Dow Theory Made Simple: How Higher Highs and Higher Lows Lead to BOS and CHoCH in Market Structure

You open a chart and still can’t tell whether it’s going up, going down, or which way it’s about to move — that’s usually the first wall every trader hits. The tool that clears this fog is Dow Theory, which has been in continuous use for over a hundred years. There’s no difficult math involved; all you do is check whether the highs and lows are rising or falling. That alone makes it much easier to read the skeleton of the market.

This page is Step 03 in our discretionary trading curriculum, and among all our teaching materials it is the single most important page connecting discretionary trading and SMC (Smart Money Concepts). We’ll walk through everything from the fundamentals of Dow Theory, through the direct translation into the terms SMC uses — BOS, CHoCH, and MSS — to how an EA mechanically judges market structure and how AI can reduce the subjectivity involved in reading highs and lows, all grounded in the logic of “why you read it that way” rather than rote memorization. If you haven’t yet read our prerequisite article on chart basics (timeframes, trends, and ranges), reading that first will make everything here much easier to follow.

Why You Need It — Dow Theory Is a Map for Reading Market Structure

In a single sentence, Dow Theory holds that price movement has direction (a trend), and that direction shows up in the way highs and lows line up. Even on a bare chart with no indicators at all, looking at how the peaks (highs) and troughs (lows) connect tells you whether the current flow is upward, downward, or a directionless range.

The reason this matters is that recognizing “what state the market is in right now” is the starting point for every kind of market analysis. Support lines, price action, and entry patterns all only make sense on top of the foundation of knowing whether this is a pullback within an uptrend or a bounce within a downtrend. If you stack individual techniques on top of a shaky foundation, you end up with fundamental mismatches — like repeatedly selling into strength during an uptrend. Dow Theory is the first step in drawing that foundation, that map.

The Six Tenets of Dow Theory (Start With the Three Beginners Need)

Classically, Dow Theory consists of six tenets, but as a beginner you only need three of them to start applying it in practice. The remaining three are fine to set aside for later, as background knowledge that deepens your understanding once you have it.

Here are the three practical tenets you should learn first.

• There are three types of trend — primary (months to years), secondary (weeks to months), and minor (days or less). It’s important to have a feel for how a smaller flow on a lower timeframe is nested inside a larger flow on a higher timeframe. This connects directly to the idea of multiple timeframes.
• A trend continues until a clear reversal signal appears — once a flow turns upward, treat it as upward until there’s evidence it has broken down. This is the single most practically useful line in Dow Theory, and it’s the core of the “continuation and reversal” discussion below.
• Volume (the market’s energy) confirms the trend — this is traditionally confirmed with volume, but since FX has no official volume figures, we substitute it by reading the momentum of price movement and the size of the candles instead.

For reference, the remaining three tenets are: “the averages (the market as a whole) discount everything,” “a primary trend unfolds in three phases (accumulation, public participation, and distribution),” and “the various market averages must confirm each other.” Since some of these assume a stock market context, it’s fine for FX beginners to focus solely on the three above.

Reading Trend Through Rising and Falling Highs and Lows

This is the heart of Dow Theory. A trend can be classified into just three types — up, down, or range — based purely on how two kinds of points, “highs” and “lows,” line up. As a starting premise, price never moves in a straight line up or down; it advances in a zigzag, forming peaks and troughs along the way. We call these peaks highs and these troughs lows, and we look at how they’re positioned relative to each other.

Uptrend (Highs and Lows Both Rise)

When the most recent high is higher than the previous high, and the most recent low is also higher than the previous low — a continuous run of “rising highs, rising lows” — that’s an uptrend. Buying pressure is outweighing selling pressure, and even when a pullback (a dip) occurs, price fails to break the previous trough and climbs again. In SMC terms, these highs are called Higher Highs (HH) and these lows are called Higher Lows (HL).

Let’s look at a numerical example. Say gold moves from 3200, pulls back to 3180, rallies to 3240, pulls back to 3210, and rallies again to 3280. The highs rise 3200 → 3240 → 3280, and the lows rise 3180 → 3210. This is a clean uptrend.

Downtrend (Highs and Lows Both Fall)

This is the mirror image of an uptrend. When the most recent high is lower than the previous high and the most recent low is also lower than the previous low — a continuous run of “falling highs, falling lows” — that’s a downtrend. Even when a bounce (a rally) occurs, price fails to clear the previous peak and falls again. In SMC terms, these are Lower Highs (LH) and Lower Lows (LL).

