XAUUSD (Gold) Trading Master Guide

Prerequisites: All three prior guides. Gold is not a beginner instrument, and this page assumes you can already calculate a position size from a stop distance.

Key Takeaways

• Gold is not a currency pair. XAUUSD looks like one, is quoted like one, and behaves nothing like one. Treating it as "just another pair" is the root error.
• Real yields are the master variable. Gold pays no interest. When inflation-adjusted bond yields rise, the opportunity cost of holding a non-yielding asset rises, and gold falls. Watch the 10-year TIPS yield, not the CPI headline.
• Inflation does not reliably lift gold. Only inflation relative to nominal rates does. Gold has fallen through inflationary periods when central banks hiked aggressively. This surprises almost everyone.
• The dollar relationship is mechanical, not mystical. Gold is priced in dollars. A stronger dollar makes gold more expensive for everyone else, suppressing demand. DXY up, gold down — usually, and with important exceptions during crises when both rise together.
• Gold's daily range is routinely 2–3× EUR/USD's in percentage terms. Every position sizing error costs you multiples of what the same error costs on a major pair.
• Contract specifications differ from FX, and "pip" is ambiguous on gold. Check your broker's spec sheet before your first trade. This single omission has cost more beginners more money on XAUUSD than any analytical mistake.
• Gold sweeps liquidity more violently than any major pair. False breakouts are its native behaviour, not an anomaly.
• We do not publish gold price predictions. Anyone who does is either guessing or selling. What we can teach is the causal framework that lets you interpret what happens.

Quick Summary

Gold is the most traded speculative instrument among retail traders and the least understood.

It attracts beginners because it moves. It destroys beginners for the same reason.

This guide is built in three layers.

Layer 1 — The fundamentals that actually drive it. Real yields, the Fed, the dollar, inflation, safe-haven demand, central bank buying. This is the layer nobody teaches, and it is why most gold traders are technically competent and directionally confused.

Layer 2 — The mechanics and the calendar. Contract specifications, pip ambiguity, position sizing for gold, spreads, sessions, and the four events that dominate its year: FOMC, CPI, NFP, and the quarterly projections.

Layer 3 — The technical execution. Market structure on gold, its violent liquidity behaviour, order blocks and fair value gaps assessed honestly, session-based approaches, scalping, swing trading, and three complete strategy specifications.

Plus: six annotated trade examples (three losses), a professional checklist, common mistakes, a glossary, and an honest answer to the question "where is gold going?"

Trading gold on margin carries a high risk of loss. Nothing here is financial advice, and nothing here is a price forecast.

Part I — What Gold Actually Is

Why Gold Is Not a Currency Pair

Your platform lists XAUUSD alongside EUR/USD and GBP/USD. It is quoted the same way. It has a bid and an ask. It looks like a peer.

Full guide: Trading gold in the US: what is actually allowed

Full guide: What is XAUUSD? How gold trading actually works

It is not.

EUR/USD is a relationship between two economies. Its price expresses the relative expected path of interest rates, growth and risk between the eurozone and the United States. Both sides have a central bank. Both sides have yields. It is a spread between two yielding assets.

XAUUSD is a relationship between a non-yielding asset and the world's reserve currency. Gold has no central bank, no earnings, no coupon, no cash flow, and no intrinsic yield. It sits in a vault and does nothing.

That single property — gold pays you nothing — determines almost everything about how it trades.

The consequence

If gold pays nothing, then the cost of owning it is whatever you could have earned by holding something else that is equally safe.

The obvious alternative is a US Treasury bond. Safe, liquid, and it pays interest.

So the question every large gold holder implicitly asks, every day, is:

"What am I giving up by holding this metal instead of a Treasury?"

The answer to that question is the real yield. And that is the whole ballgame.

What this means for you, practically

Traders who approach gold as "a pair that moves a lot" trade its volatility without any model of its direction. They apply a EUR/USD framework to an instrument with entirely different machinery, and then they are baffled when gold rips 2% higher on a day when the dollar was also higher — an event that would be near-impossible for a normal dollar pair.

Understanding gold's causal structure will not make you profitable. Nothing on this page will. But it will stop you being surprised, and being surprised is expensive.

What Moves the Price of Gold?

Real interest rates, primarily. Gold pays no yield, so when inflation-adjusted bond yields rise, the opportunity cost of holding it rises and it falls. Watch the 10-year TIPS yield published by the U.S. Treasury. Secondary drivers are the US dollar, safe-haven demand during crises, and central bank buying.

Full guide: What moves the price of gold: the full hierarchy

If you learn one thing from this guide, learn this.

Real yield  ≈  Nominal yield  −  Expected inflation

The market gives you this number directly. You do not have to compute it. US Treasury Inflation-Protected Securities (TIPS) trade on real yields. The 10-year TIPS yield is published daily by the U.S. Treasury.

The mechanism, stated plainly

Gold pays 0%. Always. Forever.

• 10-year real yield is +2.0%. Holding gold costs you 2% per year in forgone, inflation-protected, risk-free return. Gold is expensive to hold. Institutional money rotates out. Gold falls.
• 10-year real yield is −1.0%. A Treasury guarantees you lose 1% of purchasing power per year. Gold's 0% is now the better deal. Gold is cheap to hold. Money rotates in. Gold rises.

That is the entire relationship. It is not a correlation someone noticed. It is a causal chain with an economic mechanism, which is why it has held for decades and why it is the only gold relationship worth building a framework on.

Gold and real yields are strongly inversely correlated. Not perfectly, and not every week — but structurally, and over any horizon that matters.

What this means for interpreting the news

This is where it becomes useful.

A hot CPI print arrives. Inflation is higher than expected. Your instinct — and every "gold is an inflation hedge" article you have read — says gold should rally.

Watch what usually happens instead.

Higher inflation → the market prices more Fed hikes → nominal yields jump → nominal yields rise more than inflation expectations → real yields rise → gold falls.

Gold frequently drops on high inflation prints. Traders who bought "the inflation hedge" are stopped out within four minutes, wondering what happened.

The reverse also holds. A soft CPI print → the market prices cuts → nominal yields collapse → real yields fall → gold rallies on low inflation.

The honest caveats

The relationship is structural, not mechanical. It has decoupled for months at a time, most notably during periods of aggressive central bank buying or acute geopolitical stress, when safe-haven demand overwhelms the opportunity-cost calculation.

Do not trade the real yield as a signal. Trade it as a bias filter. It tells you which direction has the wind behind it, not when to click.

The Federal Reserve

Once you understand real yields, the Fed's importance becomes obvious rather than mysterious.

Full guide: How the Federal Reserve moves the gold price

The Federal Reserve controls the short end of the yield curve directly and influences the long end through expectations. The Fed does not move gold. The Fed moves real yields, and real yields move gold.

What actually matters

Not the rate decision itself. By the time the Fed hikes, the hike has been priced for weeks. Markets trade the future, not the present.

What matters is the surprise — the gap between what the Fed does or says and what the market already expected.

Three things generate that surprise:

1. The statement's language. A single changed word — "some further tightening" becoming "further tightening" — repositions the entire curve.

2. The Summary of Economic Projections (the "dot plot"), published quarterly. Each official's expectation for future rates. A shift in the median dot moves real yields immediately and violently.

3. The press conference, thirty minutes after the statement. The Chair speaks unscripted. Gold routinely does one thing at 19:00 UTC on the statement and the exact opposite at 19:35 UTC on the press conference.

Hawkish and dovish, defined for gold

A single sentence to carry with you: hawkish Fed, gold down. Dovish Fed, gold up. Because of real yields. Always because of real yields.

The Dollar and DXY

Gold is priced in US dollars. This creates a mechanical relationship that requires no theory at all.

Full guide: The dollar index (DXY) and its link to gold

The mechanism

An ounce of gold costs, say, $2,000.

For a European buyer, the cost in euros depends on EUR/USD. If the dollar strengthens, that same ounce costs more euros. Demand from every non-dollar buyer — which is most of the world — falls.

Stronger dollar → gold more expensive globally → demand falls → price falls.

