Price spends the whole London morning grinding under 1.0920. Every push into it gets sold within a few pips. Then one candle closes above, the buy orders arrive, and twenty minutes later the pair is trading back inside the range with those stops already filled. Nothing about that sequence is unusual. It is the default outcome when a clean level is touched by a market that has no reason to go anywhere.
A break has to cost someone money
A level matters because orders sit around it. Stops from short positions rest above a range high. Buy stops from breakout traders rest in the same place. Both are the same instruction to the order book: buy at market the moment this price prints. That cluster is fuel, and it burns once.
What decides the next hour is whether anything arrives after the fuel is gone. A real break has a second source of demand: a large order that could not be filled inside the range, a repricing driven by rates or data, a session opening with participants who were not in the market when the range formed. A liquidity grab has none of that. The stops fill, whoever wanted size gets a decent average price, and the market returns to the middle of the range, because that is still where both sides agree on value.
So the question is never whether the candle closed above the line. It is whether anybody kept paying up after the easy fills were taken.
Where the sweeps happen, and why
Sweeps are not random. They happen where the stop orders are visible from a mile away, and every trader draws the same lines: the high of the previous day, the boundary of the Asian range at the London open, a round number like 1.1000 or 2400 on gold, and any pair of equal highs sitting side by side on the chart. Two touches at exactly the same price is an advertisement. It tells everyone reading the chart where the shorts have hidden their protection.
The counterintuitive part is that the cleaner the level looks, the more likely the first move through it is a sweep. A messy level with overlapping wicks has stops spread over thirty pips, which makes it expensive to hunt. A razor sharp double top has them stacked in five. That is why the levels people rate highest on their support and resistance maps are so often the ones that punish the entry.
Timing matters as much as location. A break at 06:40 UTC, in the last twenty minutes of a dead Asian session, has almost no chance of being funded by real flow. The same break twelve minutes after the London open is a different event, because the participants who can carry it are actually at their desks. Knowing the rhythm of the trading sessions removes a third of the bad setups by itself.
Four things worth checking before you take one
- How price arrived. If the market ran 90 pips into the level in a straight line, the break is late and the people who will take profit are already onside. Breaks that come after a tight consolidation directly under the level are the ones with room left.
- What made the level. A high printed once during a news spike is thin. A boundary that held four separate attempts over two days has real positions behind it, and a genuine break through it forces those positions to be covered.
- Where the next opposing pocket sits. If a heavy supply zone starts 15 pips above the break, the trade has nowhere to go and the reward side of the calculation collapses before you enter.
- What is on the calendar. Breaks that occur in the ninety seconds around a data release are execution lotteries. Spreads widen, fills slip, and the level stops being the reason price moved. That is a different trade entirely, and it should be planned as one.
Break, retest, or nothing
There are three honest ways to enter and each pays for something different. Buying the close of the break candle gets you in on every real move and also on every sweep, at the worst available price, with the widest stop. Waiting for the retest of the broken level gives a cheaper entry and a tighter stop, and it filters out most sweeps, because a sweep does not come back and treat the level as support. The cost is real: strong breaks funded by an order that has to get done frequently never retest, and you watch those from the sidelines.
The third option is to mark the level on H4 or the daily and drop to M5 only to time the entry, taking the first higher low that forms above the broken boundary. It converts a guess into a structural condition, and it usually halves the stop distance. The trade off between timeframes is the same one every time: precision on the small chart, meaning on the large one.
Whichever you pick, pick one and keep it for a hundred trades. Switching entry style after two losses is how traders end up with a record they cannot learn anything from.
The stop goes where the idea dies
A breakout stop belongs below the base that produced the break, not at a round twenty pips because that is what fits the position size you wanted. If the correct stop is 45 pips away and 45 pips of risk is too much money, the answer is a smaller position, not a closer stop. Sizing rules exist precisely for this moment, and the arithmetic is set out in the risk management rules.
Two structural mistakes cost more than bad entries. The first is placing the stop exactly at the level, where the retest is designed to reach. The second is scaling in after the first stop is hit, on the theory that the second attempt must work. Second attempts do work more often than first ones, which is exactly why they deserve their own planned entry rather than an improvised average down.
Breakout signals cluster. When EURUSD, GBPUSD and the dollar index all break at the same minute, that is one position expressed three ways, and the risk on it is three times what the position sizing suggests. Firms running their own risk desks watch for this pattern in client books because it is where correlated blowups start.
Leveraged trading carries a high risk of loss, and breakout entries concentrate that risk into the noisiest few minutes of the day. The traders who make the style work are not the ones with the best level detection. They are the ones who pass on most of what they find, and who accept that a plan built on a defined reward to risk ratio will still spend long stretches being wrong.
"I stopped trading the candle that breaks the level and started trading what price does in the twenty minutes after it. Most of my old entries would never have been taken, and I do not miss them."
— Alex Onta, Executive Director, SINGUARD
Key Takeaways
- A break only continues if buying arrives after the resting stop orders have been consumed, so watch the twenty minutes after the break, not the break candle.
- The tidier the level, the tighter the stops behind it and the cheaper it is to sweep, which is why perfect double tops fail so often.
- Time of day filters more bad breaks than any indicator: a break into a dead session rarely has the flow to carry it.
- Set the stop from structure and then size the position to fit it, and treat simultaneous breaks on correlated pairs as one trade.
Frequently Asked Questions
What is a liquidity grab?
A liquidity grab is a move that pushes just far enough past an obvious level to trigger the stop orders resting there, then returns inside the prior range. The stops provide the fills that a larger participant wanted, and once they are gone there is no further buying or selling to carry price onward. On the chart it usually looks like a single candle with a long wick and a close back inside the range.
Should I wait for a retest before entering a breakout?
Waiting for a retest gives a tighter stop and a cheaper entry, and it filters out sweeps that never come back to the level as support. The cost is that strong breaks driven by news or a session open often do not retest at all, so you miss them. Many traders split the difference by taking a smaller position on the break and adding on the retest if it appears.
Which timeframe works best for breakout trading?
Levels drawn on H4 and daily charts tend to hold more resting orders than levels drawn on M5, so the break carries more meaning. A common approach is to mark the level on the higher timeframe and use M5 or M15 only to time the entry and place the stop. Trading breaks of small intraday levels produces far more false signals, and all breakout trading carries a high risk of loss.