarking and tax.

Levels that change the picture

$4,300Ceiling of the weekly call
$4,280Ceiling of today’s third scenario
$4,230The neckline
$4,200First resistance
$4,160Floor of today’s first scenario
$4,150The line that grades the weekly call
$4,131This morning’s low
$4,120Floor of today’s second scenario
$4,100The week’s low

Tonight one number grades two calls: $4,150. Above it, the weekly call is correct by chance and today’s call is correct in order. Below it, the weekly call is correct for the reason and today’s lands in its second scenario. Above $4,300 the weekly call fails. The verdict is at midnight.

Today and tomorrow, in Cairo time

Friday 5:00 PM — Michigan consumer sentiment (expected 48.1).
Friday midnight — the weekly close on which the weekly call and today’s forecast are graded.
Saturday 10 October, morning — the weekly verdict: the weekly call and the five daily calls, what was right, what was right by chance and why, both corrections, and next week’s call.
Next week: Tuesday, September CPI, the most important number before the October Fed meeting.


Sources and method

Spot price, Thursday’s close ($4,133.12), oil, Treasury yields, the dollar index, hike odds, the 30-year auction, the Waller remark and ETF flows from Trading Economics; Trump’s Truth Social statement, the Pentagon planning and Iran’s stance from Axios (8 October); the hurricane, the nine tankers and Hormuz exports from Trading Economics; technical levels ($4,200 / $4,250 / $4,300, the $4,131 low) from Mitrade; local prices from our own published price list, reset at 1:40 PM on 9 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.

← Previous report All forecasts