Market structure is the sequence of highs and lows a chart is printing. An uptrend makes higher highs and higher lows; a downtrend makes lower highs and lower lows; a range makes neither. Reading structure means naming which of the three you are in before you decide what trade is allowed.
That is the whole concept, and its value is that it is boring and repeatable. Structure does not tell you what happens next. It tells you what has been happening, in language precise enough that two traders looking at the same chart should reach the same description — which is more than can be said for most chart reading.
The three states
| State | The sequence | What it implies | What ends it |
|---|---|---|---|
| Uptrend | Higher highs and higher lows | Buyers are paying up; pullbacks find demand early | Price takes out the last higher low |
| Downtrend | Lower highs and lower lows | Sellers are accepting less; rallies get sold sooner | Price takes out the last lower high |
| Range | Highs and lows roughly level | No side is winning; edges matter, middle does not | A close outside the range that holds |
Most of the time you are in the third one, which is the state traders most want to avoid admitting to. A range with a slight tilt is not a trend, and treating it as one is how people end up buying the top of a box for the fourth time this week.
Defining a swing point so you cannot cheat
Structure is only as objective as your definition of a swing high and a swing low. The usual convention: a swing high is a candle whose high exceeds the highs of a fixed number of candles on either side — two or three is standard. A swing low is the same, inverted.
The exact number matters less than fixing it. If you use two candles when you want to see a trend and five when you do not, structure stops being a description and becomes a rationalisation. Write the number into your trading plan and apply it to every chart, including the ones you disagree with.
Break of structure vs change of character
These two phrases get used interchangeably and mean opposite things.
- Break of structure (BOS) — the trend continues. In an uptrend, price takes out the previous swing high. Nothing has changed except that the sequence extended.
- Change of character (CHoCH) — the trend's defining sequence breaks. In an uptrend, price takes out the previous higher low. The chart has not necessarily reversed, but it can no longer be described as an uptrend without lying.
The practical difference is what each one licenses. A BOS is a reason to stay in and trail. A CHoCH is a reason to stop taking longs until a new sequence establishes itself. Neither is an entry signal on its own — a level being taken out on a thin wick during the lunch hour is not the same event as a decisive close through it, which is the distinction covered in what makes a breakout real.
Structure is timeframe-specific, and that is fine
A five-minute chart can be in a clean downtrend while the daily chart is in a clean uptrend. Both descriptions are correct. The lower-timeframe downtrend is the higher-timeframe pullback, viewed at higher resolution.
What goes wrong is treating a lower-timeframe structure as permission to fight a higher one. If the daily is making higher highs and higher lows, a five-minute change of character is information about the pullback, not about the trend. Use the higher timeframe to decide direction and the lower one to decide entry — the ordering set out in multi-timeframe analysis. Two timeframes is enough. At four, something is always confirming something.
Does the sequence actually mean anything?
It is fair to ask whether "higher highs continue" is anything more than a story traders tell each other. The honest answer is that the evidence supports persistence, not prediction.
Moskowitz, Ooi and Pedersen examined 58 diverse futures and forward contracts — equity indices, currencies, commodities and sovereign bonds — over more than 25 years, and found that an instrument's past 12-month excess return was a positive predictor of its future return, with those 12-month time series momentum profits positive not just on average but for every one of the contracts studied (Moskowitz, Ooi & Pedersen, "Time Series Momentum", Journal of Financial Economics, 2012). The same paper found the effect partially reverses over longer horizons.
Two things follow, and both matter. First, trends persisting is a real, measured tendency across markets rather than folklore — so describing a chart by its sequence of highs and lows is describing something. Second, that research studied 12-month horizons across a broad basket, which is a long way from the five-minute chart in front of you. It gives you no licence to expect any individual intraday trend to continue. Structure earns its place because it is a consistent frame of reference, not because it forecasts.
How structure should change what you do
- It filters direction. In a clean uptrend, longs at support are the trade and shorts are a hobby. This single filter removes more bad trades than any indicator.
- It gives you an invalidation that is not arbitrary. If your thesis is "the uptrend continues", the last higher low is where that thesis is wrong. That is a stop with a reason behind it — the standard set out in writing your invalidation before you enter.
- It gives you a target that is not invented. The next swing high is where price is already trying to go. You do not need a round number.
- It tells you when to stand down. A range with no clean structure is a legitimate reason to trade smaller or not at all. Most people treat "no setup" as a failure rather than a result.
Frequently Asked Questions
What counts as a swing high or a swing low?
A swing high is a candle whose high is higher than a fixed number of candles on either side of it — commonly two or three. A swing low is the mirror image. The number is a convention, not a law, but you must choose one and keep it. If you allow yourself two candles on one chart and five on another, you can produce whichever trend you were hoping to see, which is the most common way traders fool themselves with structure.
What is the difference between a break of structure and a change of character?
A break of structure continues the existing trend: in an uptrend, price takes out the last swing high, confirming buyers are still in control. A change of character breaks the trend the other way: in an uptrend, price takes out the last higher low. The first says the trend is intact, the second says the sequence that defined it has ended. The distinction matters because only one of them is a warning.
Which timeframe should you read market structure on?
Read it on two: one for context and one for execution, and expect them to disagree. A five-minute downtrend inside a daily uptrend is normal, not a contradiction — it is a pullback described at higher resolution. The higher timeframe should decide which direction you are willing to trade and the lower one should decide where you enter. Traders get hurt when they take a lower-timeframe structure as permission to fight the higher one.
Does market structure actually predict anything?
Not in the sense of forecasting the next candle. What the evidence supports is persistence: trends across many liquid markets have historically continued more often than chance over horizons of roughly a year, which is a statistical tendency, not a rule about any single chart. The practical value of structure is not prediction — it is a consistent way of describing what a chart is doing so that your entries, targets and invalidation are all defined against the same reference.
Bottom line
Market structure is a naming discipline: higher highs and higher lows, lower highs and lower lows, or neither. Fix your definition of a swing point, read it on two timeframes, and treat a change of character as a warning rather than a signal. Its value is that it gives your entry, your target and your invalidation a single shared reference, which is what turns a collection of opinions into a system. Read it next to technical analysis basics, put the levels on the chart with a pre-session markup routine, and pair it with support and resistance for the prices where the swings tend to form.