Breakout vs False Breakout
Michael Whitman
·
Trading since 2017, full-time since 2020
·
Published 2026-09-08
·
Updated
·
Reviewed
A breakout is a close through a level that stays through it, while a false breakout is a close through a level that soon returns. They are the same event on the break day; the name depends on the sessions that follow.
A breakout and a false breakout start as the same bar. Only later does one of them come back. The fair question is whether anything visible on the break day hints at which ending is coming, and on SPY’s downside breaks the answer was no.
What each one is
A breakout is a close through a level that stays through. Price ends the session beyond the line and the next few closes do not return. Breakout covers the entry.
A false breakout is a close through a level that soon returns. The line is cleared and then recaptured within a few sessions. False breakout covers the upside version in detail, including how the size of the clearance relates to failure.
The label is assigned later. On the break day there is one event; the name is decided by the sessions that follow it.
Where they differ
Only in the sessions that follow. A held break keeps its closes beyond the line, while a false one prints a close back inside. Nothing else is part of either definition, and neither one can be named on the day itself.
In how quickly the answer arrives. On SPY’s downside breaks, 53.55% of the failures closed back above the level on the very next session. A failure is often known within a day; a hold is only known once the window has passed.
In what each costs. Both pay a round trip. A held break keeps a position, whereas a failed one pays that round trip and usually exits near the level, so the loss is set by where the stop was, not by the pattern.
What the break day could not tell apart
Volume was high on both kinds. On the break session, volume ran a median 1.50 times the median of the previous 20 sessions for the breaks that failed and 1.63 times for the ones that held. The difference is inside chance (Mann-Whitney p = 0.151).
The close sat near the low on both. Failed breaks closed a median 7.7% of the way up the day’s range and held ones 9.9% (p = 0.592). A close pinned to the low did not mean the break would hold.
Both were wide days. The break session’s range ran 1.382 times the prior average true range for the failures and 1.544 times for the holds (p = 0.081), again within chance.
Heavy volume did not rescue the rate much. Breaks on at least 1.5 times normal volume failed 63.83% of the time (141 breaks); the rest failed 73.81% of the time (126 breaks). Fisher p = 0.087, so even this gap could be chance.
Which one to use: Breakout or False Breakout?
Trade the break only when the size is set for the failure case. On SPY’s downside breaks the failure was the more common ending, and the break day gave no reliable sign of which one was coming.
Wait a session when you can accept a worse price. Over half the failures showed themselves on the next close, so one session of patience removes many of them and costs the move on the ones that held.
Fade the break only with its own rule. A failed break is a separate trade with a separate stop, not a reason to flip a losing position on the spot.
And when someone shows you a filter that sorts them on the day, ask for the count. Volume, range and close location did not do it here.
Two breaks that looked alike: SPY, 2025 and 2026
A convincing break that failed. On 27 February 2025 SPY closed at $585.05, below the $589.56 low of the previous 20 sessions, on 1.94 times normal volume, with the close almost at the day’s $584.65 low and a range 2.05 times the prior average true range. The next session closed at $594.18, back above $589.56.
A quieter break that held. On 19 March 2026 SPY closed at $659.80, below a 20-session low of $661.19, on 1.31 times normal volume, closing 59% of the way up a range smaller than its average. It did not close back above the level in the next five sessions.
The lesson is not the reverse. Two examples prove nothing on their own, either way; the 267 breaks above show no reliable pattern either way.
The original data
The sample. SPY daily data from Yahoo Finance with share volume included, 29 January 1993 to 2 October 2026. The first qualifying break fell on 24 February 1994, and the last one with five sessions still to come on 16 September 2026.
The rule. The level is the lowest low of the previous 20 sessions. A break is a close below it after a close at or above the previous session’s level. It is false when any of the next five closes is back above the level. This is a downside rule on purpose; the site’s false breakout page measures upside breaks of a 20-day high, including how failures compare with ordinary days, and this page does not repeat that comparison.
The features. Volume over the median volume of the previous 20 sessions; close location as close minus low over high minus low; range over the prior session’s 14-day average true range (Wilder). Groups compared with a two-sided Mann-Whitney test. Of the 267 breaks, 183 (68.54%) were false by this rule and 84 held; 53.55% of the failures came back on the very next close. All 267 breaks, with each feature and the outcome, are in the SPY 20-session-low breaks file.
When it fails
The failure is sizing up the break that looks strongest, and it puts the most money on a coin the data says is no better. A heavy-volume break that closes on its low feels like conviction, so it gets a larger position. On SPY those features appeared on the failures about as often as on the holds, so the larger size lands on a sample that fails at roughly the same rate as the rest. The loss on the failures grows; the edge it was paying for never showed up in the count.
The second failure is a stop exactly at the level. The next close back inside is the most common outcome.
A third is drawing the level after the break. Every break then looks deliberate.
A fourth is judging a break filter on a handful of trades. With 267 breaks in three decades, a few examples can point anywhere.
And a fifth is reversing on every failure. That is a second trade and needs its own rule.
Related
Breakout covers the entry through a level. False breakout measures the upside failures and the margin that mattered. And support and resistance covers drawing the level before price reaches it.
Decide how much a failed break may cost you before the break happens. On the day it prints, the failures and the holds wear the same clothes.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.