What a fair value gap is
A fair value gap (FVG) — also called an imbalance or inefficiency — is a three-candle pattern that marks an area where price moved so quickly that trading only happened in one direction.
- Bullish FVG: the high of candle 1 is below the low of candle 3. The gap between them, inside the large middle candle, is the FVG.
- Bearish FVG: the low of candle 1 is above the high of candle 3.
The concept is central to ICT concepts and Smart Money Concepts. The theory is that price may return to “rebalance” the area before continuing. In practice, an FVG is simply a measurable footprint of a fast move — useful for marking where momentum originated, but not a promise of anything.
Why FVGs are everywhere on XAUUSD
Gold regularly moves several dollars in a single candle around session opens and economic releases. That means FVGs appear constantly, on every timeframe. If you treat them all as trade signals, you’ll have dozens of “setups” per day. The real skill is filtering.
How FVGs behave on gold (XAUUSD)
| Gold trait | Effect on fair value gaps |
|---|---|
| Session opens | The London and New York mornings produce the cleanest displacement, and so the most-watched gaps. |
| 8:30 New York data | Releases such as CPI or payrolls can print gaps of many dollars in one candle. These are often revisited quickly and with wide spreads. |
| Thin Asian trade | Small gaps form on low participation. They tend to matter less than gaps created when London or New York is active. |
| Round numbers | A gap that sits just above or below 4,100 or 4,200 often gets “overfilled” as price probes the round number. |
| Dollar and yields | A bullish gap created on a sharp drop in yields can be erased just as fast if yields reverse. |
| Overshoots | Gold frequently trades through the whole gap and a little beyond before reacting, which is why stops at the gap edge get hit so often. |
Check ATR 14 on your working timeframe. A $3 gap on an M15 chart means something different when the average candle is $2 than when it’s $8.
How to mark a fair value gap step by step
- Find displacement. Look for a large-bodied candle that moved price quickly, ideally breaking a swing.
- Take the three candles. Candle 1 is the one before the big candle, candle 2 is the big candle and candle 3 is the one after.
- Check for the gap. Bullish: is candle 1’s high below candle 3’s low? Bearish: is candle 1’s low above candle 3’s high? If the wicks overlap, there’s no FVG.
- Draw the zone from candle 1’s wick to candle 3’s wick and extend it to the right.
- Mark the midpoint. ICT calls the 50% level of the gap consequent encroachment (CE).
- Track its status: untouched, partly filled, filled to CE, or fully filled. Stop extending a gap once price closes fully through it.
| Term | Meaning |
|---|---|
| BISI | Buy-side imbalance, sell-side inefficiency: ICT’s name for a bullish FVG |
| SIBI | Sell-side imbalance, buy-side inefficiency: a bearish FVG |
| Consequent encroachment | The 50% level of the gap |
| Inversion FVG | A gap that price closes through, which some traders then treat as support or resistance from the other side |
| Balanced price range | Where a bullish and a bearish gap overlap, so the area has been traded in both directions |
Variations and settings: from beginner to pro
| Choice | Beginner approach | What experienced traders refine |
|---|---|---|
| Which gaps | Every gap on the chart | Only gaps from displacement that broke structure, ideally after a sweep |
| Entry level | First touch of the gap edge | CE (50%), or the far edge for a better price and more missed trades |
| Timeframe | Same timeframe for context and entry | H1/H4 gaps for location; M5/M1 gaps inside them for timing |
| Stop | Just beyond the gap | Beyond the swing that created the displacement |
| Failed gaps | Forgotten | Tracked as possible inversion FVGs |
| Timing | Any hour | Gaps formed inside kill zones and Silver Bullet windows |
“Fill” rules are a big source of inconsistency. Decide in advance whether a gap counts as respected if price touches the edge, reaches CE or trades through completely and closes back. Then use the same rule in testing and live trading.
Filtering meaningful gaps
Traders commonly prioritize FVGs that:
- Formed during displacement that broke structure — for example the move that created a Change of Character.
- Followed a liquidity sweep — see liquidity sweeps.
- Align with the higher-timeframe direction.
- Sit in discount (for longs) or premium (for shorts) of the relevant swing.
- Formed on a meaningful timeframe — H1 or higher for context, lower timeframes only for entry refinement.
- Are still unfilled — a gap that has already been fully traded through has lost its relevance in most frameworks.
Ways traders use FVGs
- As an entry area. Placing an order inside the gap (at its edge or midpoint) in the direction of the original move.
- As confluence. Combining an FVG with an order block or a support zone.
- As a target. Some traders treat unfilled gaps on the opposite side as potential objectives.
- As a strength gauge. A gap that price respects quickly suggests strong momentum; a gap that price slices through suggests the move is weakening.
