What Is a Fair Value Gap?And Why Most Traders Get Them Wrong

A fair value gap (FVG) is a three-candle pattern that marks an imbalance in price — a band the market ran through so fast it left orders unfilled on one side. "Imbalance" and "inefficiency" are just other names for the same thing.

Here's the pattern, candle by candle, built around a strong middle candle:

• Bullish FVG: price displaces up hard. The gap is the space between the high of the first candle and the low of the third candle — and for it to be a real gap, that third candle's low has to sit above the first candle's high. The middle candle moved so fast that no trading happened in that band of price. That empty band is the FVG, and it sits below current price as a potential support zone.
• Bearish FVG: the mirror. Price displaces down hard. The gap is the space between the low of the first candle and the high of the third candle, with the third candle's high sitting below the first candle's low. It rests above price as a potential resistance zone.

The middle candle is the one that matters — it's the displacement, the violent one-sided move that creates the imbalance. The gap it leaves behind is the footprint of that aggression.

Why "fair value"? Because that fast move was unfair — only one side got filled. Price left a pocket where buyers ran over sellers (or the reverse) without a proper two-way auction. The market has a habit of returning to those pockets to rebalance — trading back through the gap to offer fair value to both sides before continuing.

The textbook story is that an FVG is unfinished business: the displacement skipped a chunk of price, orders in that zone never got filled, and price comes back to "rebalance the inefficiency" before carrying on. It is a good story. We wanted to know whether it shows up in the numbers, so we measured every fair value gap on 13 markets since July 2022: 221,004 on 15-minute charts and 52,187 on the 1-hour.

Price does come back, and fast. On the 15-minute chart, 93% of gaps were touched within 24 hours and 87% were filled completely; the median wait for the first touch was 30 minutes. On the 1-hour chart, 85% were touched within a day and 74% filled.

Then we ran the test that matters. For every real gap we drew a fake one: the same size, the same distance from price, but on the other side, where no gap exists. The fake gaps filled just as often: 87.2% within a day on the 15-minute chart against 87.3% for the real ones, and 74.0% against 73.7% on the 1-hour. Same by gap size, same in every session. A fair value gap does not pull price back to it. It gets filled because it sits close to where price is, and price revisits nearby levels all the time.

That changes what an FVG is for. It is a precise map of where price just moved fast, which is genuinely useful context for placing an entry or a stop. It is not, on its own, a reason to expect a reaction.

Fills come in degrees, too. Price might tap the near edge of the gap, push to its midpoint, or close the whole thing. A lot of traders treat the 50% midpoint, consequent encroachment in ICT terms, as the real level rather than the outer edge. In the data, the first visit rarely stops at either: once price got back to a 15-minute gap, it went on through the whole of it within five days 97% of the time.

Here's where it falls apart for most people. They learn the three-candle rule and start treating every gap on the chart as a setup. But your chart is full of fair value gaps. Drop to any timeframe and there are dozens — the overwhelming majority meaningless.

An FVG is only as good as the move that made it and the place it sits. A gap left by a tired, choppy push in the middle of a range is noise. A gap left by an explosive move that just broke structure after sweeping liquidity is a completely different animal — and treating those two the same is exactly why so many FVG trades are random.

The lazy version of FVG trading goes: see gap, expect fill, expect reversal, enter. It fails constantly, because gaps don't owe you anything. Plenty get run clean through and never look back — usually the ones that formed against the higher-timeframe direction, or with no real displacement behind them. The gap itself isn't the edge. The context around the gap is the edge.

The gaps worth trading tend to share a few traits. Stack them, and a random imbalance becomes a level with real weight.

It came from genuine displacement

The move that made the gap should be decisive — a strong, one-sided candle (or run of them) with momentum, not a lazy drift. Best case, that displacement breaks market structure: it takes out a prior swing point and shifts the trend. A gap born from a structure break is the market showing its hand. A gap from a nothing move is just a gap.

It formed after a liquidity sweep

This is the big one, and it ties straight to what liquidity is and how it moves price. The highest-probability FVGs appear right after price sweeps a pool of liquidity — takes out an obvious high or low, then reverses and displaces hard the other way, leaving a gap as it goes. Sweep, displace, leave the gap, retrace into it for the entry. The liquidity grab is what fuels the kind of move that creates a gap worth trading.

