Break of Structure vs Support and Resistance

Michael Whitman · Trading since 2017, full-time since 2020 · Published 2026-09-09 · Updated 2026-10-19 · Reviewed 2026-10-04

One of these is a price and the other is a pattern in a sequence of prices. They get spoken about in the same breath, and the place they meet is the old swing level itself: once it breaks, is it now support? On thirty-three years of SPY daily bars, that idea did not show up in the data.

What each one is

Support and resistance is a price level where the market has turned before. It is a location, and its main claim is that other people are watching it too. Support and resistance covers it.

A break of structure is something that happens within a sequence of swings. In an uptrend, price closing above the previous swing high confirms the sequence of higher highs is unbroken. Break of structure covers the definition, and change of character covers the opposite event.

One tells you where and the other tells you whether. Whereas a level answers where price might react, a break of structure answers whether this is still a trend, and no amount of the second produces an entry price.

Where they differ

Whether it names a price. A level does. A break of structure names an event that has occurred, and by the time it is confirmed price is already beyond the swing point, so the information arrives after the move it describes.

How much interpretation is required. A level is roughly observable: price turned here. Which swing high counts as the swing high is a judgment, and changing that choice changes every structural reading that follows from it.

Whether other people share it. A daily level is on a great many screens. Your swing selection is your own, so a break of structure has to earn its value by describing the market accurately rather than from other people acting at the same place.

What each is for. A level is where you might act. A break of structure is context that tells you which direction to look in, which is upstream of any entry rather than being one.

Does a broken swing level flip?

The claim joins the two ideas. Support and resistance teaching says a broken high becomes support. A break of structure is, by definition, a close through a swing high. So if the flip is real, the swing level a break of structure has just cleared should hold price on the way back more often than a price with no history.

The test asks exactly that. For every break, it finds the first session in the next 20 that traded back to the broken level, and asks whether the close that day stayed on the break side. Then it asks the same of an ordinary level: on every other session, a price the same percentage distance from the close, with no swing behind it.

The broken levels were not special. They were revisited about as often as the ordinary ones, and when revisited they held about as often. The upward ones held a little more and the downward ones a little less, and neither gap was larger than chance produces on samples this size.

Two breaks of structure from 2026

26 May 2026 held. SPY closed at $750.59, above the $749.53 swing high from 14 May, an upward break of structure. The next session’s low reached $748.22, through the level, and the close came back to $750.46, above it. That is the retest the flip idea describes.

6 January 2026 did not. SPY closed at $691.81, 15 cents above the $691.66 swing high from 26 December 2025. The next session traded down to $689.32 and closed at $689.58, below the level. Two sessions later it closed at $694.07, above it again. A reader who called the level support and a reader who called the break failed were both looking at the same three bars.

Which one to use: Break of Structure or Support and Resistance?

Use support and resistance when you need an entry or an exit. It is a price. You can rest an order at it and measure risk from it, which a sequence label cannot offer.

Use a break of structure to decide which side to trade. If the sequence of higher highs is intact, you look for longs at levels; if it is not, you stop. That is a filter applied before the level, not instead of it.

Use both together, in that order, and do not assume the broken swing is the level. On SPY it was no better a place to lean on than any price the same distance away.

And when structure is unclear, do not trade. An ambiguous sequence of swing points is the market telling you there is no trend to be aligned with.

Why confirmation cannot be a trigger

Because a break of structure is defined by having already happened. Price must close beyond the prior swing for it to exist, so entering on it means entering after the leg that created it, which means a late price and the point where the stop has to be widest.

And because the close past the level is usually small. On SPY the median upward break closed only 0.30% past its swing high, so the level sits inside the reach of the very next ordinary bar, which is why so many came back the next day.

The original data

The sample. 8,472 daily SPY bars, 29 January 1993 to 25 September 2026, with the five-bar swing rule used on the break of structure page: a swing is known five sessions after it forms, a newer swing replaces the older one the day it becomes known, a close beyond the live swing breaks it, and each level breaks once. The rule reproduces that page’s event file row for row. Breaks with 20 sessions still to come: 142 upward and 56 downward.

Retest and hold. A retest is the first session in the next 20 whose low reached a broken swing high (or whose high reached a broken swing low). It held if that session closed at or beyond the level on the break side (one upward retest closed exactly on the level and counts as held). The ordinary level for each break is the same percentage distance from the close of every other session, run through the same rule, and the break’s result is tested exactly against that rate.

None of the differences is distinguishable from chance. For coming back at all, p = 0.90 upward and 0.86 downward. The return was usually immediate: 62 of the upward breaks and 28 of the downward ones were revisited the very next session, and the median wait was one session. Only 20 upward and 8 downward breaks never came back within 20 sessions.

Why the downward side looks worse. The median downward break closed 0.69% past its level, more than twice the upward median, and SPY rose over most of the sample, so prices above the market were reclaimed often whether or not a swing low had been there. The ordinary level carries the same drift, which is why it is the fair comparison. Every break, its retest and its ordinary-level rates are in the SPY broken swing level file.

When it fails

The characteristic failure is entering on the break of structure itself. The event is confirmation of a leg that has already run, so the entry sits at the extreme of that leg with the invalidation all the way back at the previous swing, the widest stop and the latest price the setup offers. Because the structural reading is genuinely correct, it does not feel like a mistake. The framework was right and the order was in the wrong place.

A second failure is buying the retest because the level has flipped. On SPY the broken level held about as often as any nearby price.

A third is redefining which swing counts after the fact. The rules are loose enough that a chart can be relabeled to justify whatever happened.

A fourth is reading structure on a timeframe too low for it to mean anything, where runs of one or two bars produce constant false sequences.

And a fifth is placing a stop just beyond either. On SPY, 62 of the 142 broken swing highs were back in reach the very next session.

Related

Break of structure covers the continuation event. Support and resistance explains levels taken from past turns. And change of character explains the break against the trend.

This page is educational, not financial advice. Test every idea on your own charts before risking money.