Support and Resistance on XAUUSD: Draw Levels Right

Support and resistance on XAUUSD are the prices where gold has already changed hands with enough force to leave a mark. If you are asking how to draw them, the short answer is: mark zones from the daily and four-hour charts first, add the session highs and lows that traders actually defend, and keep a level only when price has reacted there with quality. A line through a single wick is a guess. A zone that has rejected price, held a close, and still sits where stops are likely to rest is a level you can plan around.

A hairline "level today" rarely survives the first London impulse. This guide is the method, not a forecast. Prices in the examples are illustrations, not calls.

Why gold respects some levels and ignores others

A level is a memory of orders. Buyers who missed a rally, and sellers who were trapped, often use the same shelf on a return. Neither group is guaranteed to be there. The level holds only while those orders, or new ones with the same idea, are still large relative to the flow hitting them.

Gold ignores a level for ordinary reasons. It was drawn on a timeframe nobody else is watching, it was a wick in a thin hour, a headline overwhelmed the resting orders, or the shelf was already used up. The first retest often carries the reaction. The third retest, with overlapping wicks and no push away, is a level being consumed.

There is no universal hit rate for "gold respects support." Anyone quoting one is inventing a statistic. What you can do is define a level in writing, log every touch in a gold trading journal, and learn how your definition behaves in Asia versus the London–New York window. That sample belongs to you. It is the only one that matches your rules.

Zones vs lines

Draw a zone, not a line. On XAUUSD the practical zone is the price band that contains the rejection, not the single print at the extreme.

A workable routine:

1. Find the swing high or swing low that actually turned the market on the timeframe you trust.
2. Mark the furthest wick and the cluster of closes or bodies that rejected from that area.
3. Shade the band between them. That band is the level.
4. Extend it to the right. Do not keep extending levels that price has closed through and never looked back at.

Example, not a signal: suppose the daily chart turns down from a cluster whose wicks reach 2,418.40 and whose bodies stall near 2,412.10. The zone is that band. A sell plan that needs a touch of exactly 2,418.40 will miss most of the rotations that still respected the shelf. A plan that says "if price trades into 2,412–2,418 and a four-hour candle rejects, I will look for a short on the 15-minute" is testable.

Named prices still help. Use the prior day high, prior day low, or session extreme as the anchor, and the zone as the band around it. A stop "a few pips" beyond that band means different dollars depending on whether your broker calls 0.01 or 0.10 a pip — see pip value on XAUUSD. Spread widens at rollover and around some data, so a line that ignores spread is tighter than the fill you will get.

When a zone is broken, decide in advance what "broken" means. A wick through the band is not automatically a break. Many gold traders require a candle close beyond the zone on the timeframe that defined it. Write that rule down before the candle prints. Changing it while the candle is open is how levels multiply until the chart is unreadable.

HTF levels first (D1/H4)

Higher-timeframe levels first is the whole edge of this method. The daily chart tells you which shelves the swing traders and the larger risk can see. The four-hour chart tells you how the current leg is negotiating those shelves. The five-minute chart is where you time an entry, not where you discover the map.

Open the daily chart and mark only the obvious swings: the last major high, the last major low, and any shelf that produced a multi-day reaction. If you cannot explain a level in one sentence ("this is the May swing high that rejected twice"), do not mark it. Then drop to the four-hour chart and mark the swings that built the current leg, especially the origin of the last impulsive move. Those four-hour zones are where an order block often sits. The block is not a substitute for the level. It is a refined location inside or next to it.

Only then open the 15-minute chart, looking for price to interact with a level you already marked. A five-minute pattern far from daily and four-hour structure is a pattern without a location.

A simple top-down pass, before London and again before New York if you trade both:

• What is the daily bias, defined as higher highs and higher lows, or the opposite? If the daily is overlapping and flat, say "range" instead of forcing a trend.
• Which daily zone is closest above and below price?
• Which four-hour zone is price approaching?
• Has that zone already been tested in this swing?

