What is a Fair Value Gap (FVG)?

A Fair Value Gap (FVG) is a three-candle price-action pattern: the wick of the first candle and the wick of the third candle do not overlap, leaving a vertical "gap" of untraded prices across the middle candle. The pattern is most often associated with smart-money-concepts (SMC) trading, but the underlying idea — that strong directional moves leave un-transacted price levels behind — is older and broader than any specific methodology.

The theory is straightforward. When price moves so quickly that one side of the order book dominates, some orders sitting in the gap range never get filled. Practitioners argue that price often returns to those un-transacted levels later to "rebalance," giving the FVG predictive properties. The honest framing: this is observation, not law. Markets are under no obligation to fill any specific gap, and many FVGs remain open for months or years.

ProChart treats FVGs the way any rigorous practitioner does — as one structural marker among several, never as a stand-alone signal.

The appeal of FVGs comes from three properties. First, they are mechanical to identify: the three-candle definition is unambiguous, so two traders reading the same chart will draw the same gaps. Second, they are visual: a gap on a chart is immediately legible without indicators, lagging averages, or oscillators. Third, they map onto an intuitive market story — that fast moves leave inefficient prices behind and that markets tend to revisit them.

That story is partially true. Returns to FVGs do happen often enough to be statistically observable in many markets. But "often enough to be observable" is not the same as "reliably tradeable." The next sections cover where the pattern is useful, where it isn't, and how to avoid the most common ways traders misuse it.

On any candlestick chart, an FVG is identified by looking at three consecutive candles. Compare the wick range of the first candle to the wick range of the third candle. If they do not overlap, the price range in between is the FVG. The middle candle is typically large — the bigger the displacement, the larger and more notable the gap.

Most charting platforms and SMC indicator packs draw FVGs automatically. ProChart's analysis pipeline includes calibrated FVG detection that filters by the minimum displacement size for the asset's volatility regime — small "noise" FVGs on low-volatility crypto pairs and on quiet forex sessions are filtered out so the surfaced gaps carry meaning.

An FVG is "unfilled" or "open" while price has not returned to its range. It becomes "filled" — sometimes called mitigated, closed, or rebalanced — when price trades back through it. Different practitioners use different fill criteria. Strict definitions require the full gap range to be traded through (i.e., wick or body has touched both extremes). Looser definitions count any single touch into the gap range as fill. Body-only criteria require a candle close inside the gap. None of these is the "correct" definition; what matters is consistency.

Two empirical observations worth keeping in mind. First, many FVGs are filled within hours or days; others remain open for months. Second, the gap getting filled does not mean the original move's thesis is invalidated — fill is just a return to the level, not a directional outcome. A bullish FVG getting filled can be a clean support test before continuation up, or it can be the start of a deeper retracement. The fill itself doesn't tell you which.

FVGs are observations of past price action. They tell you something happened — a displacement that left untraded prices behind. They do not tell you what will happen next. Specifically, they do not tell you whether price will return to the gap, when it will return, in what direction it will leave, or whether the broader trend supports your interpretation.

A research-grade approach uses FVGs as one of several inputs. Useful context to combine: the active higher-timeframe trend, key supply/demand zones, news and macro catalysts inside the holding period, an explicit invalidation level (where the thesis is wrong), and a defined reward target. Without an invalidation level, an "FVG trade" is a guess; with one, it's a setup with bounded risk that may or may not work.

FVG detection is mechanical, which makes it well-suited to AI assistance. A calibrated detector can scan many assets and timeframes consistently, filter noise FVGs below a displacement threshold, flag which open FVGs align with higher-timeframe structure, track which have been touched or mitigated since formation, and cross-reference with other signal layers (trend direction, news context, sentiment, positioning).

What AI cannot do is make any individual FVG more likely to play out. The underlying limits of the pattern — that fill is not a directional outcome, that many FVGs never fill, that hindsight bias inflates historical apparent edge — are properties of the dataset and the human reading it, not properties an AI can remove. The honest advantage AI brings is speed, consistency, and the ability to surface only the FVGs that pass higher-quality filters.