Why Does Price Reverse Right After I Enter?

You enter. Price goes against you almost immediately, takes your stop by a few points, and then travels exactly where you thought it would — without you.

If that has happened enough times that you have started to wonder whether something is watching your orders, you are not imagining the pattern. It is real, it is not paranoia, and there is a mechanical explanation for it.

This page is about that mechanism — what is actually happening when it occurs. It is a different question from whether you are making mistakes, which I have written about separately in why your ICT trades keep failing. Here I want to explain the thing the market is doing, because understanding it changed how I traded more than any pattern ever did.

The two answers you have already been given, and why both are wrong

Ask this question anywhere and you get one of two responses.

"It's confirmation bias. You only remember the times it happened." This is the sensible-sounding answer and it is usually wrong. Confirmation bias is real, but it does not explain why the reversals cluster at specific prices — just above a prior high, just below yesterday's low, a few points past the round number. Random noise does not organise itself around the levels everyone can see. If your stop-outs were random they would be scattered, and they are not.

"Your broker is hunting your stops." This is the emotionally satisfying answer and it is usually wrong too. Here is the test that settles it: the same pattern happens on NQ and ES, which trade on a centralised exchange with a single public order book. No broker can move that price. If the behaviour occurs identically on an instrument where broker manipulation is structurally impossible, the broker is not the explanation.

The real answer is less personal than the second and less dismissive than the first, and once you see it, it stops feeling like an attack.

The problem nobody buying size can avoid

Start from the other side of the screen.

Suppose you need to buy a very large position. Not ten contracts — a size that matters. You cannot simply buy at market, because there is not enough resting supply at the current price to fill you. Push into the book and you move the price against yourself with every fill, and your average gets worse the more you buy. The larger the order, the worse the problem.

So what you need is a lot of people willing to sell to you at once, at a price you like.

Where does a concentration of sellers exist? Below an obvious low. Because sitting below that low are the stop-loss orders of everyone who is long, and a stop-loss on a long position is a sell order. It is a sell order that executes automatically, without hesitation, the instant price arrives.

That is the whole mechanism. A pool of stops is a pool of guaranteed counterparty. If you need to buy size, the cheapest way to get filled is to push price into the place where a few thousand automatic sell orders are waiting, absorb them all, and then let price go where it was always going.

Why your stop, specifically

This is the part that feels personal, and it is worth explaining precisely why it is not.

Nobody knows where your stop is. Nobody looked you up. What happens instead is that you and several thousand other people, reading the same chart with the same reasonable logic, arrive at almost the same conclusion about where the invalidation sits.

Where do you put a stop on a long? Below the recent swing low. So does everyone else, because it is the correct textbook answer. Where do you put it on a breakout entry? Below the breakout level. So does everyone. Round numbers, session highs, yesterday's low, the obvious trendline — all of it produces the same clustering.

You are not being targeted. You are being aggregated. Your order is one drop in a pool, and the pool is visible from a long way off because it forms in the same predictable places every single day.

What the sequence actually looks like

Once you know what to look for, the shape is recognisable and it repeats:

1. A range forms. Price moves sideways for a while, building an obvious high and an obvious low. This is where the orders accumulate.
2. Price breaks one side. It looks like a breakout. Breakout traders enter in that direction, and everyone positioned the other way gets stopped out.
3. The move fails immediately. Price re-enters the range within minutes. This is the tell — a genuine breakout does not usually come straight back.
4. Price delivers the other way with real displacement, toward the pool on the opposite side.

Step two and three together are what ICT calls a liquidity sweep, and when it happens at a session open it is the Judas swing. Richard Wyckoff described the same event in the 1930s and called it a Spring — more on that in ICT vs Wyckoff if you want the history.

Your stopped-out trade was step two. The move you correctly predicted was step four.

The same thing happens in both directions

Worth stating explicitly, because people usually only notice it happening to them one way.

Everything above described a long getting stopped below a low. The mirror image is just as common and slightly harder to spot: you short a breakdown, price spikes up through the prior high, takes your stop, and then falls exactly as you expected.

