What Is a Breakout in Trading? Definition, a Real QQQ Example and 27 Years of Data

Michael Whitman · Trading since 2017, full-time since 2020 · YouTube creator, 137K subscribers · Published 2026-09-05 · Updated 2026-11-02 · Reviewed 2026-11-01

Covered on this page: TradingView.

A breakout is price closing beyond a level that had been holding it in: the top of a range, a resistance line, or the highest high of the last few weeks. A close below support is the same event pointing down, usually called a breakdown. Whether the break lasts is the whole question, and on QQQ’s daily chart it lasted about as often as it did not.

That second sentence is the part most definitions leave out. This page gives the definition, walks through two real QQQ breaks from September and October 2026, one that held and one that did not, and then measures every break of a 20-session high on QQQ since 1999.

How it works: orders on both sides of the line

A level is a price where orders have piled up. Sellers who defended the top of a range kept price under it; traders short from inside the range keep their stops just above it; others wait above it with buy orders for the break itself.

A close through lets all three act at once, in the same short stretch, which is why a break often moves faster than anything inside the range did.

The same mechanism explains the failure. If not much was resting above the level, the push runs out and price drifts back in. Liquidity covers that argument from the side of the orders, and false breakout is the name for the break that returns.

What counts as a breakout

A close beyond the level, not a touch. A candle that trades above the line and finishes below it has tested the level and been turned away. That is closer to the opposite of a break than a weak version of one.

There has to be something to break. Price grinding to a new high with no ceiling above it is a trend, not a breakout.

So the level must have held price first: a range with a clear top, a support or resistance line price kept stopping at, or the edge of a narrowing pattern such as a triangle or an inside bar.

A mechanical version makes it testable. For the data on this page, the level is the highest high of the previous 20 sessions, and a breakout is the first close above it. Twenty sessions is about one trading month and the same window the Turtle rules used for entries.

A worked example: two QQQ breaks, autumn 2026

The range. From late August to mid-September 2026, QQQ traded under a high of $724.13, set on 28 Aug, and above a low of $700.00 on 16 Sep. That range was 3.33% of the top, tight by QQQ’s standards. On 18 Sep QQQ closed at $721.45, still under the line.

The break. On 21 Sep 2026 QQQ opened at $727.89, already above $724.13, and closed at $741.47, 2.39% above the level. The next session closed higher again, at $747.46, which is the second close the rule at the top of this page asks for.

It held for the five sessions this page counts. The lowest close from 22 to 28 Sep was $736.53, still $12.40 above the old top. By that test, the break held.

The second break did not. The 22 Sep high of $748.35 became the new 20-session high. On 2 Oct QQQ closed at $749.58, only 0.16% above it, and on 5 Oct closed higher at $756.20, so this one was confirmed too. On 8 Oct QQQ closed at $747.58, back under $748.35: a false breakout by the same rule.

And the first break’s gains were gone a month later. Twenty sessions after 21 Sep, on 19 Oct, QQQ closed at $718.36, 3.12% below the breakout close and back under the $724.13 level. It closed at $690.49 on 30 Oct. Holding for five sessions said nothing about the next month.

Does the next candle confirm it?

The rule in Michael’s note above asks for one more close in the same direction. On QQQ, 288 of the 480 breaks got one: the session after the break closed higher than the break day. Those confirmed breaks still closed back below the level within the next six sessions 106 times (36.81%).

The 192 breaks without that second close did much worse: 150 of them (78.13%) were back under the level within six sessions, and 92 were back under on the very next close. Part of that gap is built into the split, because a lower second close sits nearer the level by definition. It still shows what the rule filters out.

The price of waiting was small. The confirming close sat a median 0.46% above the break close, which is the amount given up by entering a session later. The how to trade a breakout page sets out the entry and stop choices step by step.

