Breakout or Fakeout: How to Tell Before It Costs You
The candle that starts a new trend and the candle that traps everyone look identical at the moment they print. The difference shows up over the next three bars, and you can wait for it.
By Pavel Penev, MScFounder, TradeWize · 10+ years trading the marketsThe short answer
A breakout is a close beyond a level that holds. A fakeout is a close beyond a level, or a wick through it, that comes back inside within a few bars. You can't tell them apart on the breakout candle. You can tell them apart three bars later, by whether price is still on the far side of the level and whether a return to it held.
Every trader has bought a breakout that reversed on them within the hour. It's the most common losing trade there is, and it isn't a sign you did anything wrong. A real breakout and a fake one are the same candle. They only become different afterwards.
So the skill isn't spotting a breakout. It's knowing how long to wait, and what to wait for, before you believe one. This piece gives you the three checks, shows both outcomes on charts read by our analyzer, and ends with the rule that removes most of the traps.
What a fakeout looks like
Here's a chart that spent months under one ceiling, finally closed above it, and then gave the whole thing back. The candles are synthetic, built for this piece. The read is computed by the engine behind our chart analyzer, the same way it reads a screenshot.
What the tool reads on this chart
EXAMPLE closed above $108.02 21 days ago and fell back below it 16 days ago. A failed break. A close back above $108.02 would change that.
- Decides above:
- $108.02 (resistance)
- Decides below:
- $91.43 (support)
- Level history:
- The last three times price reached $108.02 it held every time, bouncing 7.5% on average.
- Level history:
- The last three times price reached $91.43 it held every time, bouncing 15% on average.
A synthetic chart, built for this article and read by the same engine that reads a screenshot on /analyze. No AI wrote this read; it is computed from the candles.
The headline is the engine's whole verdict. It closed above the level, it fell back below it, and the read names that a failed break. The level's history line says the ceiling has now held every time it was tested, including this one.
Read the headline. Closed above the level, then closed back below it a few days later. A failed break. The engine doesn't need a name for it beyond that, and neither do you. The level held. The people who bought the close above it are underwater, and the chart is back inside the range with a ceiling that just got stronger.
The tell was there on the day after the break: the follow-through never came. One close above, then nothing. A breakout that's real usually keeps going, or at worst holds its ground. One that stalls on the first day is asking to be sold.
What a real one looks like
Now the other outcome. Same kind of range, same kind of ceiling, but this time the break holds, comes back to the level, and bounces off it from the other side.
What the tool reads on this chart
EXAMPLE broke out of a cup and handle 31 days ago on heavier than normal volume, then retested $108.71 13 days ago and held. Below $108.71 the break has failed.
- Formation state:
- Broke above $108.71 31 days ago on heavier than normal volume, came back to it 13 days ago on light volume and held.
- Decides above:
- $119.10 (the measured move)
- Decides below:
- $108.71 (the breakout)
A synthetic chart, built for this article and read by the same engine that reads a screenshot on /analyze. No AI wrote this read; it is computed from the candles.
The engine reads the base as a cup and handle, marks the breakout bar, and reports the retest: price came back to the breakout level on light volume and held. The lower decision price is the breakout level itself; below it, the break has failed.
Three things in that read are the three checks. The break came on heavier than normal volume. Price came back to the level and held it, on light volume. And the level below which the break has failed is printed as a price, so you know exactly what would change the verdict.
Check one: the close, not the wick
A wick through a level means someone tried to push through and got rejected before the bar ended. That's not a breakout. It's the opposite: the level absorbed the attempt. Only a close beyond the level counts, and on a daily chart that means waiting until the day is over.
This costs you something. Waiting for the close means a worse price if the break is real. That's the trade you're making, and it's a good one, because the wick-only pokes are the single biggest source of false breaks and the close filters every one of them out.
Learn it by doing
Reading about it is one thing — it clicks when you do it. Learn it hands-on with free, interactive lessons on TradeWize.
