Crypto Fair Value Gaps (FVG): Three-Candle Imbalances on USDT Pairs

• Bullish FVG - candle 3's low is above candle 1's high, and candle 2 closed up. The gap runs from candle 1's high to candle 3's low.
• Bearish FVG - candle 3's high is below candle 1's low, and candle 2 closed down.
• Size filter - a gap smaller than 0.1 × ATR 14 is ignored as noise.
• Displacement - candle 2's range is at least 1.5 × ATR 14 and its body at least 60% of that range.

What happens to a gap afterwards:

• First retest - the first day that trades back into a gap nothing has touched since it formed, without closing through it.
• Filled - price reaches the gap's far edge (a bullish gap's bottom).
• Inverted - a daily close goes through the far edge. The band is then read the other way round: a broken bullish gap becomes resistance, an inverse FVG. The page lists the day it flips and stops following the gap after that.

Why crypto gaps look different

In Unfluke's archive, 99% of crypto daily candles from November 2025 to November 2026 opened within 0.5% of the previous close; every jump of 1% or more sat across a day the feed missed. With no opening gap to help, candle 3 can only clear candle 1 if candle 2 covers the whole distance. So crypto gaps are fewer for the same volatility, and the ones that form tend to come from forceful days: the 17–19 September gap below needed an 18 September candle 2.2 times its ATR.

The exception is a missing day. 5 August, 11 September and 5 November 2026 have no candle, so the days either side sit next to each other, and the chart can show a jump that never traded. A gap formed across one of those days is worth a second look on a one-hour chart.

A bullish gap from formation to inversion

• Candle 1, 17 September: high 77,179.47.
• Candle 2, 18 September: up from 76,744.40 to 80,990.25, a range of 4,998 against an ATR 14 of 2,263, with 85% of it body: displacement.
• Candle 3, 19 September: low 80,844.58.
• The gap: 77,179.47 to 80,844.58, 1.62 × ATR. On 19 September BTCUSDT was listed on New Bullish and Displacement.
• 20 September, first retest: the low of 80,126.04 traded into the gap and the close of 81,271.20 held above it, so the pair moved to Retest Bullish. It passed the volume test that Sunday at 0.84 × its average, just over the 0.8 line.
• 21–22 September: a close of 85,923.02 and a high of 87,395.67.

Over the next six weeks the gap was traded into on 14 days, and only the last of them closed below it. On 30 October, the market-wide sell-off covered in the Volume Shockers guide, BTCUSDT closed at 76,918.91, below 77,179.47: filled and broken in one day, so it appeared on Inverse Bearish. That day the crypto list had 47 rows: 19 gaps filled, 15 bullish gaps inverted, 9 first retests of older bullish gaps and 4 new bearish gaps. A flush day fills the list with old gaps being used up.

An inverse gap that failed

• The 30 October candle left a bearish gap from 77,796.01 (31 October's high) to 79,710 (29 October's low), a displacement gap.
• On 4 November BTCUSDT closed at 80,600.01, above 79,710: the bearish gap was inverted and listed on Inverse Bullish, read as support.
• The next candle in the archive, 6 November (5 November is missing), closed at 75,856.01, back below the whole band.

Gaps are places to watch, not walls, and that is just as true of inverse gaps. The page judges each one on daily closes and moves on; it does not promise that the flipped band will hold.

1. Start with the side. Use the Retest Bullish or Retest Bearish tab, and check the pair's trend on the SMC Overview; a retest with the trend is the cleaner reading.
2. Prefer displacement gaps. A gap left by a 1.5 × ATR candle shows urgency; check the Size (ATR) column for room to set risk.
3. Look at the day the gap formed. Was it a market-wide day (BTC and ETH moving together, as on 30 October) or the pair's own news? A market-wide gap tends to be retested when the market turns, whatever the pair does.
4. Mark the far edge. A daily close beyond it ends the idea. Crypto wicks run far in thin hours; the page uses the close, and so should the plan.
5. Check the futures side. A displacement day with open interest falling sharply (see the OI Buildup guide) was positions being closed, not fresh conviction.

Risk per trade - risk a small, fixed share of capital on any one idea. Crypto never closes, so a stop is the only exit while you sleep; with leverage, keep the liquidation price well beyond the stop. When selling options, prefer defined-risk structures (spreads, iron condors) over naked positions.

These are ideas to study, paper-trade and backtest, not recommendations or signals, and not a promise of any result. Every condition below can fail. Crypto trades around the clock and can move sharply at any hour; leverage and options can lose more than you expect, and past behaviour says nothing certain about the next session. Size, entries, exits and risk are your own decisions.

1. The first retest, with the trend

• Setup: a pair on Retest Bullish whose daily trend on the Overview is bullish, ideally a displacement gap.
• Entry idea: a daily close back up out of the gap, or a break of the retest day's high.
• Invalidation: a daily close below the gap's bottom (it would then be inverted).

2. Fade rallies into an inverse gap

• Setup: a pair on Inverse Bearish: a bullish gap a close went straight through, like BTCUSDT's on 30 October.
• Reading: buyers who defended the gap are trapped; a rally back into the band often meets supply.
• Invalidation: a daily close back above the band.