Three FVG setups traders use
1. Displacement continuation at consequent encroachment
| Element | Rule |
|---|---|
| Context | Clear H1/H4 direction. A displacement leg breaks structure in that direction and leaves a fresh gap. |
| Trigger | Price retraces into the gap for the first time. |
| Entry | Limit order at CE (50% of the gap). |
| Stop (invalidation) | Beyond the swing that started the displacement, not just beyond the gap. |
| Targets | The high or low made by the displacement, then the next liquidity pool. |
| Management | If price closes fully through the gap, the setup is weakening. Consider exiting early rather than waiting for the stop. |
2. Silver Bullet window FVG
| Element | Rule |
|---|---|
| Context | One of the Silver Bullet windows: 03:00–04:00, 10:00–11:00 or 14:00–15:00 New York time. A draw on liquidity is identified beforehand. |
| Trigger | Inside the window, price takes nearby liquidity and then displaces toward the draw, leaving an M1–M5 gap. |
| Entry | First retrace into that gap. |
| Stop (invalidation) | Beyond the liquidity-taking extreme. |
| Targets | The pre-identified draw on liquidity. |
| Management | Many traders exit or cancel if the window ends without a fill. See the Silver Bullet guide. |
3. Inversion FVG
| Element | Rule |
|---|---|
| Context | A bullish gap fails: price closes through it with displacement, often after a sweep of a high. |
| Trigger | Price rallies back into the failed gap from below. |
| Entry | Short at the gap, which is now treated as resistance. |
| Stop (invalidation) | Above the swing high that swept liquidity. |
| Targets | The low made by the move that broke the gap, then sell-side liquidity below. |
| Management | This is a reversal idea. Use smaller size than continuation setups until your own testing supports it. |
Entry conditions and invalidation (educational example)
On a hypothetical M15 chart during the London session, gold sweeps a prior low and then rallies hard. Candle 1 has a high of 3,811.2; candle 3 has a low of 3,814.6. The bullish FVG sits between 3,811.2 and 3,814.6, and the rally breaks the most recent lower high. A trader might place a limit order near the top of the gap (around 3,814.5) or wait for price to enter the gap and show a lower-timeframe bullish reaction. Invalidation could sit below the swing low that started the move rather than just below the gap, because price often trades through the full gap before reversing. A close well below the gap would mark the setup as failed.
Risk management
- Don’t place stops just below the gap by default. Price often fills the entire gap and a little more. Use the swing low/high that created the move as a more robust invalidation, and size down.
- Expect missed trades. In strong momentum, price may never return.
- Be careful around news. FVGs created by a data release can be revisited violently, with wide spreads.
- Fixed risk per trade. Read XAUUSD risk management.
Worked example (hypothetical)
Account: $5,000. Risk per trade: 1% ($50). Contract: 1 lot = 100 oz, so 0.01 lot = $1 per $1 move.
Context. The H1 trend is down. After the 8:30 US data, gold rallies into the 10:00–11:00 New York Silver Bullet window and takes buy-side liquidity above an intraday high at 4,186.0, reaching 4,188.4.
Setup. Price then drops sharply and breaks the M5 higher low. Candle 1’s low is 4,182.6 and candle 3’s high is 4,178.2, so the bearish FVG is 4,178.2–4,182.6. CE is 4,180.4.
Plan. - Entry: sell limit at 4,180.4 (CE). - Stop: 4,189.6, which is $1.20 above the sweep high. - Stop distance: 4,189.6 − 4,180.4 = $9.20. - Risk per 0.01 lot: $9.20. - Position size: $50 ÷ $9.20 = 5.4, rounded down to 0.05 lot. Actual risk: 5 × $9.20 = $46. - Target 1: session low at 4,166.0 ($14.40, about 1.6R). - Target 2: previous-day low at 4,158.0 ($22.40, about 2.4R).
Management. Close 0.03 lot at Target 1 ($43.20). Move the stop on the remaining 0.02 lot to 4,181.0 only after a new M5 lower low. Cancel the order if it hasn’t filled by 11:00.
What would prove it wrong. A close above 4,182.6 shows the gap was traded through. A move above 4,188.4 invalidates the idea completely. These numbers illustrate the method only, and many trades like this lose.
Combining FVGs with other concepts
- Order blocks: a gap next to an order block shows the block produced displacement. Some traders enter in the gap with the stop beyond the block.
- Liquidity sweeps: gaps created right after a sweep have a clearer story than gaps in the middle of a range.
- Premium, discount and OTE: an FVG that overlaps the 62–79% retracement band is a common ICT refinement.
- Breaker and mitigation blocks: inversion FVGs and breakers often appear together after a failed move.
- Power of 3 (AMD): the distribution leg of the daily template often leaves the gaps traders watch.
- Kill zones and Silver Bullet: a time filter that removes many low-quality gaps.
- VWAP and ATR and Bollinger volatility: useful for judging whether a gap is large or small relative to current conditions.
