Fair Value Gap (FVG) Decoded: The Institutional Imbalance Behind Precise Entries
Fair Value Gap (FVG) Decoded: The Institutional Imbalance Behind Precise Entries
Institutional algorithms leave price footprints called Fair Value Gaps. Learn the exact 3-candle matrix, the three FVG types, a real MNQ setup around $28,820 — and the four regimes where the framework quietly fails.
Once you can read structure with the CHoCH vs BOS framework and filter fake shifts with the Inducement rule, one question remains: where exactly do you enter? Chasing the last green candle is not an answer. The Fair Value Gap is one of the cleanest ways to define an execution zone that has an actual mechanical reason to exist.
This post breaks down how to identify an FVG, the three types you will see on a chart, how to combine it with higher-timeframe context on a real MNQ example, and — importantly — the market regimes where FVG logic stops working.
💡 The Core Concept: When large orders push price quickly, the tape moves through levels faster than resting limit orders can absorb. That creates a structural imbalance — a zone where one side barely participated. Algorithms and desks with unfilled resting interest are frequently drawn back to those zones to "rebalance" before the next expansion.
- Anatomy of an FVG — the 3-candle matrix and how to spot it
- Three types of FVG — bullish, bearish, and the Balanced Price Range
- The confluence entry strategy — combining FVG with HTF context
- Reality Check — where the FVG framework still fails
1. Anatomy of an FVG — the 3-candle matrix
An FVG is defined mechanically on three consecutive candles. It represents a range where price traded through so quickly that no meaningful two-sided auction occurred. That empty range often acts as a reversion zone the next time price approaches it.
The empty range between Candle 1 High ↔ Candle 3 Low is the FVG (imbalance zone)
The mechanical rule that turns three arbitrary candles into a valid FVG:
On three consecutive candles where Candle 2 is a large aggressive bullish expansion, if the High of Candle 1 does not touch or overlap the Low of Candle 3, the empty range between them is a valid bullish FVG.
Think of the gap as an unfilled delivery zone. Because one side barely got any resting orders filled during the expansion, price often revisits the range to complete that delivery before the next leg. The reversion is a probability tilt, not a guarantee.
Figure 1: Mapping the Fair Value Gap (FVG) matrix and executing an entry on Micro E-mini Nasdaq futures. Price retraces precisely to the FVG upper boundary at $28,820 before resuming the bullish expansion.
2. Three Types of FVG You Will See on a Chart
Most retail teaching only shows the bullish version. In practice, you need to recognise three variants so you do not misread the setup or confuse two unrelated gaps stacked at the same level.
① Bullish FVG
Three-candle sequence where the middle candle is a large aggressive bullish expansion. The gap sits between Candle 1 high and Candle 3 low. Bias to the upside. Price often revisits the upper boundary of the gap as a pullback entry before continuation.
② Bearish FVG
The mirror image. Middle candle is a large aggressive bearish expansion. The gap sits between Candle 1 low and Candle 3 high. Bias to the downside. Price often revisits the lower boundary as a rally-into-supply entry before the next leg down.
③ Balanced Price Range (BPR)
Not a separate FVG — a bullish FVG and a bearish FVG overlapping at the same price zone. This means the level was traded through aggressively in both directions and left imbalances on both sides. A BPR is generally a stronger reversion magnet than either FVG alone, because whichever direction price approaches from, one of the two imbalances is still unfilled.
For a rigorous definition of imbalance and inefficiency in a broader market microstructure context, see Investopedia's overview of market inefficiency.
3. The Confluence Entry Strategy — combining FVG with HTF context
An FVG on its own is a B-grade signal. It becomes a high-conviction setup only when it stacks with higher-timeframe context — an HTF order block, a completed inducement sweep, and a session where institutions actually participate. On the MNQ example in Figure 1, the entry was executed in four steps:
⚡ Identify the expansion that printed the FVG
Following the 4H institutional demand zone mitigation, MNQ printed a structural shift on the 5M chart and left behind a clean bullish FVG. The middle candle was aggressive and one-sided — not a slow drift.