This is arithmetic, not economics.

The Dollar Index

DXY measures the dollar against a basket: euro (≈57.6%), yen, sterling, Canadian dollar, Swedish krona, Swiss franc.

Note the weighting. DXY is, to a first approximation, an inverted EUR/USD chart. It is dominated by the euro. It tells you far less about the dollar's strength against the yen or emerging market currencies than its name implies.

Gold and DXY are inversely correlated, typically strongly. But:

When the relationship breaks — and this matters

During acute crises, the dollar and gold both rise.

Why? Because in a genuine panic, capital seeks safety, and both the dollar and gold are safe-haven assets. The dollar is the world's reserve currency and the deepest liquidity pool on earth. Gold is the asset with no counterparty.

Fear buys both.

A trader who is short gold because "DXY is up" during a geopolitical shock is fighting a flow they have not identified. The correlation held for months, and it broke on the day it cost the most.

Note also: if you are long gold and short USD/JPY, you hold one position, twice. Both are short-dollar expressions. See correlation-adjusted sizing in the risk masterclass.

Inflation — The Relationship Everyone Gets Wrong

"Gold is an inflation hedge."

Full guide: Real yields and gold: the relationship that matters

This is the most repeated and least examined claim in gold commentary. It is, at best, badly incomplete.

The long run versus the trade

Over decades, gold has broadly preserved purchasing power. That is the sense in which it is an inflation hedge, and it is a statement about a fifty-year horizon.

Over months and quarters — the horizon on which you trade — the relationship between gold and inflation is unreliable and frequently inverted.

Why

Return to the master variable.

Real yield ≈ Nominal yield − Expected inflation

Inflation rises. What happens to gold depends entirely on what nominal yields do in response.

Scenario A — Inflation rises, central bank hikes aggressively. Nominal yields rise faster than inflation expectations → real yields rise → gold falls.

Scenario B — Inflation rises, central bank does nothing. Nominal yields stay flat → inflation expectations rise → real yields fall → gold rises.

Same inflation. Opposite outcomes. The variable is the central bank's response, not the inflation number.

Gold has fallen through genuinely inflationary periods when policymakers responded forcefully. It has rallied hard during low-inflation periods when policymakers cut rates toward zero.

Trading the CPI release

Do not. Or, if you must, understand what you are trading.

CPI is released by the U.S. Bureau of Labor Statistics at 13:30 UTC (08:30 ET, adjusted for daylight saving).

In the seconds around it:

• Spreads on gold widen from cents to several dollars.
• Slippage on stops is severe. Your stop fills where the market next prints, not where you set it.
• Price frequently moves violently in one direction, then fully reverses within minutes.

The first move is algorithmic. It is not analysis; it is machines reading the number and repricing. Retail traders participating in that thirty-second window are providing liquidity to it.

The considered move arrives ten to forty minutes later, once the market has digested what it implies for the Fed's path. That move is more tradeable, on wider stops, at normal spreads.

Safe Haven Demand and Central Bank Buying

The two drivers that override everything else, and the two you cannot model.

Full guide: Is gold a safe haven? What crises actually show

Safe haven flows

In a genuine crisis — war, banking failure, sovereign default — capital moves to assets with no counterparty risk. Gold's defining feature is that it is nobody's liability. A bond is a promise. A deposit is a promise. Gold is a metal.

During such episodes, the real yield relationship, the DXY relationship, and every technical level you have drawn become temporarily irrelevant. Gold gaps. It runs. It ignores resistance that held for two years.

You cannot predict these events. You can only avoid being positioned catastrophically wrong when one occurs, which means: position size, and do not hold large gold positions through weekends during periods of geopolitical tension. Gold gaps on Sunday's open, and a gap fills your stop at the next available price, not at your price.

Central bank buying

Central banks — particularly in emerging markets — have been substantial net buyers of gold in recent years, motivated by reserve diversification away from dollar assets.

This is price-insensitive, structural, persistent demand. These buyers are not trading. They are accumulating according to a policy mandate, and they do not care about your resistance level.

Its effect is to provide a floor beneath dips and to explain periods when gold rises despite rising real yields — decouplings that baffle traders who have only one model.

Track it via the World Gold Council's quarterly demand data. It will not time a trade. It will tell you which way the structural wind is blowing over quarters, and it explains most of the periods when the real-yield model appears to break.

The Hierarchy of Drivers

Not everything matters equally. Rank them.

Note where technical analysis sits.

This is not a demotion of technicals. Technicals tell you where to enter, where you are wrong, and how to size. They are your execution layer, and without them you have no trade.

But when a driver above them moves, the levels below them evaporate. This is exactly the point made in the technical analysis guide: a level is a statement about orders resting under current assumptions. Change the assumptions, and the orders are pulled. The level does not fail. It ceases to exist.

Gold makes this lesson expensive, because its drivers move violently.

On Gold Price Predictions

You may have arrived here searching for one. We are not going to give you one, and this section explains why — because the reasoning is more useful than any number would be.

Why nobody can forecast gold's price

Gold's price is a function of the future path of real interest rates, which is a function of the future path of inflation and central bank policy, which is a function of future economic data, which is a function of events that have not happened.

Anyone forecasting gold's price in twelve months is forecasting the Federal Reserve's reaction function to data that does not yet exist.

Institutions publish gold targets. Read their track records. The dispersion between banks' year-ahead forecasts is routinely wider than the range gold actually trades, and revisions are frequent and large. These are not fools. They are extremely capable people demonstrating that the task is not tractable.

What a forecast is actually for

A published gold target is, almost always, one of three things:

1. A marketing asset. It generates headlines and inbound links.
2. A conditional scenario, stripped of its conditions. The analyst wrote "if real yields fall 80bp and the Fed cuts three times, gold reaches X." The headline drops everything before the comma.
3. A guess with a decimal point.

What to do instead

Replace the question "where is gold going?" with:

"What conditions would make gold rise, and are those conditions present or developing?"

That question is answerable.

Conditions favouring gold: falling real yields · a Fed pivoting dovish · a weakening dollar driven by rate differentials · escalating geopolitical stress · sustained central bank accumulation.

Conditions pressuring gold: rising real yields · a hawkish Fed · a strengthening dollar on rate differentials · risk-on equity strength · rising bond yields with contained inflation expectations.

You do not need to know where gold will be. You need to know which way the wind is blowing, take setups aligned with it, and size so that being wrong costs 1%.

That is the entire professional edge, and it does not require a crystal ball.

Part II — Mechanics

Contract Specifications and the Pip Problem

This section has cost more beginners more money on gold than any analytical error on this page. Read it slowly.

Full guide: Gold futures vs ETFs vs miners vs physical: which does what

The standard specification

For most brokers offering XAUUSD as a CFD:

• 1 standard lot (1.00) = 100 troy ounces
• A $1.00 move in gold's price = $100 profit or loss per standard lot
• A $0.10 move = $10 per standard lot
• A $0.01 move = $1 per standard lot

The pip ambiguity — the actual trap

There is no universal definition of a "pip" on gold.

• Some brokers call $0.01 one pip.
• Some call $0.10 one pip.
• Some display prices to two decimals, some to three.
• Some platforms label the smallest increment a "point" and refuse to use "pip" at all.

A trader who assumes gold's pip works like EUR/USD's, and who reads "a 200-pip stop" from a forum post, may be sizing for a $2.00 stop or a $20.00 stop. Those differ by a factor of ten.

At 0.10 lots, that is the difference between risking $20 and risking $200.

Other specifications to verify

• Minimum lot size. Some brokers' minimum on gold is 0.01 lots. Some enforce 0.10. If yours enforces 0.10, a small account cannot risk 1% properly on gold, and you should not trade it there.
• Margin requirement. Gold's leverage is frequently lower than FX majors — 20:1 rather than 30:1 under EU rules — and brokers often raise margin requirements before major events.
• Trading hours. Gold typically closes for roughly an hour each day (around 21:00–22:00 UTC) and does not trade over the weekend. It gaps on Sunday's open.
• Swap. Gold's overnight financing is frequently negative on both sides, and materially larger than FX. Holding gold for weeks has a real, compounding carrying cost that must be modelled into any swing strategy.