Range (Sideways Movement)

This is when neither the highs nor the lows rise or fall, and price oscillates around roughly the same level. The highs cluster together, and the lows cluster together. Buying and selling pressure are roughly balanced with no clear direction, and in Dow Theory terms this is a phase of “no trend,” where the default stance is to stand aside. These clusters of “roughly equal highs and lows” at the top and bottom of the range are called Equal Highs / Equal Lows in SMC, and they become pools of liquidity, as we’ll discuss below.

State	Sequence of highs	Sequence of lows	Basic stance
Uptrend	Rising (HH)	Rising (HL)	Look to buy on pullbacks
Downtrend	Falling (LH)	Falling (LL)	Look to sell on rallies
Range	Roughly flat	Roughly flat	Stand aside until direction emerges

Trend Continuation and Reversal — Where Do You Say the Flow Has Changed?

What you most want to know in practice is whether the current flow will continue or has changed. Dow Theory’s criterion is simple: an uptrend is considered broken when price clearly breaks below the most recent low (the pullback trough), and a downtrend is considered broken when price clearly breaks above the most recent high (the rally peak).

Let’s think through our earlier uptrend example (3200 → 3180 → 3240 → 3210 → 3280). Suppose price now falls and breaks below 3210, the most recent pullback low, dropping to, say, 3170. This is a sign that “the sequence of rising lows has broken — the uptrend structure is starting to fail.” Conversely, if price bounces without breaking 3210 and climbs above 3280 again, we judge the uptrend to be continuing.

The “Which High or Low Do You Use?” Problem (Dow Theory’s Subjectivity)

Here we run into Dow Theory’s biggest weakness, and the point where beginners most often stumble. Even if we say “the most recent low,” a chart contains troughs of all different sizes. On a one-minute chart, even a tiny dip counts as the “most recent low,” while on a daily chart, a large trough from several days ago is the “most recent low.” Depending on which scale of high or low you adopt, the same chart can look like either an uptrend or a downtrend — this is Dow Theory’s subjectivity.

Major educational resources tend to leave this vague, so here is our own practical framework. Fix the highs and lows you use for judgment to the peaks and troughs that stand out clearly on the timeframe one level above the one you actually trade. If you day-trade primarily on the 1-hour chart, use the highs and lows that are clearly visible on the 4-hour chart as your reference. Deciding on your reference timeframe up front like this dramatically cuts down on the hesitation over “which trough do I use?” This approach is shared with our article on drawing horizontal support and resistance lines.

Using It in Discretionary Trading — Fakeouts and the Lag in Signals

Dow Theory is powerful, but it isn’t a cure-all. When using it in discretionary trading, you need to build in an awareness of the following two weaknesses.

The first is a fakeout. Price often appears to momentarily dip below the recent low, only to snap back and continue rising. Whether it was just a brief wick poking below the level, or a solid break with the candle body and a close confirmed below it, makes all the difference. When applying Dow Theory, judging breaks by the close (a confirmed candle) rather than a momentary wick poke makes it much easier to reduce fakeouts. In SMC, the true nature of this “brief break that snaps right back” is explained as a liquidity grab, which we’ll cover below.

The second is that the signal lags. The judgment “a break of the most recent low means a reversal” is only confirmed after the break actually happens, so you can only react at a point well removed from the actual top or bottom. In other words, Dow Theory isn’t a tool for “selling at the exact top and buying at the exact bottom” — it’s a tool for “confirming that the flow has changed and then getting on board.” Accepting this lag, the standard practice is to combine it with price action and support/resistance to catch earlier signs.

Bridging to SMC — What Dow Theory’s Terms Are Called in SMC

This is the core of this page, and where we believe we can add the most value. SMC (Smart Money Concepts) and ICT can look like a completely separate, arcane methodology at first glance, but much of their foundation is simply Dow Theory restated from an institutional investor’s point of view. Now that you understand Dow Theory, you can absorb SMC’s central terms almost entirely through “translation.”

