How to trade gold (XAUUSD) with multi-timeframe analysis
On gold, the same chart can tell two opposite stories depending on the timeframe. On H1, price has been falling for three hours; on the daily chart, it is in a strong uptrend. Most beginners look at a single chart, and end up selling a simple pullback in the middle of a larger rally.
Multi-timeframe analysis solves this problem: you read the market from the top down, and only take a position when several timeframes agree.
The principle: each timeframe has one job
The simplest version uses three timeframes, each answering one specific question:
| Timeframe | Role | The question it answers |
|---|---|---|
| D1 (daily) | Direction | Which way should I trade? |
| H4 (4 hours) | Context | Is price pulling back against that trend? |
| H1 (1 hour) | Timing | Is price resuming the trend right now? |
Until all three answers are "yes", you do nothing. That is what makes the method more selective: it filters out a large share of false starts.
Step 1: the underlying trend on D1
Start with the daily chart. A simple, objective way to define the trend is to compare two exponential moving averages, for example an EMA 20 and an EMA 100:
- if the EMA 20 is above the EMA 100 and the close is above the EMA 100, the trend is up: you only look for buys;
- if it is the other way around, the trend is down: you only look for sells;
- otherwise, the market is undecided: you wait.
The key point: use the last closed daily candle, never the one in progress. Otherwise, the signal can flip several times during the day.
Step 2: wait for the pullback on H4
Buying an uptrend, yes. But buying it after a big rally often means buying the top right before a correction.
So you wait for a pullback: a moment when the market takes a breather against the trend. The RSI is a good tool to spot it:
- in an uptrend, wait for the H4 RSI to drop below 45;
- in a downtrend, wait for it to rise above 55.
These are not the classic extreme "oversold" 30/70 levels. You are looking for a simple pullback within a trend, not a reversal. The pullback stays valid for a few H4 candles, for example 9, which is a little over a day and a half.
Step 3: the trigger on H1
A pullback can also be the start of a real reversal. So you wait for proof that the trend is resuming: the first H1 close above the high of the previous 20 candles for a buy, or below their low for a sell.
Only then do you enter. At that point you have:
- the underlying trend on your side;
- a better entry price thanks to the pullback;
- confirmation that the move is resuming.
A concrete example
Imagine gold in an uptrend on the daily chart. On H4, the RSI drops to 43 after two days of decline: that is the pullback. A few hours later, an H1 candle closes above the highs of the last 20 hours: that is the buy signal.
The stop-loss is placed at a distance based on volatility (see our article on setting a stop-loss with the ATR), and the take-profit is set as a multiple of that risk.
Pitfalls to avoid
- Mixing up the roles. If H1 contradicts the daily trend, that is not a sell signal: it is often exactly the pullback you are waiting for.
- Using unclosed candles. On TradingView, a poorly coded indicator can "repaint" its signals: they appear, then disappear. All higher-timeframe data must come from candles that have already closed.
- Too many filters. Every extra condition reduces the number of signals. Three timeframes are enough; beyond that, you risk almost never trading.
- Forgetting costs. On gold, the spread and overnight fees (swap) matter. A method must stay profitable once those costs are included.
What six years of testing show
We tested this method on gold from January 2020 to September 2026, risking 1% per trade: a profit factor of 1.30 and 53% winning positions, but also two losing years. All the details are in our article on the backtest.
This is exactly the logic of the Hybrid Or indicator: it runs these three checks automatically and displays an arrow, with the stop-loss and take-profit levels, only when everything lines up.