If you cannot answer those four questions, you do not have a level. You have a chart. TradYX AI Market Analysis marks support and resistance on XAUUSD so this pass starts from a proposal instead of a blank page. The proposal is still a proposal. Compare it with the swings you would have marked yourself, and throw out anything that does not survive the "one sentence" test.

Session highs/lows as levels

Gold trades almost around the clock from the Sunday open to the Friday close, following the forex week rather than a single stock-exchange bell. Inside that week, the session extremes are levels because that is where intraday traders put stops.

Mark these as named prices each day:

• Prior day high and prior day low.
• Asia high and Asia low, using one window you keep fixed. An example window is 00:00–06:00 UTC, which lines up with the Tokyo cash session. It is a convention, not an exchange rule. If you change the window, you change the level, so do not change it mid-week.
• London session extreme once London has made its first push.
• New York session extreme, especially after the US data times that often fall in the morning.

A sweep of the Asia high that returns inside the range is a different event from a close beyond it that holds. The second is the start of the Asian range breakout, and it only makes sense if the high and low were marked before London. A daily shelf that sits on yesterday's high is a stronger location than an Asia high floating in the middle of a daily range. Write the overlap into the plan.

The best time to trade gold also changes how much you should trust a fresh session level. An Asia high printed in a narrow, quiet range is a boundary. The same style of high printed in a news spike is a wick that may never trade again. Note the session and whether a scheduled release was in the window. What moves the gold price is the checklist for those releases: yields, the dollar, and the data calendar, not a new line for every headline.

Validating levels with reaction quality

A level is a hypothesis until price reacts. Reaction quality is how you keep the hypothesis honest.

Look for four signs, and require the ones you write into your rules. You do not need all four every time, but you do need a definition you will not edit after the fact.

Departure. Price left the zone and did not linger. On gold, a useful departure is a push that breaks a nearby minor swing, not a three-tick bounce that comes back inside the zone on the next candle.

Rejection shape. Wicks into the zone with closes back outside tell you the zone was offered or bid. A full-body close through the middle of the zone tells you it was not.

Follow-through on the retest. The first return to the zone either holds or it does not. Measure this on the timeframe you trade. A five-minute pin bar against a daily zone can be noise. A four-hour rejection from a daily zone is the reaction the level was drawn for.

Failure. If price closes through the zone and the old level starts acting the other way — former support behaving as resistance, or the reverse — update the map. Leaving a broken level in its old color is how traders fade a trend that has already changed.

Write the validation rule before the session. Example: a daily zone stays active after a four-hour rejection, and it is inactive after a daily close beyond the far side. The distance you require is yours to set from your own chart.

Log the touch even when you skip the trade. A row with the level name, the session, and the result in R shows which shelves deserve attention. Without that log, "this level always works" is just the last winner remembered more clearly than the failures.

AI-marked levels vs manual

Manual levels are slow, and you tend to mark what confirms the trade you already want. AI-marked levels are faster and will also propose zones that do not fit your strategy. Use them as a second pass, not as permission. A practical split:

• You mark the daily and four-hour swings yourself, or you review the ones the tool marked and delete anything you cannot explain.
• You let the tool highlight session highs, lows, and nearby shelves so you do not miss an obvious prior-day level on a rushed morning.
• You still require your own trigger: a close, a retest, a sweep, whatever your written rule is. The level is the location. It is not the entry.

TradYX is built only for XAUUSD, and the AI analysis proposes support and resistance in that market rather than a generic oscillator on fifty symbols. That focus does not make a zone correct. It means the suggestion is at least about gold. Compare the proposal with the reaction-quality rules above. If the marked zone has never produced a departure you respect, leave it off the trade plan even if it is neatly labeled.

Screenshot the levels before the session so you cannot nudge them into the trade you already took. Then review both the levels that failed and the ones you skipped. That review is the skill. The software only stops a blank chart from eating the first hour of the day.