How Do You Calculate Lot Size for Gold?

Lots equal your risk amount divided by stop distance in dollars multiplied by 100, assuming a 100-ounce standard contract. On a $10,000 account risking 1% with a $10.60 stop: $100 ÷ ($10.60 × 100) = 0.09 lots. Ignore the word pip on gold; brokers define it inconsistently.

The formula from the risk masterclass is unchanged. Only the pip value changes. But because gold's volatility is so much larger, the discipline matters more.

Lots = Risk$ ÷ (Stop distance in dollars × $100 per lot per $1.00 move)

Simplified for gold:

Lots = Risk$ ÷ (Stop in $ × 100)

Worked example 1 — intraday

$10,000 account. 1% risk = $100.

Entry at a support zone. Structural invalidation $5.00 below entry. Daily ATR(14) on gold = $28.00. Buffer at 0.2 × ATR = $5.60.

Total stop distance = $10.60.

Lots = $100 ÷ ($10.60 × 100)
     = $100 ÷ $1,060
     = 0.094  →  trade 0.09 lots

Actual risk: $10.60 × 100 × 0.09 = $95.40. Just under 1%.

Worked example 2 — swing

Same account, same 1% risk = $100. Swing setup. Structural invalidation $34.00 below entry. Buffer 0.3 × $28 = $8.40. Total stop = $42.40.

Lots = $100 ÷ ($42.40 × 100) = 0.023  →  trade 0.02 lots

Actual risk: $84.80.

The lesson, again

Stop distance moved from $10.60 to $42.40 — four times wider. Position size fell from 0.09 to 0.02 lots — roughly four times smaller.

The risk stayed at 1%.

Level → stop → size. On gold. Every time. No exceptions.

Spreads, Slippage and Costs

Typical spreads

Figures are indicative and vary enormously by broker and account type. Measure your own.

What this means for strategy design

At 0.10 lots, a $0.30 spread costs $3.00 on entry.

A scalper targeting a $2.00 move on gold is paying 15% of their gross target in spread before anything happens. At twenty trades a day, spread is the dominant term in their P&L, not their analysis.

Gold is a poor scalping instrument at retail spreads. This is an unpopular claim and it is arithmetic, not opinion. See scalping below.

Slippage

Gold slips. Aggressively. During news, stops fill dollars from where they were placed.

Model this. If your backtest assumes stops fill at your price, your backtest is optimistic by an amount you have not measured. Record actual fill prices in your journal and compute your average slippage in R. Most gold traders discover it costs them 0.05–0.15R per trade, which is enough to convert a marginal system into a losing one.

Swap

Gold's overnight financing is frequently negative on both long and short sides, and materially larger than FX. A 0.10-lot position held for three weeks can accumulate a swap cost that consumes a meaningful fraction of the target.

Check the swap rate on your broker's spec sheet before designing any swing strategy on gold. Then include it in the backtest.

What Is the Best Time to Trade Gold?

The London/New York overlap, 12:00–16:00 UTC, where liquidity peaks and spreads are tightest. The London open at 07:00–09:00 UTC produces the day's first expansion and frequently sweeps the Asian session range. Avoid the hour around rollover near 21:00 UTC, when spreads widen sharply.

Full guide: Gold trading sessions: when XAUUSD actually moves

Gold trades 23 hours a day, and it is four different instruments across them.

The Asian range play

Gold's most consistently observed intraday structure:

1. During the Asian session, gold establishes a comparatively narrow range.
2. At the London open, price sweeps one side of that range — taking the stops of everyone who traded the Asian boundary.
3. It then frequently reverses and travels toward the opposite side, and beyond.

This is a liquidity sweep, executed on schedule. It is the single most reliable structural observation about gold's day, and it is why gold traders who fade the London-open breakout outperform those who chase it.

Caution, stated clearly: "frequently" is not "always." This is a tendency, observable in the data, not a rule. Backtest it on your own data before trusting it, and note that it fails most often on days with 07:00–09:00 UTC economic releases from Europe.

The Four Events That Define Gold's Year

Everything else is background.

1. FOMC (eight times per year)

Statement: 19:00 UTC (18:00 during US winter time — check). Press conference: 19:30 UTC. Dot plot: quarterly (March, June, September, December).

The single largest scheduled driver of gold, because it is the single largest driver of real yields.

The pattern: an initial algorithmic move on the statement, frequently reversed during the press conference.

The professional response: flat before, watch, and consider trading the move that establishes after 20:00 UTC, on wide stops, at normal spreads.

2. CPI (monthly)

13:30 UTC. Bureau of Labor Statistics.

Remember the inversion. Hot CPI → hawkish repricing → real yields up → gold frequently falls. Do not trade the "inflation hedge" story. Trade the yield reaction.

Watch core CPI more than headline. The Fed does.

3. NFP (monthly, first Friday)

13:30 UTC. The Employment Situation report.

Strong jobs → economy robust → Fed can stay hawkish → real yields up → gold down.

The nuance that catches people: average hourly earnings frequently moves gold more than the headline jobs number, because wages feed inflation expectations, which feed rate expectations, which feed real yields.

A strong headline with soft wages is not a bearish gold print. Reading only the headline is how traders end up on the wrong side of a reversal within four minutes.

4. The dot plot / Summary of Economic Projections (quarterly)

Published with the March, June, September and December FOMC statements. Each official's projected rate path.

A shift in the median projection reprices the entire curve instantly. Its effect on gold is immediate and frequently larger than the rate decision.

The rule

Do not hold gold positions through FOMC, CPI or NFP unless your strategy is explicitly a news strategy and you have backtested it as one.

The spread widening alone converts a positive-expectancy setup into a negative one. Add slippage on stops, and the arithmetic is unambiguous.

Check the calendar every single morning. This is thirty seconds of work and it is the highest-value thirty seconds in gold trading.

Part III — Technical Execution

Market Structure on Gold

Everything in the technical analysis guide applies. Higher highs and higher lows, lower highs and lower lows, break of structure. The framework is universal.

Full guide: Liquidity and market structure, explained

Three gold-specific adjustments:

1. Everything is wider

Gold's daily range is routinely 2–3× EUR/USD's in percentage terms. Your zones must be wider. Your stops must be wider. Your position sizes must therefore be smaller.

A support "zone" on gold might be $8–15 wide, not the fraction of a pip a beginner instinctively draws. Use 0.5 × ATR(14) on your analysis timeframe, as always — and be prepared for that number to look shocking.

2. Gold overshoots

Levels are respected approximately. Gold routinely pierces a level by $3–6 before reversing.

A trader placing a stop $1 beyond an obvious swing low is not trading. They are donating to the sweep.

This is the single most important reason gold destroys traders who migrate from EUR/USD: the same stop discipline that works on majors is fatally tight on gold. They keep getting stopped out on trades that would have worked, conclude the market is rigged, widen their stops without reducing size, and blow up.

3. Higher timeframes are cleaner

Gold's noise on M1–M15 is severe. Its structure on H4 and daily is remarkably clean — arguably cleaner than most FX majors, because its fundamental driver (real yields) trends over weeks.

Analyse gold on H4 and daily. Execute on H1 or M15. Analysing gold on M5 is analysing noise, and the fact that the noise is $4 wide will convince you it is signal.

Liquidity Sweeps — Gold's Native Behaviour

On EUR/USD, a liquidity sweep is a notable event.

On gold, it is the default.

Why gold sweeps so aggressively

Retail concentration. Gold is the most popular retail instrument. More retail traders means more retail stops, clustered in the same obvious places.

Obvious levels. Round numbers ($2,000, $2,050, $2,100) attract enormous stop clustering. Everyone sees them.

Volatility. A $5 sweep is trivially cheap for an institution to execute on an instrument with a $28 daily range. On EUR/USD, an equivalent-percentage sweep would be a major move.

Two-way liquidity. Gold is deeply liquid at the institutional level, which means a large seller can reach for a stop pool without slipping badly, fill their size, and let price return.

