Imbalance Candle: The 3-Candle Gap Price Left Behind
An imbalance is a run of candles where price moved so quickly in one direction that a band of prices was skipped over rather than traded through. It is measured as the gap between the first candle's wick and the third candle's wick, and it marks where trading was one-sided rather than two-sided.
Covered on this page: TradingView.
How it works
In an orderly market, price moves through each level and trades there. Buyers and sellers meet at every price on the way, and the chart shows overlapping candles — each one covering ground the previous one also covered.
An imbalance is what happens when that stops. One side overwhelms the other for a few bars, price travels a long way quickly, and a band of prices is crossed without any real two-sided trading occurring inside it.
The chart records this as a gap between wicks. Take three consecutive candles. If the first candle’s high sits below the third candle’s low, the space between them was never traded through during those bars — that space is the imbalance.
The middle candle is doing the work. It is the large one — the bar that covered so much ground the candles either side of it do not overlap. Without that displacement there is no gap and no imbalance.
The measurement, exactly
Up move: the top of the imbalance is the third candle’s low. The bottom is the first candle’s high. Everything between them is the band.
Down move: the reverse. The bottom is the third candle’s high, the top is the first candle’s low.
That is the entire calculation. No settings, no lookback period, no smoothing. Two prices read off two candles, which is why two people drawing it on the same chart should produce the same box — and why it is worth insisting on the precise definition rather than eyeballing “a fast move.”
The same measurement is called a fair value gap in most smart-money material. They are the same three candles and the same two prices; the vocabulary differs by teacher rather than by method.
This is the part most explanations get backwards. An imbalance does not mark where large orders were placed. It marks where orders were absent — where there was not enough resting interest to slow the move down. It is a photograph of a vacuum, not of a wall.
Which matters for what you can conclude from it. A gap tells you liquidity was thin through that band during those three bars. It does not tell you anyone intends to trade there again.
In practice: the claim nobody puts a number on
The standard teaching is that price returns to fill the gap. Sometimes it does. The honest position is that the return rate depends entirely on the instrument, the timeframe, how far back you look and how long you are willing to wait — and that almost nobody teaching it has counted.
Ask the question precisely and it becomes answerable. Of imbalances formed on this instrument on this timeframe over this period, what share saw price trade back into the band within twenty bars? That is a measurable figure. “Price fills imbalances” is not.
The reason it feels reliable is selection. Look at any chart and the filled ones are obvious, because price passed through them and they are visible in hindsight. The unfilled ones sit above and below the current price and get no attention at all.
Volume inside the band is low by construction. That is not a signal, it is the definition — an imbalance is a region where little traded. Presenting the low volume as confirmation is circular.
Timeframe decides how many exist. On a one-minute chart, imbalances appear constantly — any brisk thirty seconds produces one. On a daily chart there are very few, and each represents a genuine repricing rather than a moment of thin liquidity.
Which is the practical filter. An imbalance that survives on a higher timeframe is a small number of genuinely significant bands. One drawn on a fast chart is one of dozens, and treating them as equally meaningful is the mistake that makes the concept unusable.
Every attempt pays the spread and commission of a round trip. A one-minute chart offering fifteen imbalances a session is offering fifteen round trips, and the cost of testing them all exceeds what most of them are worth.
What an imbalance is not
It is not evidence of institutional activity. A fast move with thin liquidity behind it can be produced by a news headline, a thin book at an odd hour, or one impatient participant. Nothing about the gap identifies who caused it.
It is not a level. It is a band with a top and a bottom, and treating it as a line loses the only useful part — the width, which tells you how far price traveled without resistance.
It is not obliged to fill. Unfilled imbalances persist for months and years. “Eventually” is not a timeframe you can position around.
And it is not the same as an opening gap. An opening gap happens between sessions, when the market was closed. An imbalance happens during continuous trading, when the market was open and simply moved too fast. Different causes, different behavior.
How this connects
An imbalance is a range price moved through without trading much, which is the same object a fair value gap describes in the smart money concepts vocabulary.
The claim is that price returns to it, on the reasoning that the market revisits prices it skipped.
An order block is the usual companion mark - the candle before the move that created the imbalance.
Liquidity is the wider idea underneath. Both concepts are really about where resting orders are and are not.
And the base rate is what settles it. In markets that change direction as often as most do, price returning to a nearby level is ordinary, so “many imbalances fill” is not yet evidence of anything.
When it fails
The failure that costs money is waiting for a fill that does not arrive. Price moves away, the band sits there, and the position waiting for the return is a position not taken elsewhere. There is no mechanism that requires price to come back.
The second failure is drawing them everywhere. A chart marked with every imbalance on a fast timeframe has a box on nearly every screen, and a level that exists everywhere is a level that predicts nothing.
A third is entering at the edge rather than through it. Price frequently trades into a band and straight out the other side. Entering at the first touch assumes the band holds, which is the thing being tested.
A fourth is ignoring what caused it. An imbalance formed on a scheduled announcement is a repricing around new information. One formed at 3 a.m. on no news is a thin book. Both look identical and they behave differently.
And a fifth is stacking it with everything else. Imbalance plus order block plus liquidity sweep plus a moving average is four ways of drawing the same region and calling it confluence. Adding correlated tools does not add evidence.
The original data
Of the 24,971 videos measured for this site, imbalance appears almost entirely inside smart-money concepts material rather than as a standalone subject — and in that material it is nearly always presented with a fill claim and without a fill rate.
The one number this page can give you is the cost of finding out: a round trip per attempt, so fifteen attempts a session across 250 sessions is 3,750 round trips a year. The number it deliberately does not give you is a fill rate, because that figure depends on the instrument and timeframe you actually trade, and quoting somebody else’s would be worse than quoting none. Counting it on your own chart takes an afternoon and settles the question permanently — which is more than any article about imbalances will do for you.
Related
Fair value gap is the same measurement under the more common name. Order block is the level most often drawn alongside it, and the two frequently mark the same region. And liquidity is what was missing from the band in the first place.
I do not mark a gap on a chart unless I am willing to count what happens next. Every imbalance you mark is real in the sense that the gap exists, and that says nothing about what fraction of them price comes back to. Until you have that number, the boxes are decoration.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money. Some links on this page earn a commission if you buy through them. It costs you nothing and it does not decide what appears here or in what order — how these pages are made is set out in our methodology.