Market structure in trading is nothing more than a chain of swing highs and swing lows, read in order. Not a study. Not a slope calculation. Just the sequence of turns price has already made, laid out one after another. If you already know how to spot a swing high and a swing low, you have every piece you need. The only new step is connecting them.
I wrote about the individual shapes in the piece on swing highs and swing lows — a swing high is a peak with lower highs on both sides, a swing low is a trough with higher lows on both sides. That’s still true here. Nothing about the definition changes. What changes is the question you ask once you have three or four of them marked on a chart: are they climbing, falling, or going nowhere.
Structure is what swing points look like in a row
Take the last four swing points on any chart. Label them in order — low, high, low, high. Now compare each one to the one before it of the same type. If every swing low sits above the prior swing low, and every swing high sits above the prior swing high, that’s an uptrend. Higher highs, higher lows. Flip it and you get a downtrend: lower highs, lower lows, each new swing failing to reach the old ceiling and each pullback digging further beneath the old floor.
There’s a third state, and it’s the one most retail charts spend most of their time in. A swing high that’s roughly even with the last swing high. A swing low that’s roughly even with the last swing low. Neither side making progress. That’s a range, and it’s not a lesser condition — it’s just structure telling you the buyers and sellers are matched. No trend indicator is required to see any of this. You’re not smoothing anything. You’re reading turns that already happened.
Ranges get treated like dead time by traders waiting on a trend indicator to slope one way or the other. I don’t treat them that way. A range has its own structure — a ceiling that keeps rejecting price and a floor that keeps catching it — and that structure is just as tradeable as a trend, provided you accept that the edges are the whole trade. The mistake is trying to force trend language onto a market that structure is plainly telling you isn’t trending. If the swing highs aren’t climbing and the swing lows aren’t climbing, calling it an uptrend because a headline says so is wishful thinking, not analysis.
The first broken link is the earliest honest signal you get
An uptrend holds as long as each pullback stays above the last swing low. The moment a pullback prints a swing low that comes in lower than the one before it, the chain is broken. That’s the first lower low in what had been a series of higher lows, and it’s the earliest objective sign that the character of the move has shifted. Same logic in reverse for a downtrend — the first higher high, after a run of lower highs, is the tell.
What makes this useful is that it doesn’t rely on interpretation. You’re not deciding whether momentum feels weaker or whether volume looks off. You’re comparing two numbers: is this swing low higher or lower than the last one. A ten-year-old could do the comparison. The skill isn’t in the math — it’s in noticing it happened and knowing what it does and doesn’t mean.
A broken structure is a reason to watch closer, not a reason to flip
This is where I differ from how a lot of people treat structure breaks. A broken structure isn’t a new trend. It’s a crack. Plenty of uptrends take a single lower low, shake out the traders who got excited about the crack, and then resume making higher highs like nothing happened. If you treat every first break as your cue to reverse a position, you’ll get run over by exactly that shakeout, over and over.
What I do instead is treat the break as an instruction to raise my attention, not an instruction to act. The old trend has lost its perfect record. That’s real information — it means the next swing high matters more than the last five did, because it will tell me whether the buyers can still push price above the prior high or whether they’re now failing where they used to succeed. One broken swing low is a question. A failed retest of the prior swing high is closer to an answer.
Confirmation, in practice, usually looks like a second piece of structure lining up with the first crack. The lower low happens. Then price rallies but can’t clear the previous swing high before rolling over again — now you have a lower low and a lower high, the actual two-part definition of a downtrend, not just a single dented number. That’s the difference between noticing a crack and being willing to trade around it.
What this looked like in AMD, and what I did with it
AMD had been climbing through most of the spring — a clean run of higher highs and higher lows on the daily chart, nothing complicated about it. Late in the run it topped near $187, pulled back to $171, and pushed to a new high around $192. Standard uptrend structure. Then the next pullback broke it: instead of holding above $171, price sold down to $164 before finding buyers. First lower low in the sequence.
I didn’t short it. I marked $192 as the level that needed to hold for the uptrend to still be intact, and I watched the next rally. It got to $181, stalled, and rolled back over without ever testing $192. That was the second piece — lower high, lower low, both in place. I shorted at $178 with a stop at $184, above the failed rally. Price worked down to $167 over the following nine sessions, and I covered most of the position there.
The lower low by itself wasn’t the trade. It was the flag. The trade came from the failed rally that confirmed it — the point where AMD had a real chance to reclaim its old high and didn’t take it. That gap between the first crack and the confirmed trade is exactly where a lot of traders lose money jumping the gun.
It’s also worth saying what didn’t happen. The uptrend didn’t announce itself as over with a headline or an earnings miss. Nothing about the fundamentals changed between the $192 high and the $164 low. What changed was purely structural — the chain of higher highs and higher lows stopped extending, and then it inverted. Reading market structure in trading this way means the decision to short AMD had nothing to do with a story about the company. It had to do with two swing points failing to keep the old pattern alive.
Why this replaces a trend indicator entirely
A trend-following indicator is built from price, but it’s always a lagging summary of price — an average, a slope, a smoothed line trying to tell you in one number what the raw chart already showed you in three or four turns. By the time most trend tools flip, the structure has usually already broken and often already confirmed. You’re not getting new information from the indicator. You’re getting old information redelivered slower.
Reading swing points directly skips that translation step. You see the higher low become a lower low the same day it happens, not after an average catches up to it. And because you’re working from the actual turns instead of a derived line, there’s nothing to configure — no period setting, no sensitivity slider, no version of the tool that disagrees with another version of the same tool. The chart just shows you where price turned. Market structure in trading is that turn history, nothing added.
How to start reading it on your own chart
Pull up a daily chart, strip off anything that isn’t a candle, and mark the last five or six swing points using the same test from the earlier piece — higher on both sides for a high, lower on both sides for a low. Write down whether each one is higher or lower than its predecessor of the same type. That sequence is the structure. When it’s consistently climbing or consistently falling, you have a trend and you know which side of it you’re on. When one link breaks, don’t reach for a conclusion. Watch the next swing. It’ll either confirm the break or prove the old trend still has one more push left in it.
I read market structure in trading off a blank chart, no indicators, in Static, the free daily chart room run by Draw Lines Make Money. If you want to watch how a structure break gets confirmed or dismissed in real time, you can sit in:
Join the free Static chart room →
Disclosure: that’s an affiliate link — I may earn a commission if you join a paid tier later, at no extra cost to you. The free room is free.