What Is Technical Analysis? Dow Theory, Assumptions and Limits
Lesson 1 of 10 · Beginner
What technical analysis means in plain terms
Technical analysis is the study of past price and volume behavior to build probabilistic scenarios about what might happen next. Instead of asking “what is gold really worth?”, a technical analyst asks what are buyers and sellers doing right now, and how has that behavior usually played out before? The main tool is the chart: line charts, bar charts (OHLC) and candlesticks. Each candlestick records four numbers for one time period: the open, the high, the low and the close.
Fundamental analysis looks at underlying drivers instead. For gold that means real interest rates, the strength of the US dollar, inflation, central bank buying and geopolitical risk. The two approaches are not rivals. Fundamentals tend to explain why a market moves, while technical analysis helps you see where and when reactions occur and where risk can be defined.
The key point: technical analysis is not certain prediction. It is a framework for managing probabilities. A well-formed conclusion sounds like this: If price holds above this area, the bullish scenario strengthens; if it does not, that scenario is invalid. A sentence like “gold will definitely rise tomorrow” is not analysis, it is a guess.
The three core assumptions
- Price discounts everything. The assumption is that available information, expectations and even traders' fear and greed are absorbed into price. It is not perfect: a surprise CPI print or a Federal Reserve decision can scramble a chart's structure within seconds.
- Price moves in trends. Markets tend to continue in one direction for stretches, and that tendency underpins every trend-following tool. But long periods of ranging, directionless trade also exist, and telling the two regimes apart is half the job.
- History rhymes, but does not repeat exactly. Human reactions to profit and loss create recurring patterns, so similar shapes appear on charts. The repetition is never exact, and any given pattern only nudges a probability slightly up or down.
All three assumptions are “roughly true,” not “fully true.” A trader who accepts this plans from the start for the case where the analysis turns out to be wrong.
Dow Theory: the root of modern technical analysis
Charles Dow, a co-founder of The Wall Street Journal, wrote a series of editorials in the late nineteenth century about how market averages behave. Others later organized them into what we now call Dow Theory. In modern language, its main principles are:
- The market has three kinds of trend. The primary trend lasts months to years. The secondary (corrective) trend lasts weeks to months and usually retraces a meaningful part of the prior move. Minor swings last days. The modern equivalent of this idea is multi-timeframe analysis.
- The primary trend has three phases. Accumulation, when informed money quietly builds positions; public participation, when the move is obvious and fast; and distribution or excitement, when latecomers buy and informed holders gradually exit.
- Trends are defined by swing highs and lows. An uptrend is a series of higher highs and higher lows; a downtrend is a series of lower highs and lower lows. This is the simplest and most widely used definition of trend.
- Averages must confirm each other. Dow compared the industrial and transportation averages. For gold, the same idea can be applied by checking whether gold and silver are moving together, or by watching gold's usually inverse relationship with the US Dollar Index.
- Volume must confirm the trend. In a healthy trend, volume is heavier in the direction of the trend and lighter during pullbacks.
- A trend is assumed to continue until a clear reversal signal appears. One large red candle does not end an uptrend. A change of view requires a break of structure, for example losing the most recent higher low.
Gold example: reading trend structure
The numbers below are hypothetical and for teaching only. Suppose XAUUSD on the 4-hour chart rallies from 3800 to 3860, pulls back to 3830, makes a new high at 3895, then pulls back to 3855. Both the highs (3860 and 3895) and the lows (3830 and 3855) are rising, so the structure is bullish. From a Dow perspective, as long as price has not closed below the latest higher low at 3855, the default assumption is that the trend continues. If price breaks below 3855 and the next rally then fails to clear 3895, you have the first serious sign of weakening structure.
With gold you also need to respect trading sessions. Asian-session volatility is usually lower, and many of the important moves happen around the London open, the New York open and US data releases such as NFP and CPI (typically 12:30 or 13:30 UTC, depending on daylight saving). A swing high built during thin hours usually carries less weight than one built during heavy participation. It is also better to start from a higher timeframe: the daily structure sets the context, and shorter timeframes only show the detail.
Limitations and fair criticism
- Subjectivity. Two analysts can look at the same chart and draw different lines, see different patterns and reach different conclusions.
- Hindsight bias. Patterns look obvious on a finished chart but are ambiguous while they are forming.
- Mixed statistical evidence. Academic research on the profitability of technical rules is mixed, sometimes contradictory, and the outcome depends on the market, the period, transaction costs and testing method. Test any rule yourself on enough data and with realistic costs.
- Lag. Most indicators are calculated from past prices, so they inherently react late.
- News events. No pattern stops a gap or a news-driven spike. That is why risk management is part of the analysis, not a separate topic.
The aim of this course is to help you understand the classic tools precisely, with their formulas and settings, to know where they work and where they fail, and to learn how to combine them with volume and order-flow data for a fuller picture of the market.
Common mistakes
- Turning analysis into a certain forecast instead of a scenario with a defined invalidation level.
- Ignoring the higher timeframe and treating every small swing as a change of trend.
- Declaring a trend over because of one opposing candle, without a break of swing structure.
- Ignoring the US economic calendar when analyzing gold technically.
- Trusting a rule that was only seen on a few hand-picked examples, without testing it on enough data.
What does volume say here?
Dow himself said volume should confirm the trend, and today's tools make that idea far more precise. The first thing to know about gold: XAUUSD is an over-the-counter (OTC) market, and the “volume” shown on most CFD platforms is actually tick volume, the number of price updates, not the number of contracts traded. Centralized, real volume can be seen in COMEX gold futures (symbol GC), and its levels can be converted to XAUUSD using the current basis, the live price gap between futures and spot/CFD.
With real data you can ask better questions. Was the latest swing high built on genuine participation or during thin hours? Did price get “accepted” in a new area, meaning meaningful time and volume traded there, or was it rejected quickly? Volume profile shows where most trading took place, VWAP shows the average price paid by participants, and delta (the difference between aggressive buying and aggressive selling) shows who is pushing with market orders. None of these replaces price structure; they add a layer of evidence that classic analysis lacks on its own.
📊 See real gold futures volume behind these ideas: free order flow course · today's gold volume map
Frequently asked questions
Does technical analysis work on gold?
Gold is a deep, heavily traded market with broad participation, and technical tools are widely used on it. But “working” means building scenarios and managing risk, not guaranteed prediction. Test any method on historical data and a demo account before trusting it.
Which timeframe should I start my analysis on?
Usually the daily and 4-hour charts for structure and the 1-hour for detail. Very short timeframes carry more noise and are not a good place to start learning.
Is Dow Theory still useful today?
Its core ideas, defining trend by swing highs and lows, nested trends and volume confirmation, still underpin most methods. Its classic signals are late, though, so it is best used as a thinking framework.
Education only, not a recommendation to buy or sell; we never give signals. Price examples are illustrative. Test any method on a demo account first, with strict risk management.