Tight ranges, wide ranges and volume

Breaks out of a tight range slipped back less often. The 480 breaks were split into three equal groups by the height of the 20-session range before them. The tightest third, ranges under 5.69% of the level, closed back below within five sessions 45.63% of the time; the middle third 48.13%; the widest third, ranges of 8.42% or more, 58.13%.

That gap, 12.5 percentage points on 160 breaks each, is larger than chance usually gives (a two-proportion test gives z of about 2.2), but it is one fund, one window and one of several splits tried here, so it is a lead rather than a rule.

Volume barely told the groups apart. Only 22 breaks came on volume 1.5 times the 20-session average or more, and 10 of them (45.45%) failed, against 233 of 458 (50.87%) on lighter volume. Twenty-two is too few to read anything into. The volume page explains why a big bar shows activity, not direction.

The original data

The sample is QQQ’s daily bars from 10 Mar 1999 to 30 Oct 2026, 6,955 sessions (Yahoo Finance, split-adjusted, bars with no range dropped). A breakout is the first close above the highest high of the previous 20 sessions, after a close that was not above its own 20-session high. There were 480. The median break closed 0.41% above its level.

Against an ordinary day, that failure rate is not high. From every session since 1999, QQQ closed lower within five sessions by at least the margin a break sat above its level (each of the 480 margins tried on every session) 57.67% of the time. The breaks came back less often than ordinary day-to-day moves would, not more. It says nothing about which break will hold.

The month after looked like any month. The median 20-session return after a break was +1.54%, against +1.57% after any session, and the share of higher closes, 62.08% against 62.24%, is inside the noise. QQQ rose over most of this period, which lifts both numbers equally.

A second market gives the same picture. The false breakout page ran the same 20-session test on SPY and found 55.06% of 623 breaks failed within five sessions. Every QQQ break, with its range height, volume ratio and outcomes, is in the QQQ breakout file.

No trading rule was tested. The counts describe what QQQ did after a mechanical definition of a break; they are not a forecast and do not include any stop, target or cost.

Breakout videos are popular and thin on failure. Across our study of 24,971 trading videos, 396 have breakout in the title, with a median of 6,160 views. Of the 113 with description text, 11 mention false, fake or failed breakouts.

When it fails

The false breakout

Half of all breaks on this definition came back. At the break close, the held and the failed cases looked alike: the 2 Oct example cleared its level by 0.16% and got its confirming close, then failed four sessions later. Breakout vs false breakout looks at whether anything on the break day separated the two.

Entering on the wick

A spike through the level that closes back inside is a test, not a break. Waiting for the close removes those cases at the cost of part of the move.

No range to break

A new high in a steady climb has nothing above it to clear. The word only means something against a level that had been holding price.

Reading it afterward

Every real breakout looks obvious on a finished chart, and so does every false one. At the close of the break day they are the same picture, which is why the stop belongs back inside the range and risk management sets the size before the order goes in.

Questions people ask

What is a breakout in simple terms?

Price closing beyond a level it had been stuck under or above, such as the top of a range. On QQQ’s daily chart, a close above the highest high of the last 20 sessions happened 480 times from 1999 to October 2026.

How do you know if a breakout is real?

Not at the break itself. A second close in the same direction and a tight range before the break were both followed by fewer returns below the level on QQQ, but 106 of 288 confirmed breaks still came back within six sessions.

What is the difference between a breakout and a breakdown?

Direction only. A breakout closes above resistance; a breakdown closes below support. The mechanics, and the risk of a false move, are the same.

Does volume confirm a breakout?

Volume shows how much traded on the break, not whether it lasts. On QQQ only 22 breaks came on 1.5 times the average volume, too few to measure a difference.

Related

Support and resistance is where the level comes from, and a breakout cannot exist without one. False breakout measures the return trip on SPY. Volume covers the one input on the chart not calculated from price. A very wide bar told little about the next two weeks on SPY in range expansion. And most chart patterns end in a break through one line, so they share every failure mode described here.

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.