HH/HL: what trend structure tells you.

Trend is a staircase, not a forecast. Each step higher than the last. The moment a step is lower, the staircase has been broken — and the broken step is the news, not the steps that came before. "Higher Highs and Higher Lows" is the cleanest, least-decorated definition of trend that exists. It uses no math, no parameters, no smoothing, no proprietary indicator. It's a binary read of the swings the chart is already drawing for you. This lesson is how to see it, why it beats moving-average crossovers, and the moment its message changes.

What HH/HL actually means

Strip every indicator off the chart. Leave just the candles. Now look at the bumps. Every chart, on every timeframe, draws a sequence of swing highs (peaks where price reversed downward) and swing lows (troughs where price reversed upward). A swing high is just "the bar's high is higher than the highs of the few bars on either side of it." A swing low is the mirror. There's no parameter to tune. The chart prints the swings without asking you.

The trend definition is just the sequence of those swings:

• Uptrend. Each swing high is higher than the previous swing high (HH). Each swing low is higher than the previous swing low (HL). The staircase goes up.
• Downtrend. Each swing high is lower than the previous (LH). Each swing low is lower than the previous (LL). The staircase goes down.
• Range / chop. Swings drift sideways without a directional progression. HH followed by LH, or HL followed by LL, with no consistent pattern.
• Transition. A trend has been broken — the most recent swing failed the prior pattern (a LL appears in an uptrend, or a HH appears in a downtrend) — but the new direction hasn't yet confirmed.

That's the entire taxonomy. Four states. No subjective interpretation. Anyone with two eyes and a chart can mark it in thirty seconds.

How to mark it in thirty seconds

Practical instruction. Open a daily chart of any liquid stock, recent 3-6 months. Eye scan from left to right. At each obvious peak, drop a mental dot at the bar's high. At each obvious trough, drop a mental dot at the low. You don't need a tool. You don't need to be precise. The eye is good at this. (For your own journal, mark them with horizontal lines or use the platform's swing-mark tool — but the read happens before the marks.)

Now read the dots, left to right. If each high-dot is higher than the previous high-dot AND each low-dot is higher than the previous low-dot, you're in an uptrend. Mirror that for downtrend. If the highs and lows are wobbling around two horizontal levels with no progression, you're in a range. If you see one or two HH+HL pairs followed by a recent LL, you're in a transition out of an uptrend. The thirty seconds is the read. Anything that takes longer is the eye looking for things that aren't there.

Why HH/HL beats moving averages

Most retail trend signals come from moving-average relationships. The 50-day crossed the 200-day. The 20-day is above the 50. These are smoothed. Smoothing is a feature for some questions and a bug for this one. The moving average tells you where price has been on average. It does not tell you when the trend changed; it tells you a long while later, after enough new data has shifted the average. By the time the 50-day crosses the 200-day, the new trend has been visible in the swings for weeks.

HH/HL has no smoothing. The signal is the swings themselves. The first LL in an uptrend is the news, in real time, on the bar that prints it. There's no lag because there's no average. The cost of "no lag" is "no smoothing" — sometimes a single bad bar prints what looks like a swing low and you mark a LL that gets violated by the next reversal back up. That's the false-positive cost. It's small in the timeframe this curriculum focuses on (daily charts, 5-day-to-3-month holds), because the swings are large and infrequent, and you confirm structure across multiple swings before sizing into the read.

Moving averages can stay useful as a coarse trend backdrop ("price above 200-day = bullish regime"). They make a poor primary trend signal at the swing-trader timeframe. The structural read does the swing-trader work. The MA is decorative — a sanity check, not the trigger.

The transition signal

The interesting moment is the transition. An uptrend prints HH+HL+HH+HL+HH+HL for weeks. Then the next swing low is lower than the previous swing low. That's a LL. The staircase has been broken. The news is the LL itself, on the bar that prints it. The new direction hasn't confirmed yet — you'd need a second LL or a confirmed LH to call it a downtrend — but the prior trend's continuation has been falsified.

What changes for the trader at that moment? Two things. First, any open long position now sits on a violated structure; the thesis that justified the entry no longer holds, and the question is whether the stop's already hit or whether the position should be reduced. Second, no new long entries on the structure that just broke. The thesis is gone. The new thesis (downtrend) hasn't formed. This is the textbook case of "no setups today" that the framework will surface.

The cost of acting on the transition is being wrong about half the time — many transitions are just deep pullbacks that resume the prior trend. The benefit is that you stopped sizing into a structure that's no longer present. The math of small losses on broken structure beats large losses on assumed continuation.

Range and chop

The most common state, by far, is range. Stocks spend more days drifting between two horizontals than they spend trending. Range has its own structural read — swings oscillate around two levels (the range high and range low) without progression. The relevant question in a range is "are we near the top, the middle, or the bottom of the range." The trend question — uptrend / downtrend — has no answer because there is no trend.

Range is where retail traders force trends that aren't there. Three bars up looks like a trend if you squint. It's not. A trend needs at least HH+HL+HH+HL to make a structural claim. Two swings is a coincidence. Four swings starts to be a sequence.

The framework treats range as the default — most trades it would consider are rejected at the trend check because most charts most days are in chop. That's not a bug. That's the math acknowledging what's actually on the chart.

The cargo-cult callback

Lesson 7 talked about candle patterns as cargo-cult signals — bamboo runways that don't summon planes. Moving-average crossovers are the cousin: a smoothed signal that occurs after the underlying structure has already changed, but feels like the cause because it's named, computable, and graphable. "The death cross." "The golden cross." Specific names attached to a derivative measurement of something the swing structure already told you weeks earlier.

The discipline is to read the structure first and let the smoothed signals confirm — or quietly disagree — afterward. The structure is upstream. The MA is downstream. Most retail trades the downstream signal because it's named. The pros trade the upstream read because it's earlier.

What the framework does with structure

The Trap & Structure Coach computes swing structure as part of its level-confluence work. The 10 S/R sources include "prior swing high" and "prior swing low" as explicit inputs to the confluence merger. The trend state (uptrend / downtrend / range / transition) is surfaced on the candidate card; entries against the prevailing trend require an explicit override.

This is one of the gates the framework applies to a "looks good" setup. A bullish hammer at confluence support means little if the structure shows a LL just printed three bars ago. The hammer is a hammer; the trend is broken; the framework refuses the entry and shows the structural reason. Math floor messages, not vibes.

The real lesson

Trend is a staircase, not a forecast. The structure tells you where you are right now and what would have to happen for the picture to change. It does not tell you what comes next — nothing does — but it tells you which assumption the chart is currently licensing. Trade the assumption while it's licensed. Stop trading it the moment the structure breaks.

Most of the bad trades retail traders take are entries against a structure that already broke, justified by a candle pattern or an indicator they prefer. The structure was telling them. They were looking at the smoothed downstream signal. The fix isn't a better indicator. The fix is reading the staircase first and letting everything else be confirmation or disagreement, not the entry trigger.

Related: Lesson 7 — Reading candles · Lesson 4 — R:R (where the levels come from) · Trap & Structure Coach (docs)