Fair Value Gaps in Futures Trading: How to Identify and Trade FVGs on ES and NQ

What the three-candle gap actually is

A fair value gap (FVG) is a hole in a three-candle sequence. Candle two is usually a large, fast candle, and the wicks of candle one and candle three don't overlap. In a bullish FVG, the low of candle three sits above the high of candle one, and the zone between those two prices is the gap. A bearish FVG is the mirror image: the high of candle three sits below the low of candle one.

The idea comes from ICT teaching, which explains these gaps as footprints of large institutional orders. I treat that as a theory, not something a chart can prove. What you can observe is the gap itself and what price does when it comes back. Everything below is built on what you can measure.

Marking FVGs on an ES or NQ chart

1. Pick one timeframe and stay with it for the test. A 5-minute chart is common, but the choice is yours.
2. Take three consecutive candles. For a bullish gap, check whether candle three's low is higher than candle one's high. For a bearish gap, check whether candle three's high is lower than candle one's low.
3. Draw a box from candle one's edge to candle three's edge. Extend it to the right until price trades back through it.
4. Record the gap size in points and the time of day it formed.

Here's a hypothetical ES example. Candle one's high is 5,400.00 and candle three's low is 5,406.50. The gap is 6.5 points. That is $325 of price distance per ES contract ($50 per point) or $32.50 per MES contract ($5 per point). There's no universal minimum size worth trading. Compare the gap to the average candle range on your chart, because a 6-point gap means something different in a quiet midday session than right after a data release.

Approach 1: entering on the retrace

The common FVG trade waits for price to come back into a bullish gap after an impulsive move up. You don't buy before price enters the zone. You wait for it to trade into the box, then enter long with a stop beyond the far side of the gap.

Using the same hypothetical numbers, say price drops back and you buy at 5,404.00. The bottom of the gap is 5,400.00 and you put the stop at 5,399.50, which is 4.5 points of risk. That's $225 on one ES contract or $22.50 on one MES contract, before fees and slippage. If your first target is a prior swing high 9 points above entry, the reward is twice the risk. Whether you reach it often enough is a question only your data can answer.

Two failure modes matter. Price can trade straight through the gap and keep going, which is what the stop is for. Or price can never come back, and you miss the move. In a strong trend, waiting for a retrace can mean waiting all day. Neither problem has a fix, only a cost you accept.

Approach 2: gaps as reference points for exits

You can also mark unfilled gaps from earlier in the session or the prior day and use them as places to review an open trade. If you're long and an unfilled bearish gap sits a few points above, that zone is a reasonable spot to take partial profit or tighten a stop. Treat it as a reference, not a magnet. I don't have a tested fill rate for intraday FVGs, and I wouldn't trust one quoted without the data, the timeframe, the sample dates and how "filled" was defined.

Gaps versus opening gaps

An opening gap compares one session's close to the next session's open, and the result depends on which session you use. A Globex-based gap and an RTH-based gap (regular trading hours, with the US equity cash open at 9:30 a.m. ET) can look very different on the same day. Intraday FVGs are a different object, formed from three candles on your chosen timeframe. Don't assume the fill behavior of one carries over to the other. If the question matters to you, define the gap, collect a sample, and count.

Using FVGs next to KPL reference prices

KPL are fixed reference prices anchored to the RTH regular-session opening print. The standard ES and NQ map has four levels, built from full-session and after-hours travel, and the levels don't move intraday. Because the anchor is the opening print, the levels can't be completed before RTH opens. An FVG is a separate, chart-based observation, and I use it only as extra context when one overlaps a KPL price. See the KPL trading overview for how the levels are built.

One caution on tighter stops. An FVG can let you place a 3 to 5 point stop instead of a wider one, and the dollar risk per contract drops. But a tighter stop also gets hit by ordinary noise more often, so a smaller stop doesn't make the trade better by itself. Stops can slip in fast markets too.

How to test it before trusting it

Log every FVG that meets your written definition, including the ones you'd have skipped. Record entry, stop, target, outcome and the time of day. Thirty trades is not a proven edge, and fifty simulated trades are not live proof, since a simulated fill doesn't reproduce exchange queue position or slippage. Once you have a sample, run the win rate and average win and loss through the edge calculator to see what break-even looks like after costs.