Market Structure Trading: Reading HH, HL, LH and LL

If you arrived here from an economics course, this is a different subject. Nothing about monopolies or market concentration. Just the question of where price turned, and what the order of those turns tells you before you place anything.

The four abbreviations in the title do most of the work in Smart Money Concepts, and they are usually taught as if the chart handed them to you. It does not. A peak is not a higher high because it looks like one. It becomes one when a specific event happens, and that event is the part almost every explanation skips.

Worth knowing where the vocabulary comes from: HH and HL predate SMC by decades as plain price-action shorthand, and the SMC material that spread across YouTube from 2022 onwards borrowed them and then added rules on top. Those added rules are the whole difference between the two readings, so this page covers the labels, the event behind each one, and what happens when you leave the events out.

What is market structure on a price chart?

Market structure is the sequence of swing highs and swing lows that price leaves behind, read as a pattern. Two things make it stricter than it first sounds. A swing point is defined by pullbacks rather than by the shape of three or five candles, and every structural mark is placed to the left of price, only after the event that creates it.

The pullback part matters because it rules out the shortcut most people use. Counting a peak with two lower candles on each side gives you a zigzag, but not this one: a pullback only counts when its candle takes the liquidity of the candle that ended the impulse, wick or body, colour irrelevant. A candle that never reaches that extreme, or an inside bar, is not a pullback at all, so the extreme it produced is not a swing point.

The left-side rule is the other half. While a rally is still going, there is no higher high yet, and drawing one is drawing a forecast. Structure is a record. It tells you where liquidity was taken, and that is the only reason the method bothers with it.

What do HH, HL, LH and LL actually mark?

Four abbreviations, two comparisons. HH is a higher high: a swing high above the previous swing high. HL is a higher low, a swing low above the previous swing low. LH and LL are their mirror images. Each label is relative to the last point of the same kind, and each one waits for its own event before it exists.

Read as pairs rather than as single marks. A higher high followed by a higher low is a sequence, and that is what a trader means by saying the trend is up. One label on its own describes a single turn.

Two things the labels do not do. They do not tell you the move will continue, because a sequence describes the past. And they do not rank the points by importance: the ones that count are the ones the working timeframe produced, and mistaking the small marks inside a leg for the main ones is the usual reason a bias flips every few bars, which is the subject of internal versus external structure.

When does each label actually appear?

Each one waits for a different event, and the order never changes: the inducement is armed, price takes it, the higher high appears, a body close makes the break of structure, and the low that did the taking becomes the higher low. Nothing skips a stage.

The two confirmations are deliberately asymmetric, and it is worth knowing which is which. The inducement side accepts a wick: any touch through that low takes the liquidity sitting there, and the higher high is granted. The break side requires a body close beyond the level. A wick through the break level takes nothing and grants nothing; it just moves the level to the tip of that wick.

There is one more condition on the higher high that catches people out. Taking the inducement is not enough on its own. The new peak also has to reach the break level, which means the leg has to actually do something after the sweep. A rally that dips through the inducement and then stalls below the previous peak has not produced a higher high.

The practical effect is a delay you cannot remove. Labels arrive a few bars after the moment they refer to, and the marker is then placed backwards, onto the bar where the extreme really was. That is not a tool being slow, and it is not repainting either. Worth being strict about the difference, because one is honest and the other quietly rewrites your backtest: the way to settle it is to run a 100-bar replay test and watch whether anything already printed moves.

Why does the classic reading disagree with this one?

Because the classic reading skips the events. It labels every local peak and trough it can see, which produces a tidier-looking chart and one specific, expensive error: the pullback that takes the inducement looks exactly like the trend breaking down.

Look at what happens in the middle of that pair. Price makes a low beneath the previous one. On the classic reading that is a lower low inside an uptrend, the staircase breaking, and the honest conclusion is that the trend is over. On the reading with the inducement in it, that low is where the stops of everyone who bought the pullback were sitting. They get taken, and the move straight after is the one that breaks structure for real.

None of that makes the classic labels useless as shorthand. It makes them a description of shape, while the method needs a description of liquidity. The full mechanics of the trap itself, including how to tell an inducement from an ordinary pullback, belong to inducement explained, and the repeatable order you mark it all in is a rule set of its own.

What are the three states a market can be in?

Three, not two. An uptrend is a sequence of higher highs and higher lows. A downtrend is lower highs and lower lows. Everything else is a range, where highs and lows overlap enough that neither sequence holds. Most explanations cover the first two and leave you unprepared for the third, which is where charts spend a large share of their time.

Reading a range as a broken trend is an expensive habit. In a range the swing points are still real and still measurable, they just do not stack, so levels matter more than direction and the same zone gets tested from both sides.

Which state you are in depends on the timeframe you ask, and the method is specific about how to handle that: mark structure completely on one working timeframe, and drop to a lower one only once price has reached a zone worth watching. Rebuilding structure on the lower timeframe instead of reading the reaction there is how a bias ends up changing several times an hour.

How does structure connect to the rest of SMC?

Structure is the context layer. It tells you which direction the sequence currently favours and where the levels that define it sit, and then other concepts supply the actual location of a trade. Used the other way around, as a signal in itself, it produces entries with no reason behind them beyond "the trend is up".

An order block is a zone price left behind before a move, and structure decides whether that zone is worth watching or sits against the prevailing sequence. A fair value gap marks an imbalance inside a leg, and the same filter applies when you get to trading a fair value gap. Order flow describes the correction that produced the last valid leg, which is a structural object measured with structural rules.

Liquidity is not a separate topic here either. Swing highs and lows are where stop orders cluster, so structure and liquidity are two descriptions of the same points, and the inducement is simply the pool the method insists you account for before believing a break.

What structure buys you, then, is a shared coordinate system. Every other concept in the method has to be placed somewhere, and these points are what you place it against. Automating it does not mean the tool sees something you cannot. It means one rule set gets applied to every chart at every hour without skipping the inducement because the rally looked convincing.