Order blocks and fair value gaps explained

By AnalytikCapital · Updated October 3, 2026

An order block is the last opposite-colored candle, or group of candles, before a strong move that breaks market structure, read as the zone where large orders were placed. A fair value gap is a three-candle imbalance in which the first and third candles do not overlap, leaving a price range the middle candle crossed in a single push.

What is an order block?

An order block is a zone left behind by the last opposing candles before an impulsive move. In the usual teaching:

• A bullish order block is the last down-close candle (or run of down-close candles) before a strong rally that breaks a prior swing high. It is read as demand, a zone price may come back to before continuing up.
• A bearish order block is the last up-close candle (or run) before a strong drop that breaks a prior swing low. It is read as supply.

The move away from the block is what qualifies it. A candle before a weak drift does not count; the departure has to be fast and has to break market structure. In Episode 3 of his 2022 mentorship, ICT adds more context to a high-probability block: the move left a gap, it took liquidity first, and it produced a market structure shift. He also warns against the shortcut of treating every down-close candle as a bullish order block.

Where do sources disagree on order blocks?

Almost every detail has more than one version:

What AnalytikCapital’s engine does. It marks an order block only behind a structure break made with displacement, meaning the breaking candle’s body is larger than 1.5 times the 14-period average true range. From that candle it walks back to the last opposite-colored candle and uses that single candle; grouping a run of candles into one block is deferred to a later version. The zone spans the candle’s full high-to-low range by default (a body-only setting exists in configuration). Each block starts active, becomes mitigated on the first touch, and is invalidated when a candle closes beyond its far edge. A wick through the zone alone never invalidates it.

What is a fair value gap?

A fair value gap (FVG) is a three-candle pattern:

• Bullish FVG: the high of candle 1 is below the low of candle 3. The gap runs from candle 1’s high up to candle 3’s low.
• Bearish FVG: the low of candle 1 is above the high of candle 3. The gap runs from candle 3’s high up to candle 1’s low.

The middle candle traveled through that range so fast that the candles on either side never overlapped it. ICT describes it the same way in Episode 3 and, in Episode 2, says he introduced the term in 2016. Other sources call it an imbalance or an inefficiency.

It is not the classic chart gap, where no trade happened at all between two bars. In an FVG the range was traded, but only in one direction.

Sources then diverge on qualifiers: whether the middle candle must be a large displacement candle, whether a minimum gap size applies, and how much of the gap must be filled before it counts as used. AnalytikCapital’s engine applies no size minimum and no displacement filter to gaps: any wick-to-wick non-overlap is a gap. It records the 50% midline and tracks how deep price has filled it, as open, partially filled or filled.

What are inverse FVGs and balanced price ranges?

Inverse fair value gap (IFVG). When price breaks through a gap instead of respecting it, the gap is said to flip roles: a bearish gap that price breaks above becomes possible support, and a bullish gap broken below becomes possible resistance. FXOpen’s explanation notes that traders often track the flip after a close or a strong wick beyond the gap. AnalytikCapital takes the stricter reading: a gap inverts only when a candle closes beyond its far edge. The inverted zone is then tracked like an order block (active, mitigated, invalidated).

Balanced price range (BPR). Where a bullish and a bearish gap overlap in price, the overlap is called a balanced price range, sometimes described as a double FVG (ATAS). The engine pairs each gap with the most recent opposite gap and records their overlap. Because a BPR belongs to both sides, the plan engine uses it only as supporting evidence, never as an entry on its own.

How are they used?

In SMC practice, both zones are places where price may react on a return:

1. Identify the trend and the structure break that created the zone.
2. Prefer zones on the cheap side of the range for longs and the expensive side for shorts (see premium and discount).
3. Watch for a reaction when price returns, rather than assuming the zone will hold.
4. Look at nearby liquidity for where the move might be heading.

On AnalytikCapital, live order blocks and gaps are the Smart Money half of a plan’s entry: a zone counts only where it overlaps the Elliott Wave retracement band and sits on the correct side of the premium and discount midpoint.

What are the limits?

• Many zones, few reactions. Lower timeframes print gaps and opposite candles constantly. Without filters, a chart fills with zones, and price will touch some of them by chance.
• Hindsight. Textbook examples show blocks that held. Blocks that failed are rarely shown, and a failed zone is often given a new label after the fact instead of being counted as a miss.
• No standard. The table above is the reason two traders can mark different zones on the same chart. A rule-based engine fixes one reading; it does not prove that reading is the right one.
• No academic validation of order blocks or fair value gaps as trading edges exists that we are aware of.

The methodology page shows how these zones combine with the Elliott count into a plan, or into a stated rejection.

Educational content, not financial advice. Trading involves substantial risk of loss.