The structure

Identical to the general case in the technical analysis guide, executed more violently:

1. The push — price extends beyond an obvious high or low, taking stops.
2. The failure — price closes back inside. The close is the signal. The wick is not.
3. The move — trapped traders exit, adding fuel to the reversal.

Where gold sweeps most reliably

• The Asian session high and low, at the London open.
• The prior day's high and low.
• Round numbers — the $50 and $100 increments especially.
• The prior week's high and low, on Monday.
• Obvious swing points that have been touched twice and are visible on every chart.

Trading it

Do not place your stop one dollar beyond the obvious level. That is the pool.

Do place it 0.3–0.5 × ATR beyond, accept the smaller position size, and stop funding other people's entries.

Or trade the sweep itself: wait for the push beyond the level, wait for the close back inside, enter against the trapped crowd with a stop above the sweep wick.

That stop is unusually precise — if price reclaims the swept high, the thesis is simply wrong — which makes it unusually tight, which produces unusually high reward-to-risk. This is where gold's R:R actually comes from: not from ambitious targets, but from setups where "wrong" is nearby and knowable.

Order Blocks and Fair Value Gaps, Honestly Assessed

These concepts dominate gold trading content. They deserve a fair hearing and an honest one. This section is labelled opinion where it is opinion.

Full guide: Order blocks and fair value gaps, assessed honestly

What they claim to be

Order block. The last opposing candle before a strong impulsive move. The theory: institutions accumulated a position there, and unfilled orders remain, so price returning to that zone finds buyers (or sellers) again.

Fair value gap (FVG) / imbalance. A three-candle pattern where the first candle's wick and the third candle's wick do not overlap, leaving a "gap" that price traversed too quickly for orders to fill on both sides. The theory: price is drawn back to "rebalance" it.

What they actually are

Order blocks are supply and demand zones with a specific identification rule.

That is not a criticism. It is a genuine improvement over "draw a box where price bounced," because it is mechanically definable — the last down-candle before the impulse — which means it is backtestable. Recall from the technical analysis guide that mechanical definability is the property that separates a strategy from an anecdote.

Fair value gaps are volatility artefacts. They mark where price moved fast. That is all they are, definitionally. The claim that price is "drawn back to fill them" is a claim about mean reversion after volatility expansion — which is a real, documented phenomenon in some conditions, and not a special property of the three-candle pattern.

Our honest position

In favour: - Both are mechanically definable and therefore testable, which puts them ahead of most retail chart patterns. - Both identify zones where price moved impulsively, and impulsive moves genuinely indicate the presence of large orders. - Enough traders now watch them that they carry the same reflexive power as the 200 EMA. This is circular, and it is real.

Against: - The institutional narrative is unfalsifiable. No one can see the institutional order book in an OTC market. "Institutions left orders here" is a story attached to a pattern, and the story is doing no analytical work. - Selection bias is severe. Look at any chart and you will find order blocks price respected. The ones it ignored are invisible in hindsight. Backtest them prospectively, bar by bar, before believing them. - FVGs "fill" frequently because price fills most nearby areas eventually. The relevant question is whether price fills them more often than it fills an arbitrary equivalent zone, and over what horizon. Most proponents have never asked this question in a form that could return "no." - The terminology creates an illusion of precision and institutional insight where none exists.

How to actually use them

Treat an order block as a supply/demand zone with a strict identification rule.

Then subject it to the identical requirements you apply to any other level, from the technical analysis guide:

• Does it align with higher-timeframe structure? Longs at demand only in a daily uptrend.
• Is it untested? A zone touched three times is a zone about to break — each touch consumes the resting orders.
• Is there a trigger on arrival, or are you entering on hope?
• Is invalidation precise?
• Is the calendar clear?
• Are you sized so that being wrong costs 1%?

If you apply that filter, order blocks work approximately as well as any other well-identified supply/demand zone — which is to say: usefully, as a location, within a complete system.

Kill Zones — An Honest Assessment

"Kill zones" designate specific windows in which trades are permitted — commonly the London open (07:00–10:00 UTC) and the New York open (12:00–15:00 UTC).

What is defensible

Volatility and liquidity genuinely concentrate in these windows. That is a measurable fact, not a theory. Spreads are tightest during the overlap. Real institutional order flow arrives at the London and New York opens. Breakouts occurring in these windows have participation behind them; breakouts at 03:00 UTC generally do not.

A session filter is one of the cheapest, most robust improvements available to any strategy. The risk masterclass makes this point in general terms: track expectancy by session, delete the negative ones. It requires no new skill, only data you already have.

What is not defensible

The mystique. The branding, the proprietary terminology, the implication that these windows contain a secret known to a select few.

They are the London and New York sessions, which have been documented in every market microstructure text for forty years, renamed.

The honest position

Use a session filter. Trade gold during the London session and the overlap. Avoid the Asian session unless your strategy is specifically a range strategy designed for it. Avoid the hour around rollover.

Do not pay for this information. It is in the beginner guide, for free, under the heading "Market Sessions."

And backtest your own hours. Your setup may have positive expectancy at 08:00 and negative at 14:00. Nobody's proprietary window is a substitute for your own data.

Gold Scalping

An honest analysis, which will be unpopular.

Full guide: Scalping vs swing trading gold: which suits you

The arithmetic

Assume a gold scalper targeting $2.00 moves with $1.00 stops (2:1 R:R), at 0.10 lots.

• Target: $2.00 × 100 × 0.10 = $20
• Risk: $1.00 × 100 × 0.10 = $10
• Spread: $0.30 → $0.30 × 100 × 0.10 = $3.00

Spread consumes 15% of the gross target on every trade.

Now add slippage on stops. Gold slips. Say 0.08R average. Your effective loss is 1.08R, not 1.00R.

Now compute expectancy. A 50% win rate at nominal 2:1:

• Gross: (0.50 × 2) − (0.50 × 1) = +0.50R
• After spread (0.15R per trade) and slippage (0.04R average): +0.31R

Still positive — if you can maintain a 50% win rate on $1.00 stops on an instrument whose ordinary noise is $2–4 wide.

You cannot. A $1.00 stop on gold is inside the bid-ask oscillation during any active session. It is not a stop; it is a coin flip with a fee attached.

Widen the stop, and the arithmetic changes

Use a $3.00 stop and a $6.00 target. Now the noise problem is solved. But:

• Spread as a fraction of the target falls to 5%. Better.
• Frequency collapses, because $6.00 moves in your direction, from your level, are far rarer than $2.00 moves.

You have not built a scalping strategy. You have built an intraday strategy and called it scalping.

Our position

Gold is a poor scalping instrument at retail costs. Not impossible — traders with institutional spreads and low latency scalp it. At $0.20–0.40 retail spreads, with a $2–4 noise band, the arithmetic is hostile.

If you want to scalp, scalp EUR/USD, where the spread is 0.1–0.5 pips and the noise band is proportionally narrower.

If you want to trade gold, trade it intraday or on swing. Its clean H4 structure and strong fundamental trends are where its edge lives.

Swing Trading Gold

Where gold's edge actually lives, in our assessment.

Why gold suits swing trading

Its fundamental driver trends. Real yields do not oscillate randomly hour to hour; they trend over weeks as the market reprices the Fed's path. That produces multi-week directional moves in gold, which is exactly what a swing trader needs and exactly what an intraday scalper cannot use.

Its H4 and daily structure is clean. Higher highs, higher lows, clear breaks of structure, well-respected zones — arguably cleaner than most FX majors.

Costs become negligible. A $0.30 spread against a $60.00 target is 0.5%. Against a $2.00 scalp target it was 15%.

Volatility works for you. Gold's large ranges mean a correct swing trade delivers R-multiples that FX majors rarely produce in the same holding period.

What you must account for

Swap. Gold's overnight financing is frequently negative on both sides and materially larger than FX. Held for three weeks, it accumulates. Model it in the backtest, or your live results will diverge from your test results in one direction only.

Weekend gaps. Gold gaps on Sunday's open, and gaps jump over stop losses — your stop fills at the next available price, not at your price. Geopolitical news over a weekend can produce a gap of several dollars.