🎯 Mark the zone and wait — do not chase
Instead of chasing the expansion candle at the top of the move, draw a rectangle across the imbalance and wait. Most entries are lost by acting during the impulse instead of during the retracement.
✅ Execute at the FVG boundary tap
As the New York session progressed, price retraced back to the imbalance. When price tapped the upper boundary of the FVG at $28,820, the resting bids were triggered and continuation resumed.
🛡️ Define risk against structure, not the FVG
Stop loss sits below the structural low at $28,680, not below the FVG lower boundary. Anchoring the stop to the structural low keeps the setup honest — if the low breaks, the thesis was wrong regardless of what the FVG did.
💎 Why this works: a standalone FVG is a B-grade signal. An FVG stacked inside an HTF order block, on the correct side of the inducement sweep, during the NY session is a high-confluence trade. Stop trading single signals — stack at least three independent reasons before you click.
- ❌ Never trade an FVG without an HTF directional bias
- ❌ Never chase the candle that just printed the imbalance
- ✅ Wait for price to tap the FVG boundary — never enter mid-air
- ✅ Stack confluence: FVG + order block + inducement sweep + session timing
- ✅ Stop loss below the structural low — never inside the FVG itself
4. Reality Check — Where the FVG Framework Still Fails
FVGs are useful precisely because they are mechanical. But the mechanical rule does not care about the underlying regime, and the framework has real failure modes you need to name before you trade real capital against them.
① Strong trends leave FVGs unfilled
In a persistent trend, FVGs can remain unfilled for days or weeks. Waiting patiently for the retrace can mean missing the entire move. When higher timeframes are in an obvious expansion, expect fewer retracements and treat each unfilled FVG as evidence of trend strength, not as a guaranteed reversion.
② News-driven gaps behave differently
FVGs printed on the response to FOMC, CPI, NFP, or Sunday globex open are driven by information and repricing, not by liquidity mechanics. They can get filled quickly, take months to fill, or never fill at all. Treat headline-response FVGs with more caution than session-driven ones.
③ HTF and LTF FVGs are not equal
A 4H or Daily FVG carries far more weight than a 1M FVG. On low timeframes the market prints dozens of small imbalances per session, most of which are noise. Treat every timeframe with the same signal severity and you will over-trade. HTF imbalances get respected; LTF imbalances need HTF confluence to matter.
④ Confirmation bias when marking FVGs
If you already want to be long, every 3-candle sequence starts to look like a valid bullish FVG. The rule ("Candle 1 high does not overlap Candle 3 low") is mechanical for a reason. If you catch yourself squinting to make an FVG appear, the setup is not there.
Even A-grade FVG setups — HTF order block plus valid FVG plus inducement sweep plus session timing — still lose approximately 40 to 45% of the time in normal market conditions. The framework does not eliminate losing trades. It ensures that the losing trades come from setups where the analytical work was sound, which makes them recoverable rather than random.
Chasing candles is emotional. Waiting for the imbalance to be revisited is mechanical. The FVG is not a magic pattern — it is a rule that turns "somewhere near the last big move" into a specific price zone with a specific reason to expect participation. Sit at the boundary, define risk against structure, and let the setup either happen or not happen.
📚 Continue the SMC Journey
Deepen your Smart Money Concepts and volume decision-making with these companion posts:
⚠️ Disclaimer: This article is for educational and informational purposes only and does not constitute financial, investment, or trading advice. Trading involves substantial risk of loss. Always conduct your own research and consult a licensed financial advisor before making any investment decision. Read the full disclaimer →
About the Author
Dongmin Park is a software engineer with over 15 years in embedded systems (automotive and defense industries) and 10+ years of active trading across Korean equities, US options, MNQ futures, and crypto. He started trading on a Kiwoom Securities account in Seoul in 2016 and now lives in Ingolstadt, Germany, after relocating in 2022.
Coder Trader is an ongoing project to document where systematic engineering discipline meets discretionary trading. Say hi on X, look at the code on GitHub, or email hello@codertrader.com.
Comments
Post a Comment