Trend Continuation = BOS (Break of Structure)

BOS stands for “Break of Structure” and refers to updating the most recent high (during an uptrend) or low (during a downtrend) in the direction of the trend. If the previous HH is taken out during an uptrend, that’s a bullish BOS confirming continuation of the uptrend; if the previous LL is taken out during a downtrend, that’s a bearish BOS confirming continuation of the downtrend. In short, BOS is Dow Theory’s “a trend continues” condensed into a single term.

Early Sign of a Shift = CHoCH / Structural Reversal = MSS

CHoCH stands for “Change of Character” and refers to the moment when price first breaks below or above a recent significant high or low in the direction opposite to the prevailing trend. If price breaks below the most recent HL for the first time during an uptrend, that’s a bearish CHoCH — exactly the same phenomenon as Dow Theory’s “the uptrend structure breaks once the most recent low is broken.” CHoCH is the first sign that “the flow may be about to change.”

MSS stands for “Market Structure Shift” and refers to a full-fledged structural reversal — a step more confirmed than CHoCH. Definitions vary somewhat depending on who’s using the term, but it’s easiest to understand as the stage where “after CHoCH gives an early sign, a trend in the opposite direction (a switch in the sequence of highs and lows) has actually begun.” In Dow Theory’s terms, CHoCH is roughly “the moment the pullback low of an uptrend is broken,” while MSS is closer to “the state after that where the highs have also started falling and a downtrend has become established.”

The translation table below summarizes these correspondences in one place. It lines up the terms discretionary traders have traditionally used, their SMC names, and how an automated trading program (EA) handles each one.

Common discretionary term	SMC term	How an EA (automated trading) handles it
Rising high	Higher High (HH)	Numerically compares whether the highest price of the last N candles has been updated
Rising low	Higher Low (HL)	Holds swing lows in an array and compares against the previous value
Falling high / falling low	Lower High (LH) / Lower Low (LL)	Applies the same logic in the downward direction
Trend continuation (previous high/low updated)	BOS (Break of Structure)	Judges a break of the most recent swing high/low by the close and updates a flag
Early sign of a shift (recent high/low broken in the opposite direction)	CHoCH (Change of Character)	Implemented as the condition that flips the trend-direction flag
Full-fledged reversal / trend switch	MSS (Market Structure Shift)	Confirms a subsequent break after CHoCH in a two-stage check
Highs/lows lined up side by side (range edges)	Equal Highs / Equal Lows	Groups clustered nearby highs/lows within a tolerance band

If this table has clicked for you, the door to SMC is already half open. The remaining terms — order blocks, FVGs, liquidity grabs — can be translated in exactly the same way. From here, our SMC/ICT beginner’s roadmap uses this translation table as the entry point into our full library of SMC material.

Useful for Evaluating EAs — How an EA Mechanically Judges “Structure”

Once you understand Dow Theory, the way you see the inner workings of an EA (an automated trading program) changes. When you come across a description like “an SMC-based EA” or “an EA that recognizes market structure,” you’ll be able to tell exactly what it’s calculating under the hood.

An EA can’t look at “peaks and troughs” intuitively the way a human can. Instead, it defines swing highs and lows with numerical rules. A typical approach is: “if a candle’s high is higher than the highs of several candles on both its left and right, treat it as a swing high.” The highs and lows detected this way are recorded in order in an array (a list), and the EA judges whether a new high is higher than the previous high, or whether a new low has broken the previous low, through simple numerical comparisons. BOS and CHoCH are implemented as conditional branches that set or flip a trend-direction flag depending on the result of these comparisons.

This is where the “subjectivity problem” you learned about in discretionary trading comes back into play. The EA’s setting for “how many candles to look at on each side to define a swing” — the so-called period parameter — corresponds exactly to the human dilemma of “which high or low do I use?” Set this value small, and the EA picks up even small troughs and reacts quickly, at the cost of more fakeouts; set it large, and it only looks at large troughs and becomes more stable, at the cost of slower reactions. Dow Theory’s weakness — the tradeoff between lag and fakeouts — carries straight through into how an EA’s parameters are designed. Once you understand this relationship, you’ll be able to infer, just from an EA’s track record, what kind of structural judgment it’s making internally. We cover how to read an EA’s numbers themselves in detail in how to read EA performance (profit factor, max drawdown, and expected value).

Applying This to AI Analysis — What Criteria Does AI Use to Judge Highs and Lows? (Reducing Subjectivity)

Dow Theory’s biggest weakness was the subjectivity of “which high or low do you use.” This is exactly the area where having AI handle the translation adds the most value. What we’re aiming for is using AI to reduce the variance in human subjectivity, and to put our judgment criteria into words and publish them openly.