Reduce size or close before weekends during periods of tension. This is the one risk a stop cannot protect against.

Events. A three-week hold will span an FOMC or a CPI. Decide in advance: hold through it with reduced size, or close before it. Do not decide on the day.

The core swing setup

Daily structure bullish (HH, HL). Real yields falling over the past two weeks. Price retraces into a demand zone confluent with the 50% retracement and a prior swing high. H4 shows a rejection candle or an internal break of structure.

Stop below the zone, plus 0.3 × daily ATR. Size from the stop. Target the prior high.

Nothing exotic. It is the pullback trade, executed on an instrument whose fundamentals you have checked.

Three Complete Strategy Specifications

These are templates demonstrating the required precision, not recommendations. They have not been validated for current conditions and may have negative expectancy. Backtest your own version and discard them if the numbers say so.

Specification 1 — Asian Range Sweep Reversal (Intraday)

Specification 2 — Real Yield Aligned Pullback (Swing)

Specification 3 — Post-Event Continuation (Intraday)

Part IV — Application

Six Trade Examples

Three of these are losses. These are constructed composites illustrating recurring structures on gold. They are educational illustrations, not a trading record, and they are not evidence that any method is profitable.

Example 1 — Asian Range Sweep (Win, +2.6R)

Context. Daily structure bullish. Real yields falling for eight sessions. Asian session range: $18 wide, roughly 0.6× daily ATR — enough to have built a liquidity pool on both boundaries.

The sweep. At 07:14 UTC, price pushed $4.30 below the Asian low, taking stops. By 07:45, the M15 candle closed back inside the range.

Execution. Long on the close back inside. Stop $1.90 below the sweep extreme (0.3 × H1 ATR). Total stop distance: $6.20.

Sizing on a $10,000 account, 1% risk: $100 ÷ ($6.20 × 100) = 0.16 lots.

Outcome. Price travelled to the Asian range high and beyond, reaching $16.10 above entry before stalling at the prior day's high. Exit at the range high. +$16.10 gross, +2.6R.

Why it worked. Not the pattern. The alignment. Daily structure said long. Real yields said long. The sweep provided a precise invalidation and therefore a tight stop. The same sweep against a bearish daily structure and rising real yields would have been a much worse trade.

Example 2 — The Inflation Hedge (Loss, −1.2R)

Context. CPI released at 13:30 UTC, materially hotter than expected. Headline and core both above consensus.

The reasoning (wrong). "Gold is an inflation hedge. High inflation. Buy gold."

Execution. Long within ninety seconds of the release. Stop $7.00 below.

Outcome. Nominal yields spiked as the market priced additional Fed tightening. Real yields rose sharply. Gold fell $22 in eleven minutes. Stopped out. Slippage of $1.40 on the fill. Actual loss: −1.20R.

What went wrong. Every part of the analysis.

Gold is not an inflation hedge on a trading horizon. Gold is a negative-real-yield hedge. A hot CPI print raises expected policy rates, which raises nominal yields faster than it raises inflation expectations, which raises real yields, which sinks gold.

Note also the slippage: entering within ninety seconds meant a widened spread on entry, and the stop filled $1.40 beyond its level. The loss exceeded 1R because the trade was placed in the worst execution window of the month.

The lesson. Trade the yield reaction, not the inflation narrative. And wait forty minutes.

Example 3 — EUR/USD Sizing on Gold (Loss, −5.3R)

Context. A trader profitable on EUR/USD migrates to gold. Their normal EUR/USD position is 0.50 lots.

The setup. A reasonable H4 demand zone. Structural invalidation $9.00 below entry, plus buffer. Stop distance $10.60.

The error. They traded 0.50 lots, because that is their normal size and 0.09 lots "looked pointless."

The arithmetic. $10.60 × 100 × 0.50 = $530 risk on a $10,000 account. 5.3%.

Outcome. The zone failed on a hawkish Fed speaker at 15:20 UTC. Stopped out. −$530. −5.3% of the account, on a single trade.

What went wrong. Nothing analytical. The zone was reasonable; it simply lost, as zones do. What made this catastrophic was that gold's lot sizes look small because gold moves enormously, and the trader adjusted upward to satisfy an intuition trained on a different instrument.

The lesson. The correct gold position size will offend you. Trade it anyway. Three losses like this and the drawdown table demands a 17% recovery.

Example 4 — Real Yield Aligned Swing (Win, +3.4R)

Context. 10y TIPS real yield had fallen 34 basis points over three weeks. Daily gold structure: clear HH, HL. Daily 50 EMA sloping up.

The setup. Retracement into a zone confluent with: the prior swing high, the 61.8% retracement, the daily 20 EMA, and a $50 round number.

Trigger. H4 bullish engulfing candle at the zone.

Execution. Long. Invalidation $27 below (the swing low that formed the last HL). Buffer 0.3 × $28 ATR = $8.40. Stop distance $35.40.

$100 ÷ ($35.40 × 100) = 0.028 → 0.02 lots. Actual risk $70.80.

Outcome. Held eleven days. Price reached the prior high. Swap cost over the hold: $9.40, roughly 0.13R. Exit at target. +$120 gross, −$9.40 swap, +3.4R net.

Why it worked. The fundamental filter did the heavy lifting. Real yields falling meant the structural wind was behind every long. The technical setup was ordinary — a textbook pullback. The edge was in not fighting the master variable.

Note the swap. It consumed 0.13R. Over forty such trades that is −5.2R. Model it.

Example 5 — The FOMC Reversal (Loss, −1R)

Context. FOMC statement, 19:00 UTC. Language read as more hawkish than expected. Gold dropped $18 within four minutes.

Execution. Short at 19:06 UTC, chasing the move. Stop $6.00 above.

Outcome. At 19:31 UTC, the press conference began. The Chair struck a notably softer tone in the Q&A, emphasising data dependence and downside employment risks. Gold reversed the entire move and travelled $22 higher. Stopped out at 19:44. Slippage $0.90.

What went wrong. Trading the statement rather than the reaction function.

The statement is the headline. The press conference is the information. Gold's most reliable FOMC-day pattern is that the initial move is frequently reversed during the Q&A — because the market's first read is algorithmic and its second read is considered.

The lesson. The professional response to FOMC is not a clever strategy. It is to be flat, watch, and consider a trade after 20:00 UTC at normal spreads. The expected value of trading a coin flip with $2+ spreads is negative regardless of how good your analysis is.

Example 6 — The Untested Order Block (Win, +2.1R)

Context. Daily structure bullish. Real yields flat — no fundamental tailwind, but no headwind. Strong impulsive H4 rally from a clearly defined origin.

The zone. The last bearish H4 candle before the impulse — an order block by the standard definition. Untested. Price had not returned to it.

The filter applied. Aligned with daily structure? Yes. Untested? Yes. Trigger on arrival? Awaited. Calendar clear? Verified. Sized at 1%? Yes.

Trigger. Price returned eight sessions later. H1 rejection candle formed at the zone's upper boundary.

Execution. Long. Stop below the zone, plus 0.3 × H1 ATR. Stop distance $11.20. $100 ÷ ($11.20 × 100) = 0.089 → 0.08 lots.

Outcome. Price reversed from the zone and reached the prior high. +2.1R.

The honest assessment. This worked. It would also have worked if the zone had been identified as an ordinary demand zone, drawn by eye, at the origin of the impulse. The order block terminology added a mechanical identification rule — which is genuinely valuable, because it makes the setup backtestable — and added nothing else.

The trade was carried by structural alignment, an untested zone, a trigger, a precise invalidation, and correct sizing. Remove any one of those and it fails. Remove the word "order block" and it is unchanged.

Risk Management for Gold Specifically

Everything in the risk masterclass applies without amendment. Gold changes the magnitude of every error, not the principles.

Full guide: Risk-reward ratio: sizing a gold trade properly

The gold-specific rules

1. Recalculate pip value from the contract spec. Never carry FX intuition across. Lots = Risk$ ÷ (Stop in $ × 100) for a 100oz contract. Verify your broker's spec first.