3. Weekend caution

• Idea: a gap that forms over a thin weekend, on a pair that only just passed the volume test, says less than a weekday gap on heavy volume. Treat it as a watchlist entry until the week's volume confirms it.

4. Treat Filled as done

A filled gap has done what SMC expects of it. It no longer marks an imbalance and the page stops following it.

Test before you trade - write the rules down exactly, test them on past data (see how to backtest an options strategy and the Option Simulator, which supports crypto), then paper-trade at least 20 occurrences before committing real money. Keep a journal of every trade, including the ones you skipped.

Every Smart Money page has a Candles setting: Daily (the default) or Weekly. Weekly runs exactly the same rules on weekly candles built from the same IST daily bars. On crypto:

• A week is Monday 00:00 to Sunday 23:59 IST, seven daily candles: Monday's open, the week's high and low, Sunday's close, the volume added up.
• The current week is the week so far. Its candle is dated the newest day, so on a Monday it is a one-day candle; a weekly row seen mid-week can change, or drop off, by Sunday's close.
• Dates in a weekly row are the week's last day: a Sunday, or the newest day for the current week.
• Counts become weeks: a swing needs five weeks on each side, ATR 14 is fourteen weeks, and every look-back counts weeks.
• What stays daily: the liquidity filter (the newest day's turnover and volume), LTP and change, and the Vol Ratio column.
• History: the scan pages build weeks from about 400 days of daily bars, so a weekly swing or zone there is at most about a year old. The SMC Overview reads up to about three years where the archive has them (158 weeks for BTCUSDT).

A weekly idea for gaps: a weekly gap is the band between one week's high and the low two weeks later, a move that ran for a whole week without looking back. BTCUSDT's open weekly gap on 8 November ran from 65,474.46 to 76,670.01, from the August rally. A daily first retest inside a weekly gap of the same direction has two timeframes pointing at one zone; judge a weekly retest or fill on Sunday's close, not mid-week.

The tabs

• All - every gap event of the day.
• New Bullish / New Bearish - a gap formed today.
• Retest Bullish / Retest Bearish - the first trade back into an untouched gap.
• Filled - price reached the far edge today.
• Inverse Bullish / Inverse Bearish - a close went through a gap today, flipping it.
• Displacement - new gaps made by a 1.5 × ATR candle with a strong body.

The columns

• Side - Bullish or Bearish.
• Gap Bottom / Gap Top - the band's edges, in USDT.
• Size (ATR) - the gap's height in today's ATR 14 units.
• Formed / Age - the day candle 3 closed, and days since.
• Event - what happened today.
• Chart - a pop-up with the gap shaded from its first candle and the middle candle marked.

Signed in, the date box shows the list for any past IST day your plan includes.

Which pairs are scanned

• The live USDT pairs in Unfluke's crypto archive, from BTCUSDT and ETHUSDT to mid-caps such as GRTUSDT and THETAUSDT: 33 were scanned on 8 November 2026.
• Stablecoins are left out by behaviour: a pair whose average daily range over 20 days is under 0.25% (USDCUSDT's is a few hundredths of a percent) is not scanned. A day with under US$100,000 traded is skipped as untraded.
• The liquidity filter: a 20-day average turnover above US$500,000, and the day's volume above 0.8 times its 20-day average. Crypto has no price bands, so there is no circuit check.
• The universe box: All, Futures or Options (pairs that also have a perpetual future, or options, in the archive).

Three things that are different on crypto

• A day is an IST day. Each daily candle runs from 00:00 to 23:59 IST, rolled up from the one-minute spot archive (Binance data). Another chart that closes its days at 00:00 UTC (05:30 IST) will show different candles, and so different swings and zones.
• Weekends thin the lists. The 0.8 × volume test is the same as on NSE, but crypto trades every day and volume falls away at weekends. From 10 August to 8 November 2026 an average of 21.2 pairs passed on a weekday and 10.8 on a Saturday or Sunday; BTCUSDT itself was off the lists on 27 of 89 days, 21 of them weekends. On Sunday 8 November no pair passed and every list was empty. The SMC Overview has no liquidity filter, so a pair's zones are always there.
• Missing days. The archive has no candle for 5 August, 11 September or 5 November 2026, when the feed stopped. The candles on either side sit next to each other, and a swing count of "five days" can span six calendar days there.

The newest day can also be unfinished: the feed runs about a day behind, and the newest stored day is topped up until it is complete, so its volume, and with it the liquidity test, can change during the day.

Smart Money Concepts are a way of reading a chart, popularised by the trader known as ICT. There is no official rulebook and no published evidence that they work better than other methods, on crypto or anywhere else; the pages apply one fixed reading so a row can be checked by hand. They describe what price did, not what it will do.

Open the page at Smart Money → Fair Value Gaps (switch the market to Crypto), or sign up free to open the Market Terminal.

The other crypto Smart Money guides: SMC Overview, Order Blocks, Breaker Blocks, Liquidity Sweeps. The NSE version of this page, with its own worked examples, is explained in the Fair Value Gaps guide. For what traders were doing in the futures market on the same days, see the OI Buildup guide and the Binance Futures guide.