Market structure: higher highs, lower lows, and reading the trend from swings

The short answer

Market structure is the sequence of a chart's swing highs and swing lows, read in order. That sequence, and nothing else, defines the trend: higher highs and higher lows make an uptrend, lower highs and lower lows make a downtrend, and no clean sequence is a range. It reads the pivots themselves rather than an average of them, so it states, objectively and without an indicator, what a chart is doing on the timeframe you are looking at.

The idea is old and the vocabulary is new, and the two get conflated constantly. The grammar of higher highs and lower lows is the Dow theory definition of trend, more than a century old. The terms a modern price-action or smart-money trader reaches for, break of structure and change of character, are recent labels for the same events Dow described. This guide separates the durable mechanism from the fashionable naming: what a swing actually is, how the four labels define the four possible regimes, exactly what a break of structure and a change of character are and how they differ, and, at the end, the honest limits of a framework that describes the past and does not predict the future.

Educational only. Bharath Shiksha is an educational publisher, not a SEBI-registered Investment Adviser or Research Analyst. Every chart description here is illustrative and labelled. Nothing in this article is a recommendation to enter, exit, or hold any position, and no win rate, accuracy or performance of any kind is claimed for any pattern described.

The core grammar: four labels, four regimes

Strip a chart of colour, volume and every overlay, and what remains is a series of turns: points where price stopped rising and fell, or stopped falling and rose. Each of those turns is a swing point, a swing high or a swing low. Market structure is the practice of reading those points as a sentence, asking one question at every new swing: is it higher or lower than the last one of its own kind?

That single question produces four labels. A high above the prior high is a higher high (HH). A low that holds above the prior low is a higher low (HL). A high that fails below the prior high is a lower high (LH). A low that breaks below the prior low is a lower low (LL). Four labels applied to two kinds of pivot, and from them every regime a chart can occupy.

The labels are not decoration. Higher highs and higher lows describe a market in which the marginal buyer keeps paying more on each cycle and each dip is bought sooner than the last, so control sits with demand. Lower highs and lower lows describe the reverse, supply in control, each rally sold before it reaches the previous peak. A range is the honest third state: swings that alternate without a clean directional sequence, neither side yet in control. Reading the pivots reads that balance directly, without the lag a moving average imposes or the smoothing an oscillator applies, because structure is price, not a function of price.

The four structures drawn as swing sequences An uptrend is a staircase of higher highs and higher lows. A downtrend is a staircase of lower highs and lower lows. A range is a sideways band where swing highs and swing lows repeat at similar levels with no directional sequence. Four sequences, four regimes UPTREND · HH + HL HH HL DOWNTREND · LH + LL LH LL RANGE · NO CLEAN SEQUENCE equal highs equal lows HOW TO READ IT Highs rising and lows rising: uptrend. Highs falling and lows falling: downtrend. Neither, held between two levels: range. Green dots are swing highs, gold are swing lows. Illustrative. The label is decided only after the next swing of the same kind confirms the direction.
The whole framework in four pictures. Every chart, on every timeframe, is one of these three states or in transition between them. The transition itself has a name, covered below, but the resting states are just ascending swings, descending swings, or swings going nowhere.
The three structures, by swing pattern and what each implies. Illustrative.
StructureSwing patternWhat it means
UptrendHigher highs and higher lows (HH, HL)Demand in control; each dip bought above the last, each rally exceeds the last peak
DowntrendLower highs and lower lows (LH, LL)Supply in control; each rally sold below the last, each decline undercuts the last low
RangeNo clean sequence; swings near equal levelsNeither side in control; highs and lows repeat around the same band until one breaks

Notice what the definition does not require: no indicator, no line-of-best-fit, no parameter to tune beyond the choice of which swings count. That is the appeal. A chart that has risen for three weeks but whose latest pullback dipped below the prior swing low is, by structural definition, no longer in an uptrend, even if a moving average still slopes up. The pivots lead; the average follows. That directness is also the framework's weakness, because the choice of which swings count is a judgement, and the section on honest limits returns to it.

Swing points: what actually counts as a pivot

Everything above rests on being able to say, without hand-waving, where a swing high or swing low is. The common definition is a local extreme flanked by less extreme bars on both sides. A swing high is a bar whose high is higher than a set number of bars on each side, typically at least two bars to the left with lower highs and two to the right with lower highs. A swing low is the mirror: a bar whose low is below at least two bars on each side. This is the five-bar or fractal definition, and it is the discipline that separates reading structure from drawing lines on noise.

The count is a setting, not a law, and this is the first place subjectivity enters. Requiring two bars either side marks many small swings; requiring five either side marks fewer, larger ones. Neither is correct in the abstract; they simply describe structure at different resolutions. A trader watching a three-bar swing definition and a trader watching a ten-bar one are reading two different, equally valid sentences on the same chart.