2. Expect stops to be 3–10× wider than EUR/USD in absolute terms, and positions to be proportionally smaller. The small position size is correct. It will feel wrong. Trade it anyway.

3. Never hold through FOMC, CPI or NFP unless the strategy is explicitly a news strategy that has been backtested as one, with realistic spreads and slippage.

4. Model slippage explicitly. Gold slips. Record actual fill prices in your journal and compute your average slippage in R. If it exceeds 0.10R, your effective loss is 1.10R and your expectancy calculation is wrong.

5. Model swap for any hold beyond two days. Gold's carry is frequently negative on both sides and larger than FX.

6. Reduce size or close before weekends during geopolitical tension. Gold gaps. Gaps jump stops.

7. Correlation. Long gold is a short-dollar position. Do not simultaneously hold short USD/JPY, long EUR/USD and long gold and call it three trades. Total correlated exposure ≤ 2R.

8. Halve your normal risk for your first fifty gold trades. 0.5% instead of 1%. You are learning a new instrument's noise band, and tuition should be cheap.

Common Mistakes

1. Treating XAUUSD as a currency pair. It is a non-yielding asset priced in dollars. Its driver is real interest rates, not a relative economy.

2. Buying gold on a hot CPI print. The single most common gold error. Gold is a negative-real-yield hedge, not an inflation hedge. Hot CPI → hawkish repricing → real yields up → gold down.

3. Carrying EUR/USD position sizes across. Example 3. A 5.3% loss on a single trade, with no analytical error at all.

4. Not verifying the contract specification. "Pip" is ambiguous on gold. Work in dollars of price movement. Confirm with a 0.01-lot demo trade.

5. Placing stops one dollar beyond obvious levels. That is the liquidity pool. Use 0.3–0.5 × ATR and accept the smaller size.

6. Scalping gold at retail spreads. A $0.30 spread against a $2.00 target is 15% of gross. Add slippage. The arithmetic is hostile.

7. Trading FOMC, CPI, or NFP. Spreads widen tenfold. Stops slip. The initial move frequently reverses. The expected value is negative.

8. Trading the FOMC statement and ignoring the press conference. Example 5. The statement is the headline; the Q&A is the information.

9. Analysing gold on M5. Its noise band is $2–4 wide. You will mistake noise for structure because the noise is large enough to look like structure.

10. Assuming DXY up always means gold down. During crises, both rise. Ask why the dollar is moving before assuming what gold will do.

11. Ignoring swap on swing positions. Gold's carry is frequently negative on both sides and materially larger than FX. Over a three-week hold it can consume a fifth of your target.

12. Holding gold over a weekend during geopolitical tension. Gaps jump stops. This is the one risk a stop cannot protect against.

13. Accepting order block and FVG theory without testing it. The past can be annotated to prove anything. Backtest prospectively, bar by bar.

14. Reading a gold price prediction and positioning around it. Nobody can forecast the Fed's reaction function to data that does not yet exist.

15. Trading gold as a beginner. It is the most popular retail instrument and among the most punishing. Learn on EUR/USD.

Pro Tips

Check the 10-year TIPS real yield every morning before you look at a gold chart. Ninety seconds. Almost nobody does it. Falling real yields = structural tailwind for longs.

Delete the word "pip" from your gold vocabulary. Work in dollars of price movement. Every ambiguity vanishes.

Mark the Asian range every day. 00:00–07:00 UTC. Then watch what happens to its boundaries between 07:00 and 09:00.

Halve your risk for your first fifty gold trades. 0.5%. You are learning a new instrument's noise band. Make the tuition cheap.

Record actual fill prices, not intended prices. Compute your average slippage in R after fifty trades. It is almost certainly costing you more than you think, and it belongs in your expectancy calculation.

Read average hourly earnings before the NFP headline. Wages feed inflation expectations, which feed rate expectations, which feed real yields, which move gold. The headline is the distraction.

Never trade in the first forty minutes after a major release. The initial move is algorithmic. The considered move arrives later, at normal spreads, on wider stops. You have all day.

Analyse on H4 and daily. Execute on H1 or M15. Gold's higher timeframe structure is unusually clean. Its lower timeframe noise is unusually violent.

Ask why the dollar is moving before assuming what gold will do. Rate differentials → gold down. Fear → gold up, alongside the dollar. Same DXY candle, opposite implications.

Model swap before you design any swing strategy on gold. Then include it in the backtest, or your live results will diverge from your test results in one direction only, permanently.

Expert Insights

On why gold is the graveyard of intermediate traders. Gold attracts traders at exactly the moment they become dangerous — after six months, when the mechanics are automatic, the chart reading is competent, and the EUR/USD returns feel modest. Gold moves three times as much, and the arithmetic of that fact is not intuitive. They carry their position sizes across. They carry their stop distances across. Both are catastrophically wrong, and the failure arrives not as a single dramatic blow-up but as a series of "unlucky" 4% losses that they attribute to volatility rather than to the sizing decision they made before the market opened. Gold does not punish new mistakes. It punishes old ones, at three times the price.

On the real yield insight, and what happens after you have it. There is a moment, when a gold trader first understands the real yield relationship, when everything on the chart reorganises. The CPI reversals make sense. The FOMC whipsaws make sense. The months when gold rallied through a strong dollar make sense. This feels like acquiring an edge. It is not an edge. It is the absence of confusion, which is a different and more modest thing. Everyone who reads the Treasury's page has it. What you do with it — whether you use it as a bias filter, size correctly, and wait for your setup — is where the edge would have to live, and that part has nothing to do with gold.

On why we wrote the longest guide on the site about an instrument we tell people not to trade. Because they will trade it anyway. XAUUSD is the most popular speculative instrument in retail trading, and telling a determined trader to stay away is less useful than telling them what will happen. Every trader who has lost money on gold lost it to the same short list of causes: sizing carried across from FX, the inflation-hedge myth, trading a news release, and a stop placed one dollar beyond an obvious low. Four causes. All avoidable. None of them requires a better indicator.

On order blocks, and the broader question they raise. A mechanical identification rule is genuinely valuable — it makes a setup testable, and testability is the only thing separating a strategy from a story. But notice how much of the appeal lies in the narrative: institutions, smart money, unfilled orders, a hidden structure visible to the initiated. That narrative is unfalsifiable in an OTC market with no visible order book. It is doing no analytical work whatsoever, and it is doing enormous emotional work. Be suspicious of any framework whose primary product is the feeling of insight. The market does not pay for feeling correct.

On the only question that matters. After all of this — the real yields, the sessions, the sweeps, the specifications — the question that determines whether you make money on gold is the same one that determines it on every other instrument: what happens to your account when this trade is wrong? If the answer is "I lose 1% and take the next setup," you have a chance. If the answer is anything else, none of the preceding nine thousand words will save you.

The Professional Gold Checklist

Weekend

• 10y TIPS real yield: direction over the last 10 sessions noted.
• DXY: direction, and why it is moving — rate differentials or fear?
• Weekly and daily gold structure marked. HH/HL/LH/LL.
• Three to five significant zones marked, sized from daily ATR.
• Daily ATR(14) noted.
• Next week's calendar: FOMC, CPI, NFP, PPI dates and times marked on the chart.
• Geopolitical tension assessment. Any weekend gap risk?
• One-paragraph written bias.

Every morning (5 minutes)

• 10y real yield checked. Direction unchanged?
• Economic calendar checked. Any high-impact release today, and at what time?
• Asian range marked (00:00–07:00 UTC).
• Current spread checked. Normal?
• Account equity confirmed. 1% risk amount recalculated.
• Emotional state 1–5. Below 3 → do not trade.

