Technical analysis is the study of price and volume history to make judgements about what price is likely to do next. It ignores what the company sells, who runs it, and whether the business is any good. To someone who has just learned to read a balance sheet, that sounds absurd. It is worth understanding why it is not.
The three assumptions
- 1Price discounts everything
Every fact, rumour, forecast and emotion held by every participant is already expressed in the price. You do not need to know why a large fund is accumulating — the accumulation shows up as rising price on rising volume regardless. The chart is the aggregate opinion of everyone who has money at stake.
- 2Prices move in trends
Price is not random noise around a fair value. It develops directional persistence, because information spreads gradually and money moves in size over days and weeks, not instants. A trend, once established, is more likely to continue than to reverse — until it is not.
- 3History repeats, because people do
Chart patterns are not magic geometry. They are the visual signature of recurring human behaviour: hesitation at a level where people previously lost money, capitulation when a support finally breaks, greed accelerating into a top. Markets change; the emotions of the people in them do not.
The honest objections
Any serious treatment has to state the case against. There are three good arguments, and they deserve real answers rather than dismissal.
| The objection | The fair response |
|---|---|
| If patterns worked, they would be arbitraged away. Anyone can see them. | Partially true, and this is why simple published systems degrade over time. What does not get arbitraged away is discipline — most people who see the pattern still fail to act on it consistently. |
| You can find any pattern you look for. Human beings see faces in clouds. | Entirely true, and the single biggest failure mode in practice. The defence is to define rules in advance and test them, rather than narrating a chart after the fact. |
| Backtests are riddled with hindsight bias. Every strategy looks brilliant on past data. | True, and unavoidable without discipline: out-of-sample testing, honest cost assumptions, and accepting that a strategy with 40 parameters has been fitted to noise. |
Where technical analysis works best
- Timing. You have decided a company is worth owning; a chart helps you avoid buying into a falling knife.
- Exits. Fundamentals rarely tell you when to sell. Structure does.
- Risk definition. A chart gives you an objective, non-negotiable point at which your idea is wrong.
- Liquid instruments. Indices and largecaps, where thousands of participants make the crowd behave like a crowd.
- Illiquid smallcaps. Twelve trades a day is not a crowd; it is a handful of people, and patterns there are noise.
- Predicting news. No chart foresaw the 2020 lockdown announcement.
- Deciding what to own for a decade. That is a business question.
- Anything on a one-minute chart without a genuine edge. Costs and randomness dominate.
You arrive at a station and see everyone on the platform shuffling towards the far end, luggage in hand, with growing urgency. You do not know why. But you can reasonably infer that the train is expected to stop there — and the longer you watch the same movement, the stronger the inference. If a few people start drifting back, you notice that too.
That is technical analysis. You cannot see the announcement board; you can see the crowd reacting to it. Volume tells you how many people are moving, price tells you which way, and structure tells you whether the movement is continuing or losing conviction.
A stock jumps 8% on massive volume with no public news. What does a technical analyst conclude?
Naye sheher mein do restaurant hain — ek khaali, ek mein line lagi hai. Aapne dono ka khaana nahi chakha, par aap line waale mein jaate ho. Technical analysis yahi hai: aap company ka khaana nahi chakh rahe, bheed dekh rahe ho. Kabhi bheed sahi hoti hai, kabhi sirf bheed hoti hai.
- Technical analysis rests on three claims: price reflects everything, trends persist, and crowd behaviour repeats.
- It is a probability and risk framework, not a prediction machine.
- Its greatest weakness is pattern-hunting after the fact — rules must be defined in advance.
- It works best in liquid instruments and worst in thin ones.
- Its most valuable contribution to most investors is timing and exits, not stock selection.
Mark it done to track your progress through the curriculum.
Common questions
Short, direct answers to what people ask about this topic.
- what are the three assumptions of technical analysis
- Technical analysis rests on three claims: that price already discounts everything known and felt about a company, that prices move in trends rather than wandering randomly around a fair value, and that chart patterns repeat because the human behaviour producing them repeats. Remove any one of the three and the rest of the discipline stops making sense. None of them is proven — they are working assumptions, and each holds better on some timeframes than on others.
- price discounts everything means that
- Every fact, rumour, forecast and emotion held by everyone with money at stake is already expressed in the traded price. The practical consequence is that you do not need to know why a large fund is accumulating — the accumulation shows up as rising price on rising volume regardless. It is also the assumption that breaks most visibly when genuinely new information arrives, which is why prices gap.
- dow theory meaning in stock market
- Dow Theory is the set of principles drawn from Charles Dow’s newspaper editorials at the end of the 1800s, and it is the origin of most modern technical analysis. Its core ideas are that the averages discount everything, that a market moves in a primary trend with smaller corrections inside it, and that a trend should be assumed to continue until there is definite evidence it has reversed.
- does technical analysis ignore a company’s fundamentals
- Yes, deliberately. A chart reader does not ask what the company sells, who runs it or what it earns; the argument is that everyone who does care about those things has already acted, and their actions are visible in price and volume. That is a real limitation as well as the source of the method’s speed — a chart cannot tell you a business is fraudulent, only that people are selling it.
- does technical analysis work if the market is efficient
- The weak form of the efficient market hypothesis says past prices cannot be used to predict future returns, which contradicts technical analysis directly. The honest position is somewhere in between: markets are efficient enough that obvious, easily coded edges get competed away quickly, but not so efficient that price always equals value, which is why trends and levels persist at all.