Best timeframes and sessions
| Timeframe | Typical use on gold |
|---|---|
| Daily / H4 | Major gaps that act as context or targets. Often left after big macro events. |
| H1 / M15 | Setup gaps for intraday and short swing trades. |
| M5 / M1 | Entry timing inside a higher-timeframe area. Gaps here are extremely common. |
Sessions (New York time, GMT when the US is on standard time): - London kill zone, about 02:00–05:00 (07:00–10:00 GMT): displacement after Asian-range sweeps. - New York AM, about 07:00–10:00 (12:00–15:00 GMT): 8:30 data produces the biggest gaps of the day. - Silver Bullet windows: 03:00–04:00, 10:00–11:00 and 14:00–15:00. - Treat with caution: Asian-session gaps (about 20:00–00:00), the 17:00–18:00 daily break, and gaps formed in the minute of a release, when spreads are widest.
Convert these to your own clock with the gold market hours tool.
Common mistakes and limitations
- “Gaps must fill.” They don’t. Treating fills as inevitable leads to counter-trend trades against strong moves.
- Trading every gap. Without filters, FVGs become noise.
- Ignoring timeframe. An M1 gap has little to do with H4 structure.
- Using inconsistent levels — the edge on one trade, the midpoint on the next.
- Hindsight. The gaps that “worked” stand out; the ones that failed are forgotten. Journal every FVG you would have traded, not only the ones you did.
Fair value gaps are best thought of as a precise way to describe where a fast move began. Combined with structure, liquidity and strict risk rules, they can help define locations. On their own, they’re not a strategy.
Limitations and honest caveats
- The “rebalancing” story is a model. Nobody can see from a retail chart that institutions “need” to return to a gap. Price may revisit a gap simply because fast moves often retrace.
- Broker feeds differ. Gold is traded over the counter, so candle highs and lows vary slightly between brokers. A gap on one platform may be a tiny overlap on another, especially on M1 and M5.
- Timeframe choice changes everything. The same move shows a gap on M5 and none on H1. Your rules must name the timeframe.
- No verified universal hit rate exists. Figures quoted online are rarely reproducible. Build your own sample of 50–100 rule-based examples before trusting any setup.
- Context beats the pattern. A perfect gap in front of a Federal Reserve decision is still a coin toss on the event.
News-driven gaps on gold can be revisited in seconds, with slippage well beyond your planned stop. Avoid resting limit orders in a gap right before a scheduled release unless you accept that risk and have sized for it.
Pro tips
- Judge the gap by what created it. A gap from a structure-breaking move after a sweep is a different object from a gap in a drifting range. Grade them separately in your journal.
- Watch reactions at CE. Price that turns cleanly at the 50% level often shows stronger intent than price that grinds to the far edge. Test whether that holds on your timeframe.
- Use the far edge as an early warning. A close beyond the far edge doesn’t always hit your stop, but it tells you the gap has lost its usefulness.
- Nest gaps across timeframes. An M5 gap inside an H1 gap gives you a precise entry while keeping the higher-timeframe reason.
- Note unfilled gaps on the opposite side. They are reasonable candidates for targets, though never guaranteed.
- Record missed trades too. Many good gaps never get revisited. Knowing how often that happens tells you whether limit entries or confirmation entries suit you better.
Quick checklist
- Did the gap form during displacement that broke structure?
- Did a liquidity sweep come before it?
- Does it agree with the higher-timeframe direction?
- Is it in discount for longs or premium for shorts?
- Is it fresh and still unfilled?
- Have I chosen my entry level (edge, CE or far edge) in advance?
- Is my stop beyond the swing that created the move, with size calculated from it?
- Is a high-impact release due during the trade?
- Is the gap inside a session window I’ve tested?
- Will I log the result, including if the gap is never revisited?
Frequently asked questions
What is a fair value gap?
A fair value gap (FVG) is a three-candle pattern in which the first candle’s high and the third candle’s low (bullish) — or the first candle’s low and the third candle’s high (bearish) — do not overlap. The space between them, inside the middle candle, is the gap.
Do fair value gaps always get filled?
No. Many FVGs are only partly revisited and many are never revisited at all, especially in strong trends. The idea that gaps must be filled is a common and costly misconception.
Is an FVG the same as a price gap?
No. A classic price gap is a jump between one candle’s close and the next candle’s open, often over a weekend. An FVG is formed within continuous trading by three overlapping candles.
What timeframe is best for FVGs on gold?
Gaps on higher timeframes (H1 and above) are generally considered more significant than those on M1 or M5, where gaps are extremely common. Many traders use higher-timeframe gaps as areas of interest and lower timeframes to time entries.
What is consequent encroachment in an FVG?
Consequent encroachment (CE) is ICT’s term for the 50% level of a fair value gap. Some traders use it as an entry level, and others watch whether price closes beyond it as an early sign that the gap is failing. It’s a convention, not a level with proven special properties.
What is an inversion fair value gap?
An inversion FVG is a gap that price closes straight through, after which some traders treat it as support or resistance from the other side. For example, a failed bullish gap may later act as resistance. Like all FVG ideas, it should be tested with written rules before you rely on it.
Sources and further reading
- CME Group — Gold futures overview — Contract specifications and trading hours for COMEX gold
- CFTC — Commitments of Traders — Weekly futures positioning, including gold
- Investor.gov — Stop order — How stop orders work, and their limits in fast markets
External sites are independent of XAUUSD Global. Check dates and data yourself.