The fractal count also imposes a delay that trips up every beginner. A swing high cannot be confirmed until the bars to its right have printed, so if your rule needs two bars to the right, the swing is only confirmable two bars after it forms. The pivot existed in real time; the recognition that it was a pivot arrives later. Structure, read honestly, always lags its own turns by the confirmation window, which is precisely why it looks so much cleaner in hindsight than it feels live.

Because a swing is defined the same way at every scale, structure is fractal, or nested: a single higher low on the daily chart is itself an entire sequence of swings, some up, some down, some sideways, on the five-minute chart inside it. The same price can be a higher low on one timeframe and a completed small downtrend on another, with no contradiction. By convention the higher timeframe dominates, because its swings represent more participation and more committed orders behind them. A structural signal on a low timeframe that runs against the higher-timeframe trend is the weaker of the two, and disciplined readers treat it as counter-trend rather than as a peer. That top-down alignment, letting the larger structure set the bias and the smaller one refine the reading, is where structure stops being a definition and becomes a way of seeing.

Break of structure and change of character

A resting trend eventually ends, and the transition has two named events that beginners mix up constantly. They are opposites, and the entire difference is the direction of the break relative to the trend. Get that one axis right and the rest follows.

A break of structure (BOS) is a break with the trend, and it confirms continuation. In an uptrend, when price breaks above the most recent swing high, that is a bullish BOS: the higher-high sequence has extended, and the uptrend is, by definition, still intact. In a downtrend, price breaking below the most recent swing low is a bearish BOS, extending the lower-low sequence. Stripped of the vocabulary, a BOS is simply the trend doing what a trend does, making a new extreme in its own direction. It is the price-action name for classical trend continuation.

A change of character (CHoCH) is the first break against the prevailing structure, and it is a potential-reversal signal, not a continuation one. In an uptrend, the CHoCH is the first time price breaks below the most recent higher low instead of making a new high. The sequence of higher lows has been interrupted for the first time; the character of the swings has changed. In a downtrend, the CHoCH is the mirror, the first break above the most recent lower high. It is the earliest structural hint that the trend may be changing character, and the word hint is load-bearing: it is a warning, not a confirmation. In the classical vocabulary, a change of character is close to what a trader means by the first break of a trendline, the earliest sign the old pattern has failed, well before any new trend is proven.

Break of structure versus change of character Both panels start from the same uptrend. On the left, price breaks above the last swing high, a break of structure that confirms the uptrend continuing. On the right, price breaks below the last higher low, a change of character that is the first counter-trend break and a potential reversal signal. Same uptrend, two different breaks BOS · BREAK WITH THE TREND last swing high breaks above → BOS Continuation: the uptrend is intact CHoCH · BREAK AGAINST THE TREND last higher low breaks below → CHoCH Potential reversal: character has changed Illustrative. Most readers require a candle close beyond the level, not just a wick, before marking either break.
One axis tells them apart. A break above the last swing high extends the trend, that is a BOS and it means continuation. A break below the last higher low is the first crack against the trend, that is a CHoCH and it means the character has changed. The BOS runs with the trend; the CHoCH runs against it.
Break of structure versus change of character, on an uptrend and a downtrend. Illustrative.
EventWhich level breaksRelative to trendWhat it signals
Bullish BOSMost recent swing high, in an uptrendWith the trendContinuation: higher-high sequence extends
Bearish BOSMost recent swing low, in a downtrendWith the trendContinuation: lower-low sequence extends
Bearish CHoCHMost recent higher low, in an uptrendAgainst the trendPotential reversal: first break of the up-sequence
Bullish CHoCHMost recent lower high, in a downtrendAgainst the trendPotential reversal: first break of the down-sequence

The relationship between the two is a sequence, not a rivalry. A change of character is the first sign; a break of structure in the new direction, if it comes, is the confirmation. In an uptrend that is topping, the bearish CHoCH breaks the last higher low, and only later, if a lower high then gives way to a lower low, does a bearish BOS confirm that a downtrend has actually begun. The CHoCH is early and therefore uncertain; the BOS is later and therefore more settled. Neither is a signal to act in the sense of a recommendation, and this guide makes no claim about how often either resolves as expected. They are ways to describe context, and the honest reader holds them loosely.

It is worth naming the lineage plainly, because the marketing around it is loud. Break of structure and change of character are smart-money-concepts (SMC) terminology for ideas that long predate SMC. A trend that continues to make new highs is trend continuation, described by Dow more than a century ago. A first failure of the swing sequence is a trendline break, taught in classical technical analysis for generations. The SMC framing repackages this durable price-action grammar in newer language. The vocabulary is useful shorthand; it is not a discovery, and it does not deserve the hype it often carries.