The difference from a human is that when you have AI judge highs and lows, you can fix the criteria explicitly. Rules like “on the 4-hour chart, a high that exceeds six or more candles on both sides counts as a swing high,” “always judge a break using the close of a confirmed candle,” or “only treat it as a reversal when structure agrees across multiple timeframes” can be applied by AI with the exact same standard every single time. Humans drift in their standards due to fatigue or emotion; AI doesn’t. In our own morning market analysis, AI first extracts the highs and lows across multiple timeframes under one consistent rule set, then puts the structure into words layer by layer — for example, “the higher timeframe shows continuation of the uptrend (BOS), while the lower timeframe shows a break of the recent low (CHoCH), an early sign of a short-term move against the trend” — before publishing it.

It’s worth stressing, however, that AI is not a magic wand that erases subjectivity entirely. The rule itself — “how many candles on each side counts as the standard” — is still designed by a human, so the designer’s judgment remains embedded in it. What AI can do is “apply a fixed standard consistently, without emotion, and with a full record.” The real value lies in being able to preserve both the outcome of that application and the actual subsequent price movement as data. Being able to verify things through both the theory (Dow Theory) and the real data (verification records) is what distinguishes this from rote memorization, and it’s the position we take as a verification-focused media outlet.

Summary

Dow Theory is a map for reading the skeleton of the market using nothing more than the sequence of two points — rising and falling highs and lows. If both highs and lows rise, that’s an uptrend; if both fall, that’s a downtrend; if they stay flat, that’s a range. And when price clearly breaks below the recent pullback low or above the recent rally high, you suspect the flow has changed — that was our basic framework for judging continuation versus reversal. Its weaknesses are the subjectivity of “which high or low to use” and the lag in its signals. Fixing your reference timeframe in advance and judging by the close keeps both of these in check.

And once you run continuation = BOS, an early sign of change = CHoCH, and a full reversal = MSS through this translation, the entrance to SMC — which once looked impenetrable — suddenly feels much closer. An EA judges this structure mechanically through numerical rules, while AI applies a consistent standard to reduce the variance from subjectivity and keeps a record. If you now feel that discretionary trading, SMC, EAs, and AI are all connected by this single point — reading highs and lows — then this page has done its job. That said, none of these are tools that eliminate risk; keep in mind that they’re all meant to be used on the premise of proper money management, including the possibility of losses.

As a next step, we recommend moving on to support and resistance lines (bounces, breaks, and liquidity grabs), which makes concrete exactly where these structural “joints” show up. The “moves that hunt your stop-loss” you’ll learn about there are the very same fakeouts we touched on in this page. If you’d like to survey the full map of SMC first, head to our SMC/ICT beginner’s roadmap; if you’d rather turn your entry and exit points into a repeatable framework, go to stop-losses and money management. You can review the full curriculum from our discretionary trading fundamentals hub.

Frequently Asked Questions

• Q. Can I win using Dow Theory alone?
  A. Dow Theory is a foundation for reading “which flow the market is currently in,” and on its own it doesn’t guarantee precise entries. Because it has weaknesses like signal lag and fakeouts, in practice it’s meant to be combined with support/resistance, price action, and stop-losses and money management. Think of it strictly as a tool for aligning your decision-making framework, not a standalone edge.
• Q. I get confused about the difference between BOS and CHoCH.
  A. It helps to remember it roughly as: “updating a high/low in the direction of the trend = BOS (continuation)” and “breaking the most recent high/low against the trend direction = CHoCH (an early sign of change).” If price clears the previous high during an uptrend, that’s BOS; if it breaks the previous pullback low during that same uptrend, that’s CHoCH. These are just different names for the exact same continuation-versus-breakdown judgment from Dow Theory. We cover this in more depth in our SMC beginner’s roadmap.
• Q. Which timeframe should I use to read highs and lows?
  A. In practice, it works best to fix your reference to the highs and lows that are clearly visible one timeframe above the one you mainly trade. If you day-trade on the 1-hour chart, use the 4-hour highs and lows, for example — deciding on this in advance greatly reduces the hesitation over “which trough do I use?” We explain how to use timeframes as a foundation for this in chart basics.