Try the free lesson →Check two: the volume, against the chart's own average
A level breaks when one side runs out of orders. If that's genuinely happened, the break should come with more trading than usual, because the buyers who were waiting for it all act at once. A break on volume well above the chart's average has that behind it. A break on ordinary volume has nobody behind it, and those are the ones that fail.
Measure against the chart's own average, not an absolute number. The engine calls a break's volume unconvincing, ordinary or strong by comparing the break bar to the twenty bars before it. Two times the average is a real signal. One point one times is noise.
Check three: what happens next
This is the one that decides it, and it's the one that requires patience. After a real break, price either keeps going or comes back to the level and holds it from the other side. That return is the retest, and a retest that holds is the strongest confirmation there is. We've written a whole piece on why the retest is where the entry usually goes.
After a fake break, price comes back through the level and closes on the original side. Once that's happened, the break is over. Don't wait for it to try again. The level just proved it's still there.
| Check | Real breakout | Fakeout |
|---|---|---|
| The close | Closes beyond the level, and the next bar doesn't close back inside | Wicks through only, or closes through and the next bar or two close back inside |
| The volume | Well above the chart's average on the break bar | Ordinary, or below average |
| What happens next | Follow-through, or a return to the level that holds | A return to the level that cuts straight through it |
No single check is enough on its own. A break that passes all three has done everything a break can do to earn your trust.
Why fake breaks happen at all
It isn't bad luck. A level is where the orders sit. Stop-losses cluster just past it, and breakout orders sit just beyond it, and both kinds become market orders the moment price touches them. So a push through a level fires a burst of buying that looks exactly like a real breakout for a bar or two, and if there's nobody behind that burst, price sags straight back. Our piece on stop-loss hunting covers the mechanism and the research behind it.
The useful consequence: a fakeout is information. A level that just survived a break attempt has been tested harder than one that hasn't, and the trapped traders on the wrong side of it will fuel the move back the other way.
The rule that removes most of the traps
Don't buy the break. Buy the retest, or the second close. If price closes above resistance, wait. Either it comes back to the level and holds, and you enter there with a tight stop just under it, or it closes above the level again on the next bar and you enter with the first close as your line in the sand. You'll miss the fastest breakouts. You'll also miss almost every fakeout, and the second matters more than the first.
The one-line version
A break is a close, not a wick. A real one has volume behind it and holds when price comes back. If it comes back through the level, it was never a break.
What is a fakeout in trading?
A false breakout: price closes through a support or resistance level, or pokes past it on a wick, and then comes back inside within a few bars. The traders who acted on the break are trapped, and the level has held. A fakeout above resistance is a bull trap; one below support is a bear trap.
How do you tell a real breakout from a fake one?
Three checks. The break must be a close beyond the level, not a wick. The break bar should carry volume well above the chart's average. And over the next few bars, price should either keep going or return to the level and hold it. A break that fails any of the three is suspect; one that fails the third is over.
Should you wait for a close before trading a breakout?
Yes. A wick through a level is a rejection, not a break. Waiting for the close costs a little price on a real break and saves you from almost every wick-only fakeout, which is the single most common kind.
Does volume confirm a breakout?
It helps a lot. A break on volume two or three times the chart's recent average has real participation behind it. A break on ordinary volume has nobody behind it and is far more likely to fail. Measure it against the chart's own average, never an absolute number.
What is a retest after a breakout?
Price coming back to the level it just broke and holding it from the other side, so old resistance acts as new support. A retest that holds is the strongest confirmation a break can get, and it's where many traders place the entry, because the stop can sit just under the level.
Can an AI tell a breakout from a fakeout?
It can tell you what has happened so far, which is the honest version. Our free chart analyzer reports whether price closed through a level, whether it came back and held or failed, and how the break's volume compared to the chart's average. It doesn't predict which way the next bar goes; it shows you the facts the three checks are built on.
Is your chart's break real?
Upload the screenshot. The chart analyzer reads the real candles and tells you whether the level was closed through, whether the break's volume was anything special, and whether the retest held or failed. Free, no account, one read a week. It reports the facts; the call is yours.