Gold Week 34/2026: 19-Year Yield High, Gold Won't FallGoldOANDA:XAUUSDtohaitrieuLast week I set a Sell on gold at 4,416, with one condition: there had to be a Sell signal candle there before I entered. On Wednesday (12 August), price reached 4,449. I sat watching every candle and never saw what I needed. So I stayed out. The next day gold dropped straight back to 4,329. Two weeks ago I bragged that the mindless rule had saved me. This week I have to say it straight: that same rule just cost me a good trade. Anyone who has followed me a while will understand, that is the price of never entering on emotion. But the real story this week isn't the trade I missed. It's the US Treasury market. 📉 THE US BOND MARKET BREAKS, YIELDS BACK AT 2007 LEVELS On Monday (17 August), the yield on 30-year US Treasuries touched 5.31%. That is the highest reading since July 2007. There is more. This long-dated maturity has now held above the 5% mark for the longest stretch in nearly two decades. The 10-year sits at 4.69% as well. The reasons aren't new, but they are stacking up. US government spending keeps swelling, forcing enormous issuance of long-dated bonds to cover the deficit. More supply means sellers have to cut the price, and a falling bond price is exactly what a rising yield looks like. Inflation, meanwhile, has sat above the Fed's 2% target for five years. The war with Iran has just delivered an oil price shock on top of that. ⚖️ YIELDS ROSE AND GOLD DIDN'T FALL, THAT'S THE REAL NEWS Normally this is bad news for gold. Yields and gold carry a fairly strong inverse correlation, somewhere around minus 0.5 to minus 0.7. Yields up, gold down, that is what the textbook says. And yet gold is sitting at 4,396, barely moved from last week. I don't read this as gold being stubborn. I read it as the reason behind this rate rise being different. If yields were climbing because the Fed was about to tighten, the dollar would have to strengthen. The opposite is happening: the DXY is at 99.57 and has fallen for three straight sessions. Which means long-dated bonds are not being sold on growth expectations. They are being sold because the market is demanding a bigger premium for America's own budget and inflation risk. When US government bonds get questioned in the very role of safe haven, money doesn't go hunting for higher yield. Money goes hunting for the thing nobody can print more of. That is why the inverse correlation broke, and it broke in gold's favour. 🥇 GOLD ETFS ADDED OVER A BILLION DOLLARS IN ONE SESSION If you still doubt that reading, look at the Gold ETFs flow. In the 17 August session alone, the SPDR fund took in a net $1.010 billion, equal to 7.13 tonnes of gold. That was their strongest buying session of the entire month. Across the last six sessions, SPDR bought a net $1.85 billion, lifting holdings from 1,020.96 tonnes to 1,030.66 tonnes. The whole run contains exactly one session of net outflow. What stands out is the timing. The session SPDR poured in a billion dollars was the same session the 30-year yield printed its 19-year high. Big money did not treat that yield spike as a reason to sell gold. It treated it as a reason to buy more. 🌍 THE MIDDLE EAST JUST MISSED A DEADLINE There is one more reason money is finding its way into gold, and it doesn't sit in America. 17 August was the deadline on the 60-day memorandum of understanding between the US and Iran. The two sides signed it in order to negotiate a broader peace deal. The deadline passed with nothing signed. The war has now entered its 24th week, and the bottleneck is still the Strait of Hormuz. Oman has been negotiating separately with Iran over how to manage the strait. Trump promptly went on Fox News and threatened to bomb Oman if they got in America's way. Markets reacted within the session. Brent crude rose 2.7% to $90.87, dragging US stocks down as another layer of inflation pressure landed. This is the link that ties the two stories together. Expensive oil means sticky inflation, sticky inflation means more bond selling, and that leaves gold with more ground to stand on. 📊 A BULL FLAG IS FORMING ON THE H1 CHART Technically, what is drawing itself on the chart is a fairly clean Bullish Flag. The flag pole is the vertical run from 4,020 to 4,452 earlier this month, nearly $432 in just a few sessions. The flag is the seven sessions of sideways action since, boxed between 4,317 and 4,452. A bull flag is a continuation pattern, not a reversal. It signals a market catching its breath rather than changing direction, and its trap is convincing newcomers that this is a distribution zone. The breakout zone sits at 4,434 to 4,452. The flag's support zone sits at 4,296 to 4,317. Between them is 4,358 to 4,379, where the H1 rising trend line is climbing up to meet price. The technical table says the same thing. Daily S3 support sits at 4,369, weekly S1 support at 4,373. Two numbers landing on exactly that zone. ⚠️ WHAT WOULD MAKE ME WRONG I am not walking into this trade on blind faith. The daily RSI is at 63.61, already creeping toward overbought. Last week I used that very indicator to doubt the buyers, and I don't intend to forget it just because I have switched sides this week. More worrying, the 4,434 to 4,452 zone has capped price three times in seven sessions. A flag that drifts sideways too long right under resistance can be a sign of distribution rather than accumulation. There is also a 20-year bond auction this week. If it goes worse than expected, the market can shift from demanding a risk premium to dumping everything for cash, and gold gets sold along with the rest. 🎯 MY PLAN FOR THIS WEEK I have two Buy scenarios ready, not one. The first is to wait for a pullback. If price eases back to 4,368, right where daily support, weekly support and the rising trend line meet, I enter a Buy. The second is to wait for a breakout. If price doesn't pull back and simply runs, I wait for a 4-hour candle to close decisively above 4,452 before entering. Buy on pullback — Entry: 4,368, Stop loss: 4,348, TP1: 4,434, TP2: 4,452, TP3: 4,564 Buy on breakout — Entry: 4,452, Stop loss: 4,432, TP1: 4,564, TP2: 4,613, TP3: 4,664 Invalidation — Entry: A daily close below 4,317 breaks the flag and voids both scenarios Each scenario uses 0.06 lot with a 200 pip stop loss, risking $120. Together they still come to exactly 0.12 lot and $240, the same as every week. Volume splits into three parts of 0.01, 0.02 and 0.03, taking profit in stages at each target. For the breakout scenario specifically, a touch of 4,487 moves my stop loss to breakeven. The 4,564 target comes from the Fibonacci extension of this very rally, not a number I pulled out of the air. If price breaks higher before it ever eases back to 4,368, the first scenario simply doesn't exist. No chasing, no averaging in. Last week the rule cost me a good trade. This week I am keeping that same rule, because what has kept me alive for nine years was never the trades I called right. And you, do you think a 5.31% yield is what will sink gold, or what will lift it? --- P/S: Don't forget to leave a Like and ask anything you'd like to discuss to trade better every day! Follow tohaitrieu on TradingView to stay connected for the long haul, and talk through each trade together. I believe sharing and discussion help us learn more, and make every analysis and every comment on TradingView more useful for everyone ⚡️