The honest caveats: what structure cannot do

A framework is only as trustworthy as its stated limits, and market structure has several that the enthusiastic tutorials skip. Taking them seriously is what separates using structure as context from mistaking it for a signal.

First, labelling is subjective at the input. The rule is objective once you fix a swing definition and a timeframe, but which swings count is a choice, and two disciplined readers who choose different swing sensitivities will label the same chart differently and both be defensible. There is no privileged, true set of swings floating above the chart waiting to be found. This is the single most important thing to internalise: the objectivity of market structure is conditional on inputs that are themselves discretionary.

Second, structure is clearer in hindsight than live. On a printed historical chart every swing is obvious and every BOS and CHoCH looks inevitable. In real time the confirming bars have not arrived, the latest swing might still extend, and the label you assign now may be invalidated by the next few bars. The clean staircase is a property of the completed past, not of the live right edge.

Third, a change of character can be a fakeout, especially in a range. Inside a sideways band, price routinely pokes through a prior swing and reverses, which will trip a CHoCH rule that then fails as the range simply continues. A wick through a level is often a stop hunt rather than a structural break, which is why most readers require a candle to close beyond the level before marking BOS or CHoCH, and why a CHoCH taken at face value inside a range is one of the most reliable ways to be repeatedly wrong.

A change of character that fails inside a range Inside a sideways range, price wicks below the lower boundary and triggers a change of character signal. Price then closes back inside the range and continues sideways, so the signal was a fakeout. This shows that a change of character read at face value inside a range is unreliable. The same break, read two ways A wick through a level inside a range: signal, or stop hunt? range high range low wick below → CHoCH triggers closes back inside → it fails Illustrative. The break looked like a change of character; the close revealed a fakeout, and the range continued.
A CHoCH is not a promise. Inside a range, the same excursion that trips a change-of-character rule is, more often than the tutorials admit, a wick that closes straight back inside. Structure describes context; it does not certify the next move, and the honest reader treats an in-range break as suspect until a close proves it.
The honest limits of market structure, and what each one means for how you read it.
CaveatWhy it holdsWhat it means in practice
Subjective inputsWhich swings count depends on the sensitivity and timeframe you chooseTwo valid labellings can disagree; fix your swing rule before you read, and hold the label loosely
Clearer in hindsightConfirming bars only arrive after the swing formsThe live right edge is always ambiguous; a label can be invalidated by the next few bars
Fakeouts in a rangePrice wicks through prior swings without a directional sequenceAn in-range CHoCH is unreliable; require a close beyond the level, not a wick
Describes, does not predictStructure is a record of what price has doneIt is context for a decision, never a signal on its own, and implies no win rate

Fourth, and most important, structure describes the past; it does not predict the future. An uptrend is a statement that price has, so far, made higher highs and higher lows. It is not a forecast that it will continue to. A BOS confirms that the trend has, to this point, extended; a CHoCH notes that the sequence has, for the first time, broken. Both are descriptions of what has already happened. Market structure is a framework for reading context, not a mechanism for generating signals, and any source that presents it with a hit rate attached has quietly changed it from a description into a promise it cannot keep.

Where structure misleads. The recurring error is treating a label as a prediction: entering the moment a BOS prints because "the trend continues", or flipping bias the instant a CHoCH appears because "the trend has reversed". Structure justifies neither. It tells you the state of the sequence up to now, on the timeframe and swing setting you chose. The interpretation, the risk, and whether an idea is worth acting on at all are separate questions it does not answer.

Where it fits, and how we teach it

Market structure is the first thing worth learning because everything else on a chart is read through it. A support or resistance level matters differently in an uptrend than in a range. A breakout means one thing when it extends a higher-high sequence and another when it merely pushes at the top of a band. A trendline is, at heart, a hand-drawn approximation of the swing-low sequence that structure names precisely. Reading the sequence first gives every later tool its context.

That is also why the framework rewards being read against a stated method rather than in isolation. Fixing a swing definition, deciding which timeframe sets the bias, and knowing in advance what would count as a break, and then treating all of it as context rather than instruction, is upstream judgement work, and that judgement is exactly what the method we teach is built around. In the Bharath Shiksha curriculum, chart structure is introduced early in Stage 1 and becomes the substrate under the systematic material in Stage 2, where continuation and reversal events are studied as things to be defined and journalled, never as signals with a promised outcome. The framework is the easy half; reading it honestly, and knowing what it cannot tell you, is the part that takes the work.

Ready to read a chart the way this article describes?