Before every gold entry

• Setup matches the written specification. All conditions present.
• Daily structure aligns with trade direction.
• Real yield direction does not oppose the trade.
• Zone is untested, or tested fewer than three times.
• Trigger present — not entering on arrival alone.
• Invalidation identified precisely. I can state what would prove me wrong.
• Stop placed beyond invalidation + (0.3–0.5 × ATR). Not one dollar beyond the obvious low.
• Position size calculated in dollars of movement, not "pips." Risk$ ÷ (Stop in $ × 100)
• Risk ≤ 1% of equity. (≤ 0.5% for the first fifty gold trades.)
• No high-impact news within 60 minutes.
• Spread < $0.50.
• Not correlated with an open short-dollar position elsewhere.
• If held overnight: swap cost modelled. If held over a weekend: gap risk accepted or size reduced.
• Screenshot taken. Thesis written in one sentence.

After

• Exit screenshot.
• Result in R.
• Actual fill prices recorded. Slippage computed.
• Swap paid recorded.
• MAE and MFE.
• Process scored: plan followed? Yes/No — independent of outcome.

End of day

• −2R limit respected. If hit, platform closed.
• Every trade journaled.

Cheat Sheet

The master variable Real yield ≈ Nominal yield − Expected inflation Real yields up → gold down. Real yields down → gold up. Watch the 10y TIPS yield (U.S. Treasury, daily).

The hierarchy Real yields > Fed > DXY > safe-haven flow > central bank buying > ETF flows > technicals

The reframe Gold is not an inflation hedge. Gold is a negative-real-yield hedge. Hot CPI → hawkish repricing → real yields up → gold falls.

Contract 1.00 lot = 100 oz. A $1.00 move = $100 per standard lot. Ignore the word "pip." Work in dollars of price movement.

Position size Lots = Risk$ ÷ (Stop in $ × 100) Level → stop → size. The small number is correct.

Stops 0.3–0.5 × ATR(14) beyond invalidation. Never $1 beyond an obvious low. Gold overshoots levels by $3–6 routinely.

Sessions (UTC) Asian 00–07 (range builds) · London open 07–09 (sweep) · Overlap 12–16 (peak) · Rollover ~21 (avoid)

The four events FOMC (19:00, presser 19:30) · CPI (13:30) · NFP (13:30, first Friday) · Dot plot (quarterly) Do not trade them. Wait 40+ minutes.

DXY Usually inverse to gold. Both rise in a crisis. Ask why the dollar is moving.

Correlation Long gold = short dollar. Do not stack it with other short-dollar positions.

Gold-specific rules Halve risk for the first 50 trades · Model slippage · Model swap · Close or reduce before weekends during tension · Analyse H4/daily, execute H1/M15

Glossary

Basis point (bp) — One hundredth of a percentage point. A 34bp fall in real yields is 0.34%.

Full guide: The full forex and gold trading glossary

Central bank buying — Sovereign accumulation of gold reserves. Price-insensitive, structural demand that can decouple gold from the real yield relationship for extended periods.

COMEX — The exchange where gold futures (ticker GC) trade. The source of true volume data for gold, unlike spot CFDs.

Dot plot — The Summary of Economic Projections. Each Fed official's projected rate path, published quarterly. Moves real yields immediately.

DXY — The US Dollar Index. Dominated by the euro (≈57.6%), making it approximately an inverted EUR/USD chart.

Fair value gap (FVG) — A three-candle pattern where the outer wicks do not overlap. A volatility artefact marking where price moved fast. Its "magnetic" properties are asserted more often than tested.

Nominal yield — The stated yield on a bond, before adjusting for inflation.

Order block — The last opposing candle before an impulsive move. A supply/demand zone with a mechanical identification rule. The institutional narrative attached to it is unfalsifiable in an OTC market.

Real yield — The inflation-adjusted return on a bond. Directly observable via TIPS. The master variable for gold.

Safe haven — An asset that appreciates during crises. Gold and the US dollar are both safe havens, which is why they sometimes rise together.

Sweep / liquidity sweep — Price extending beyond an obvious level to trigger resting stop orders, providing counterparties for a large order, then reversing. Gold's native behaviour.

Swap — Overnight financing. On gold, frequently negative on both long and short sides, and materially larger than FX.

TIPS — Treasury Inflation-Protected Securities. Bonds whose principal adjusts with inflation. Their yield is the real yield.

Troy ounce — The unit gold is priced in. Approximately 31.1 grams. One standard CFD lot is 100 troy ounces.

XAU — The ISO code for one troy ounce of gold. XAUUSD is the price of one ounce in US dollars.

People Also Ask

Why is gold falling when inflation is high?

Because gold tracks real yields, not inflation. A hot inflation print causes markets to price more central bank tightening, which lifts nominal bond yields faster than it lifts inflation expectations. Real yields therefore rise, the opportunity cost of holding a non-yielding metal rises, and gold falls. Gold is a negative-real-yield hedge, not an inflation hedge.

What is the best indicator for gold trading?

The 10-year TIPS real yield, published daily by the U.S. Treasury, and it is not on your chart. Check its direction each morning before you look at gold. Falling real yields put a structural tailwind behind long setups. On the chart itself, ATR matters more than any oscillator, because it sizes your stop.

How much is 1 lot of gold worth?

For most brokers, one standard lot of XAUUSD is 100 troy ounces, so a $1.00 move in the gold price equals $100 of profit or loss. A 0.10 lot moves $10 per dollar; a 0.01 lot moves $1. Verify this on your broker's contract specification, because the term pip is defined inconsistently across brokers on gold.

Can you day trade gold with $100?

Mechanically, if your broker permits 0.01 lots. Sensibly, no. At 1% risk that is $1 per trade, and a typical gold stop of $10 in price movement would require a position size below the 0.01 minimum. Gold's volatility makes small accounts structurally unable to size correctly. Learn on EUR/USD.

Should I trade gold during NFP?

Almost certainly not. Spreads widen from cents to several dollars, slippage on stops is severe, and the initial move is algorithmic and frequently reversed within minutes. The considered move arrives forty minutes to two hours later at normal spreads. Note that average hourly earnings often moves gold more than the headline jobs number.

Frequently Asked Questions

What moves the price of gold? Primarily real interest rates — the inflation-adjusted yield on government bonds, observable via 10-year TIPS. Gold pays no yield, so when real yields rise, the opportunity cost of holding it rises and it falls; when real yields fall, gold rises. Secondary drivers are the US dollar (gold is priced in dollars, so a stronger dollar mechanically suppresses demand), safe-haven flows during crises, and central bank buying. The Federal Reserve matters enormously, but only because the Fed moves real yields.

Why does gold drop when interest rates rise? Because gold pays no interest. If a risk-free, inflation-protected government bond yields 2% in real terms, holding a non-yielding metal costs you 2% per year in forgone return. Institutional capital rotates out. When real yields fall or turn negative, the bond guarantees a loss of purchasing power and gold's 0% becomes competitive, so capital rotates in.

Is gold an inflation hedge? Over decades, broadly. Over the horizon on which you trade, unreliably, and frequently in the opposite direction. What matters is inflation relative to nominal yields. Hot inflation typically causes central banks to hike, which raises nominal yields faster than inflation expectations, which raises real yields, which pushes gold down. Gold has fallen through genuinely inflationary periods. Gold is a negative-real-yield hedge, not an inflation hedge.

Why does gold fall when the dollar rises? Mechanically: gold is priced in dollars, so a stronger dollar makes the same ounce more expensive for every non-dollar buyer, suppressing demand. But this holds only when the dollar is moving on interest rate differentials. During acute crises, capital seeks safety and both gold and the dollar rise together. Always ask why the dollar is moving before assuming what gold will do.

Is XAUUSD good for beginners? No, despite being the most popular retail instrument. Gold's daily range is routinely two to three times EUR/USD's in percentage terms, which amplifies every position-sizing error proportionally. Its contract specification differs from FX and the term "pip" is ambiguous on it. It sweeps liquidity aggressively, punishing tight stops. Learn on EUR/USD, and come to gold after you can size a position from a stop distance in your sleep.

How much is one pip in gold? This question has no universal answer, and that ambiguity is dangerous. Some brokers define a pip on gold as $0.01, others as $0.10. Ignore the term entirely. Work in dollars of price movement. For a standard 100-ounce contract, a $1.00 move equals $100 per lot, a $0.10 move equals $10, and a $0.01 move equals $1. Verify your broker's contract specification before your first trade.