It sits in the right place

A gap that runs with the higher-timeframe trend, in a sensible location — below price in an uptrend, above price in a downtrend — carries far more weight than one fighting the bias or floating in the middle of a range. Context first, gap second.

It's still fresh

An FVG price hasn't returned to yet — unmitigated, untouched — is far more likely to produce a clean reaction than one price has already traded back through. Once a gap has been filled and used, most of its charge is spent.

None of these are boxes you tick off a screenshot. You build the eye for them by watching the sequence repeat, which is the whole reason backtesting ICT concepts beats watching another video on them.

The general sequence most FVG entries follow:

1. Wait for a liquidity sweep. Price takes out an obvious high or low. Not mandatory, but it's the cleanest context you can ask for.
2. Look for displacement that shifts structure. A strong move the other way that breaks a recent swing point — and leaves an FVG in its wake.
3. Mark the gap. The imbalance left by that displacement is your zone.
4. Wait for the retrace. Let price come back into the gap. You're not chasing the displacement candle — you're waiting for the pullback into the inefficiency.
5. Enter on the reaction, target the next liquidity. The stop sits on the far side of the gap or beyond the swing that created it, where the idea is plainly wrong. The target is the opposite liquidity pool the move is reaching for.

Your exact entry trigger, how deep into the gap you wait, and how you manage the trade are yours to sharpen — that's the gap between knowing the pattern and actually having an edge. If your model runs on Candle Range Theory, the same displacement-and-return logic sits underneath it; we cover that in What Is CRT (Candle Range Theory)?.

A fair value gap is simple to define and easy to draw. That's exactly why so many traders lose on them — the drawing is trivial, the judgment is everything. The pattern tells you where an imbalance is. Only context — the displacement, the liquidity, the structure, the location — tells you whether it's worth a trade.

You can't shortcut that judgment. It's pattern recognition, and pattern recognition is reps. The fastest way to get them without burning real money is to replay price candle by candle and watch FVGs form, fill, and fail in real time — hundreds of them, until you can separate the clean ones from the traps at a glance.

That's what the FVG backtester is built for. Pick a market, replay it bar by bar with the candle-by-candle replay engine, and mark the sweep, the displacement and the gap yourself as they form. It's made for ICT backtesting, so the fair value gap is exactly the kind of concept you'll drill until reading it is automatic. Study FVGs once; watch a thousand of them play out. That's how a definition turns into a skill.

What is a fair value gap in simple terms? A fair value gap is a three-candle pattern where price moves so fast in one direction that it leaves a gap — a band of price the middle candle skipped, where one side never got to trade. It marks an imbalance. Price usually trades back into it, but measured, no more often than it trades back to any level the same distance away.

How do you identify a fair value gap? Look at three consecutive candles with a strong middle candle. For a bullish FVG, there's a gap between the high of the first candle and the low of the third (the third's low sits above the first's high). For a bearish FVG, it's the gap between the low of the first candle and the high of the third. That empty band is the FVG.

Do fair value gaps always get filled? Not always, but most do, and quickly. Of 221,004 gaps on 15-minute charts, 87% were filled completely within 24 hours and 94% within five days; on 1-hour charts, 74% and 88%. The catch is that a fake gap drawn the same distance away filled just as often, so the fill tells you price is close, not that the gap is special. Expecting a fill and a reversal from it is the fastest way to lose on them.

Does trading the FVG retest work? Not by itself. We tested the version usually taught: a limit order at the gap's near edge, stop at the far edge, target 2R. Over 189,423 trades on 15-minute charts it won 26% of the time and averaged −0.21R before costs, and small gaps did worst. Large gaps came out at roughly zero. Any edge lives in what you add — the sweep, the displacement, the session — which is why you backtest your own FVG rules against that baseline.

What's the difference between a fair value gap and an imbalance? They're the same thing. "Imbalance" and "inefficiency" are just other names for the pocket left when price displaces too fast for a proper two-way auction. A fair value gap is the specific three-candle version traders mark and trade.

Is a fair value gap bullish or bearish? Either — it depends on the direction of the move that created it. A bullish FVG forms in an up-move and acts as potential support below price; a bearish FVG forms in a down-move and acts as potential resistance above it.