Market structure is taught from first principles early in Stage 1 and carried through the systematic material in Stage 2, with illustrative, labelled charts and a fixed method for reading swings before any other tool. Bharath Shiksha is a 30-volume curriculum across 6 stages, from chart structure (Stage 1 at ₹14,999) through advanced systematisation, or the full bundle at ₹1,49,999. Every volume includes a companion worksheet and a gate quiz.

Take the free diagnostic →

Common questions

Frequently asked questions

Market structure is the sequence of a chart's swing highs and swing lows read in order. That sequence defines the trend directly from price, with no indicator. Higher highs and higher lows describe an uptrend, lower highs and lower lows a downtrend, and the absence of any clean sequence a range. Because it reads the pivots themselves rather than an average of them, market structure is the most direct way to state, objectively, what a chart is doing on a given timeframe.

An uptrend is a sequence in which each swing high exceeds the prior swing high and each swing low holds above the prior swing low, abbreviated HH and HL. A downtrend is the mirror: each swing high fails below the prior high and each swing low breaks below the prior low, abbreviated LH and LL. This is the Dow theory definition of trend, which predates modern price-action vocabulary. The trend is considered intact until the sequence breaks, and describes only the timeframe you are reading.

A swing high is a local price peak with lower highs on both sides, commonly at least two bars whose highs are lower to its left and two to its right, the five-bar or fractal definition. A swing low is the mirror, a trough with higher lows on each side. The count of bars you require is a setting, not a law, and a stricter count marks fewer, larger swings. This is why the same chart yields a different structure to two readers who choose different swing sensitivities.

A break of structure is a break in the direction of the prevailing trend, so it confirms continuation. In an uptrend, price breaking above the most recent swing high is a bullish BOS, extending the higher-high sequence. In a downtrend, price breaking below the most recent swing low is a bearish BOS, extending the lower-low sequence. It is the price-action name for what classical technicians simply called the trend continuing to make new highs or new lows. A wick alone is not usually treated as a break; most readers require a close beyond the level.

A change of character is the first break against the prevailing structure, so it is a potential-reversal signal rather than a continuation one. In an uptrend, the CHoCH is the first time price breaks below the most recent higher low, instead of making a new high. In a downtrend, it is the first time price breaks above the most recent lower high. It is the earliest structural hint that the trend may be changing character. It is a hint, not a confirmation: many change-of-character events fail, especially inside a range.

The single difference is the direction of the break relative to the trend. A break of structure runs with the trend and confirms continuation: a new swing high in an uptrend, a new swing low in a downtrend. A change of character runs against the trend and flags a possible reversal: the first broken higher low in an uptrend, the first broken lower high in a downtrend. In classical terms a BOS is the trend making a new extreme, and a CHoCH is the first failure of that pattern, close to what a trader means by an early trendline break.

Structure is fractal: an uptrend on the daily chart is built from smaller swings that, on a five-minute chart, form their own uptrends, ranges and downtrends. The same price can be a higher low on one timeframe and a completed downtrend on another. By convention the higher timeframe dominates, because its swings represent more participation and more committed orders. A structural signal on a low timeframe that contradicts the higher-timeframe trend is the weaker of the two and is read as counter-trend.

Both, and the distinction matters. The rule is objective: once you fix a swing definition and a timeframe, HH, HL, LH and LL follow mechanically. The inputs are subjective: which swings count depends on the sensitivity you choose, and structure is far clearer in hindsight than live, because the swing that confirms a label only completes after later bars. Two disciplined readers can label the same chart differently and both be defensible. Structure describes what price has done; it does not predict what it will do.

Where the facts come from

Sources

  • Dow theory definition of trend. An uptrend is a series of higher highs and higher lows, a downtrend a series of lower highs and lower lows, and a trend is assumed intact until the sequence reverses. This is the century-old lineage of the higher-high, lower-low grammar. incrediblecharts.com
  • Break of structure versus change of character. BOS is a break in the direction of the existing trend and confirms continuation; CHoCH is a break against it and signals a potential reversal. The single difference is the direction of the break relative to the trend, and most readers require a candle close, not a wick, to mark either. innercircletrader.net
  • Swing points and the fractal, multi-timeframe reading. A swing high has lower highs on both sides, commonly two bars each side, and swings are fractal across timeframes, with top-down alignment letting the higher timeframe set the bias. tradingsim.com
  • The live base rate for retail derivatives. SEBI's FY25 study, published July 2025, found that 91 percent of individual equity-derivatives traders lost money, a net ₹1,05,603 crore, the sober context for treating any chart framework as description, not a promise. theweek.in
Educational note. This guide explains how market structure describes trend from a chart's swings. It is not a recommendation to trade or invest, and it is not investment advice. No win rate, accuracy, or performance is claimed for any pattern described. Bharath Shiksha is an educational publisher, not a SEBI-registered investment adviser or research analyst, and all charts here are illustrative and labelled.

Related reading