Séance d'exemple · 2026-10-09
Three things dominate this week's gold headlines.
- The Federal Reserve: officials have signalled that more rate cuts are possible but depend on the data. Markets care because lower rates usually mean lower real yields (see the fundamentals report).
- US inflation: the CPI report is due next Wednesday. In the past, a hotter number than expected has pushed yields and the dollar up and gold down within minutes. A softer one has done the opposite.
- Geopolitics: ongoing tensions keep some safe-haven demand in place. History shows this kind of demand can fade quickly when headlines calm down.
Also on the calendar: the US jobs report (NFP) at the start of next month and the minutes of the last Fed meeting. Scheduled events like these cause the biggest moves, so a learner should always know what is due before the week starts.
Gold's daily chart shows an uptrend that has lost some speed. The price is still above its 200-day moving average (the average closing price of roughly the last ten months, a common way to judge the long-term trend), and the 50-day average sits above the 200-day, which traders call a rising structure.
Momentum is cooling. RSI, a 0-100 momentum gauge, has eased from above 70 to about 57. That says the rally is less stretched, not that it has ended. MACD, which compares a fast and a slow average, is still positive, but its histogram is shrinking.
Volatility is elevated. ATR, the average daily range, is near 45 dollars, so an ordinary day can move about 1%. Beginners often underestimate how far that is.
In the example data, buyers stepped in twice near 3,980 in late September, while sellers appeared near the recent high around 4,160. The round number 4,000 sits in between and has been crossed several times.
What indicators cannot tell you: why the price moves. That is the next analysts' job.
The mood around gold is optimistic, bordering on crowded. Headlines this week use words like "record" and "safe haven", and social posts lean heavily one way: most of the chatter in the example data expects higher prices.
Why that matters for a learner: when nearly everyone already agrees, many of the people who wanted gold may already own it. That leaves fewer new buyers, and a surprise in the other direction can trigger fast moves as crowded positions unwind. Contrarian traders treat extreme agreement as a warning sign rather than a confirmation.
Sentiment is not a timing tool. Crowds can stay one-sided for weeks. Use it as a question: "Who is left to change their mind?"
Gold pays no interest, so its biggest slow-moving driver is the real yield: the interest rate on US government bonds after inflation. When real yields fall, owning gold costs less in missed interest, and demand has historically improved. In this example, the 10-year real yield has drifted lower over the past two months as markets priced in further Federal Reserve rate cuts.
The US dollar is the second driver. Gold is priced in dollars, so a weaker dollar makes it cheaper for people using other currencies. The dollar index has been flat to slightly weaker.
Third, central banks. Their gold purchases have been strong for several years, led by emerging-market central banks diversifying their reserves. This is slow, steady demand that does not react to daily news.
Pulling the other way: if inflation data surprises to the upside, the Fed could slow its cuts, real yields could rise, and that support would weaken. Fundamentals explain the backdrop. They rarely time anything.
Let me make the bullish case. The trend is intact: price is above both the 50-day and 200-day averages, and the cooling RSI looks like a pause, not exhaustion. The macro backdrop supports it: real yields are drifting lower, the dollar isn't strong, and central banks keep adding gold steadily. The buyers who appeared twice near 3,980 show there is demand on dips.
In this scenario, a soft CPI print next week confirms the rate-cut path, real yields slip further, and gold returns to the area of the recent high near 4,160.
What would prove me wrong: a hot inflation report that pushes real yields up, or a daily close below the 3,980 area where buyers stepped in.
I see the same chart differently. Momentum is fading while the price is still near its highs, and a shrinking MACD histogram is a classic early sign of a tiring move. The sentiment report says the crowd is already one-sided, so who is left to push it higher?
The bullish case also leans on one assumption: that inflation keeps cooling. If next week's CPI is hotter than expected, rate cuts get priced out, real yields rise, the dollar firms, and gold loses its main support. Safe-haven demand can also evaporate when the headlines calm.
In this scenario, the 4,000 round number gives way and the price returns to the 3,980 area or lower.
What would prove me wrong: a soft CPI and a decisive daily close above the 4,160 high.
From a high-risk-tolerance view, uncertainty is where opportunity lives: a big scheduled event like CPI can create sharp moves in either direction. But let's be honest about the price of that mindset. A 1% daily range becomes a 10% swing in an account using 10:1 leverage, in a single day. Gaps around news can jump past any exit you planned. The aggressive traders who survive treat every idea as possibly wrong and decide their maximum loss before anything else.
The conservative lens starts with what could go wrong. Gold is near its highs, the crowd is one-sided, and a major data release is days away: three reasons volatility could spike. Event risk cuts both ways, so even a correct reading of the backdrop can be overwhelmed by one number. Liquidity can thin out around the release, and prices can jump. From this view, waiting until after CPI for clearer information is a fully legitimate choice. Not acting is also a decision, and often the cheapest one.
Both sides make fair points, and that is the lesson: the evidence is genuinely mixed. Known: the long-term trend is up and the macro backdrop has been supportive. Uncertain: whether inflation keeps cooling, and how crowded the consensus really is. Time horizon changes everything. Someone thinking in years cares about central-bank demand; someone thinking in hours cares about next Wednesday at 8:30 New York time. The single piece of information most likely to change the balance this week is the CPI report.
Contenu éducatif, pas un conseil en investissement.