ICT Concepts for Futures Trading: Liquidity Grabs and Displacement Explained

ICT, short for Inner Circle Trader, is a body of ideas associated with Michael Huddleston, and its vocabulary now shows up in most futures chat rooms. The terms describe patterns you can see: price spikes through an obvious level and snaps back, or moves away from a range in a hurry. The explanations attached to them are a different matter. Claims that large institutions are deliberately hunting your stop can't be confirmed from a chart or from public volume data. I'd treat them as a hypothesis and trade only what can be defined and tested.

Liquidity grab (sweep)

The idea is that stop orders and breakout entries rest beyond obvious levels like the prior day's high, a round number or a multi-day swing. Price trades through the level, fills those orders, and reverses back. In chart terms, a sweep is a bar that trades beyond a defined level and closes back on the original side within one or two bars.

Make it objective. Pick the level in advance, the prior session high for example. Define penetration as at least a certain number of ticks. Define the reversal as a close back inside the level within a set number of bars. Also record the volume on the wick bar and the bars around it. The source material claims volume spikes at the penetration and fades after, but that's something to measure, not assume. Real breakouts and sweeps can look identical on the way through.

A worked example (hypothetical)

The prior session high in ES is 5,250. A 5-minute bar trades to 5,254.50 and closes at 5,249. You wait for the first lower high on the 2-minute chart and short at 5,247. The stop goes above the wick at 5,255, so risk is 8 points. That's $400 on one ES contract or $40 on one MES. A target at 5,232 offers 15 points, roughly 1.9R. Commissions and slippage reduce that, and a stop order can fill beyond 5,255. If price trades back above the wick high, the sweep reading has failed and you're out.

Displacement

Displacement means a strong, directional move away from a range, with large-bodied candles and little overlap. You can define it numerically: for example, a move of at least X points within Y bars, or candles whose range is at least some multiple of the recent average. What it tells you is that price moved decisively. It doesn't tell you who was buying. Attributing it to a specific class of participant is a story, and the chart can't prove it.

Putting a sequence together

The tidy sequence goes like this: consolidation near an obvious level, a sweep in one direction, displacement in the other, then an entry on the first pullback with a stop beyond the sweep's extreme. When it works it gives a well-defined stop. But plenty of sweeps are just the start of a real breakout, and plenty of consolidations simply keep chopping. The way to find out how often each happens in your instrument is to count them.

Order blocks

An order block is the last down candle before a bullish displacement, or the last up candle before a bearish one. The premise is that the zone later attracts price again. Treat it as a candidate zone with a rule: a displacement of at least a stated number of points, say 20 to 30 as a starting parameter on the 15-minute or 30-minute chart, with the zone defined by the candle's range. Whether price respects it is a testable question. Don't assume that smaller zones are retail noise or that larger ones are institutional, because that can't be observed.

How to test it

1. Define the level, penetration size, reversal condition, trigger, stop and target in writing.
2. Mark every qualifying instance on historical charts, including the ones that continued through.
3. Record the result in R after estimated fees and slippage.
4. Split the sample into an earlier part and a later part, and check whether the result holds.
5. Run a small live size before scaling, since simulated fills don't validate exchange queue position or slippage.

Thirty examples don't prove an edge. If you want to turn your results into an expectancy figure, use the edge calculator.

How this relates to KPL

KPL levels are fixed reference prices anchored to the regular-session opening print, with four levels in the standard ES and NQ map. They are not institutional order-flow levels and they aren't predictions of where stops sit. You can use them alongside sweeps as one more reference, which is worth testing, and the KPL trading page explains how they're built. Treat any overlap as a hypothesis, not as confirmation of what large participants are doing.