The mechanism is identical with the signs flipped. A large sell order has the same filling problem — it needs buyers — and a pool of buyers exists above an obvious high, because the stops of everyone who is short are buy orders. Push price up into them, absorb the buying, then sell into the market you have just created.

ICT names these two pools buy-side and sell-side liquidity, and the naming is worth getting the right way round because it trips almost everyone up at first. Buy-side liquidity sits above price, not below, because it is made of buy orders. Sell-side sits below. The label describes the orders resting there, not the direction price will travel.

Once you can read the chart this way, the question you ask before entering changes. It stops being "which way is this going" and becomes "which pool has not been taken yet" — because the untaken one is usually where price is being delivered, and it is the one you do not want to be sitting in front of.

That reframe is what ICT calls the draw on liquidity, and it is the single most useful habit to take from this page. Mark both pools before the session. If price is reaching for the one above, you do not want to be short underneath it, however good the setup looks.

The boring explanations, which are sometimes the real one

Before attributing everything to liquidity mechanics, it is worth ruling out the unglamorous causes. In my experience a meaningful share of "the market took my stop" is actually one of these.

Your stop was inside normal noise. If ES routinely swings eight points in a five-minute window and your stop was six points away, it was never a stop. It was a coin flip with extra steps. Check the average range of the session you trade before deciding the market is out to get you.

Spread and slippage on the wick. On forex and CFDs the price your broker fills you at includes the spread, which widens at exactly the moments this happens. Your chart shows the low at 1.0798; your stop triggered at the bid, which was two pips lower. The candle can genuinely not reach your level while your stop still fills.

Feed differences. Two brokers can print slightly different highs and lows on the same forex pair, because there is no single central price. A wick that took your stop may not appear on someone else's chart at all. This one does not apply to futures, where there is one exchange and one price.

You moved it. Worth being honest with yourself about how often the stop that got hit was where the plan put it, versus where anxiety put it after the trade went slightly wrong.

What to actually do about it

Three changes, in order of how much they helped me.

1. Stop putting your stop where everyone else puts theirs

Not further away arbitrarily — that just costs more when you are wrong. The change is to place it beyond the level that invalidates the idea, with room for the wick, rather than at the nearest tight point that makes the risk-to-reward look good.

If your reason for being long is that price swept a low and reversed, then the trade is wrong only if price goes back below that sweep. That is where the stop belongs. A stop above it is not protecting your idea, it is protecting your ratio, and the market does not care about your ratio.

2. Wait for the sweep instead of being in front of it

This is the actual solution and it is what the whole ICT framework is organised around. If pools of stops get taken before the real move, then the sequence you want is: let the pool get taken, wait for the reversal to confirm, and enter after.

You give up the best price. In exchange you stop being the fuel. Entering before the sweep means being on the wrong side of the exact event that is about to happen, which is why a correct directional read keeps producing losing trades.

What "confirm" means specifically: a market structure shift on a body close, then an entry from the array the displacement leaves behind — an order block or a fair value gap. That is the 2022 model in one sentence, and it exists precisely to solve this problem.

3. Notice when it is happening

These sweeps are not evenly distributed through the day. They cluster at session opens and at the start of the kill zones — the London open, the 9:30 cash open — because that is when there is enough participation to make absorbing a pool worthwhile.

If you keep getting stopped between 09:30 and 09:50 and then watching the move leave without you, that is not bad luck. That is the manipulation phase, on schedule, and the answer is to treat that window as the setup forming rather than the trade to take.

What I am not saying

Two clarifications, because this topic attracts overstatement.

This is not a conspiracy. Nobody is coordinating against retail traders. What is happening is that large orders have a filling problem, resting stops solve it, and stops accumulate in predictable places. That is a structural feature of a market with a visible order book, not a plot.

Understanding it does not make you profitable. Knowing why you got stopped out removes a source of frustration and points you at a better stop placement. It does not give you an edge on its own, and plenty of people who can explain this perfectly still lose money. The mechanism is the beginning of the answer, not the whole of it.

Frequently Asked Questions