How do I calculate lot size for gold? Lots = Risk amount ÷ (Stop distance in dollars × 100), assuming a 100-ounce standard contract. On a $10,000 account risking 1% ($100) with a $10.60 stop: $100 ÷ ($10.60 × 100) = 0.09 lots. That number will look small if you are used to FX. It is correct. Do not adjust it upward.

What is the best time to trade gold? The London session (07:00–16:00 UTC) and particularly the London/New York overlap (12:00–16:00 UTC), where liquidity peaks and spreads are tightest. The London open (07:00–09:00 UTC) produces the day's first expansion and frequently a liquidity sweep of the Asian range. Avoid the Asian session unless you are running a range strategy, and avoid the hour around rollover (~21:00 UTC) when spreads blow out.

Should I trade gold during NFP or CPI? Almost certainly not. Spreads widen from cents to several dollars, slippage on stops is severe, and the initial move is algorithmic and frequently fully reversed within minutes. The considered move arrives forty minutes to two hours later, at normal spreads, and is far more tradeable. The expected value of trading the release window is negative regardless of the quality of your analysis.

Is gold good for scalping? At retail spreads, the arithmetic is hostile. A $0.30 spread against a $2.00 target consumes 15% of gross profit before slippage. Meanwhile a $1.00 stop sits inside gold's ordinary noise band, which is $2–4 wide during active sessions. Widening the stop to solve the noise problem converts the strategy into intraday trading. Gold's edge lives in intraday and swing timeframes, where its clean H4 structure and multi-week fundamental trends can be exploited.

What are order blocks in gold trading? The last opposing candle before a strong impulsive move — a supply or demand zone with a mechanical identification rule. That rule is genuinely useful because it makes the setup backtestable. The accompanying narrative — that institutions left unfilled orders there — is unfalsifiable in an OTC market with no visible order book. Treat an order block as a well-defined zone, apply the same filters you apply to any level (higher-timeframe alignment, untested, trigger, precise invalidation, calendar clear, 1% risk), and backtest it prospectively before believing it.

What is a fair value gap? A three-candle pattern where the first and third candles' wicks do not overlap, leaving an unfilled range. It is a volatility artefact marking where price moved quickly. The claim that price is "drawn back" to fill it is a claim about mean reversion after volatility expansion — a real phenomenon in some conditions, but not a special property of the pattern. Price fills most nearby areas eventually; the untested question is whether it fills FVGs more often than equivalent arbitrary zones.

Are kill zones real? Volatility and liquidity genuinely concentrate at the London and New York opens. That is measurable and a session filter is one of the cheapest improvements available to any strategy. The mystique is not real — these are the London and New York sessions, documented in market microstructure literature for decades, renamed. Use a session filter, backtest your own hours, and do not pay anyone for this information.

Where is the price of gold going? Nobody knows, and anyone telling you otherwise is guessing or selling. Gold's price depends on the future path of real interest rates, which depends on the Federal Reserve's reaction to economic data that has not yet been released. Replace the question with a better one: what conditions would make gold rise, and are they present? Falling real yields, a dovish Fed, a weakening dollar on rate differentials, geopolitical stress, and central bank accumulation favour gold. Their opposites pressure it. You do not need a forecast. You need a bias, a setup, and a position size that makes being wrong cost 1%.

Why is gold so volatile? Because its price is a real-time referendum on the future path of real interest rates, and that path is repriced continuously as data arrives. It carries no yield to anchor it, no earnings, and no cash flow — only expectations. Add heavy retail participation clustered around obvious levels, deep institutional liquidity willing to reach for those stops, and a large average daily range, and you have an instrument that moves violently in both directions with no fundamental floor other than sentiment about central banks.

Resources

The four guides

• Forex Trading Education — Homepage — the full curriculum and learning path.
• The Complete Beginner Forex Trading Guide — mechanics, pips, lots, leverage, margin.
• Complete Technical Analysis Guide — structure, levels, sweeps, indicators, multi-timeframe.
• Forex Trading Strategy & Risk Management Masterclass — expectancy, sizing, journaling, backtesting, psychology.

Free tools

Gold Position Size Calculator · Position Size Calculator · Risk of Ruin Simulator · Trading Journal Template · Session Clock

Primary data — bookmark these

• U.S. Treasury — Daily Treasury Real Yield Curve Rates. The single most useful page on the internet for a gold trader. Check it every morning.
• Federal Reserve — FOMC calendar, statements, minutes, Summary of Economic Projections.
• U.S. Bureau of Labor Statistics — CPI and Employment Situation release schedule.
• World Gold Council — quarterly gold demand trends, central bank purchases, ETF flows.
• CME Group — COMEX gold futures (GC) contract specifications and true volume data.
• ICE — US Dollar Index composition and weightings.

Books

• The New Case for Gold — James Rickards. Read critically; the author has a position and a thesis.
• Thinking in Bets — Annie Duke. Process versus outcome. Essential for an instrument this volatile.
• A Man for All Markets — Ed Thorp. Position sizing, and why full Kelly is unusable.
• Fooled by Randomness — Nassim Taleb. On mistaking luck for skill, which gold's volatility makes very easy.

Conclusion & Next Steps

Gold is the most popular instrument in retail trading and the least understood, and those two facts are causally connected.

It attracts traders because it moves. It removes them for the same reason.

But the reasons people lose money on gold are not mysterious, and there are only about four of them.

They carry their EUR/USD position sizes across. A stop that is $10.60 wide on gold, traded at a EUR/USD-sized 0.5 lots, is a 5.3% loss on a single ordinary trade. No analytical error was required.

They believe gold is an inflation hedge. So they buy a hot CPI print, and the market prices Fed hikes, and nominal yields rise faster than inflation expectations, and real yields rise, and gold falls, and they are stopped out in four minutes wondering what happened. Gold is a negative-real-yield hedge. Once you see that, the CPI reversals and the FOMC whipsaws and the months when gold rallied through a strong dollar all resolve into sense.

They trade the news release. Spreads widen tenfold, stops slip, the initial move is algorithmic and frequently reversed. They had a view. The view was irrelevant.

They place stops one dollar beyond an obvious low. That is not a stop. That is the liquidity pool, advertised on every chart, on an instrument that sweeps more aggressively than any major pair.

Four causes. All avoidable. None of them requires a better indicator, a proprietary framework, or a price prediction.

And beneath all four sits the sentence that four guides on this site have been building toward:

The market does not pay for being right. It pays for being sized to survive being wrong, and for still being there when the setup with the edge finally arrives.

Gold merely charges more for forgetting it.

What to do now

This week. Open your broker's contract specification sheet. Confirm ounces per lot and the P&L of a $1.00 move. Place one 0.01-lot demo trade and verify the arithmetic with your own eyes. Then delete the word "pip" from your gold vocabulary permanently.

Every morning from now on. Check the 10-year TIPS real yield. Check the economic calendar. Mark the Asian range. Ninety seconds, three habits, and you are already doing more than most people trading this instrument today.

This month. Mark gold's daily structure every day. Note its ATR. Watch what happens to the Asian range boundaries between 07:00 and 09:00 UTC. Do not trade. Watch.

Then. Backtest one setup, manually, bar by bar, 100 trades, with realistic spreads and modelled slippage and modelled swap. If expectancy after all three is negative, discard it and design another. That outcome is a success — you have just saved an account.

Then, and only then. Forward test at 0.5% risk, not 1%. Thirty trades on demo. Thirty on 0.01 lots with real money. Score process, not P&L.

Gold will still be here. It has been here for six thousand years, and it has never once been in a hurry.

Start Here

→ Return to the Strategy & Risk Management Masterclass Expectancy, sizing, journaling, backtesting. Everything on this page assumes it. If any of it was unfamiliar, go back — gold is the wrong place to learn it.

→ Download the XAUUSD Cheat Sheet (PDF) The real yield relationship, contract specs, sizing formula, session clock, and the daily five-minute routine. One page. Free, no email.

→ Gold Position Size Calculator Works in dollars of price movement, as it should. Then learn to do it without the tool.