Technical Analysis: Charts, Trends and Indicators

Technical analysis studies the history of price and volume in order to form a view about what price might do next. It makes no attempt to value the business. Where fundamental analysis asks what a company is worth, technical analysis asks what the market is doing.

It is widely used and genuinely contested. This lesson explains the tools and then sets out the objections to them fairly, because a lesson that presented only the case in favour would not be education.

What technical analysis claims

The approach rests on three assumptions. That price already reflects everything known about a security. That prices move in trends which tend to persist for a time. And that patterns in participant behaviour repeat, because the psychology behind them repeats.

Each of these is arguable, and the third is the weakest. But the tools built on them are used by enough market participants that they influence behaviour regardless of whether the underlying theory holds — which is itself a reason to understand them.

Chart types

A line chart joins closing prices. It is the clearest way to see a long-run direction and discards everything else.

A bar chart shows open, high, low and close for each period.

A candlestick chart shows the same four values in a form that is faster to read, and is what most people mean by a stock chart.

Reading a candlestick

Each candle covers one period — a day, an hour, a week. The body spans the opening and closing prices. The thin lines above and below, the wicks, reach the high and low.

Conventionally a candle whose close is above its open is drawn light or green, and one closing below its open is drawn dark or red. A long body means the period ended far from where it started. A short body with long wicks means price travelled a long way and came back — indecision rather than direction.

That is the whole vocabulary. Everything else in candlestick analysis is interpretation layered on top of these four numbers, and the interpretations are considerably less reliable than the numbers.

Trends

An uptrend is a sequence of higher highs and higher lows. A downtrend is lower highs and lower lows. A range is neither: price oscillating between roughly constant boundaries.

Trends are identified after they have begun and are only definitively identified after they have ended, which is the central difficulty of the entire method. A trendline drawn on a chart looks obvious in hindsight and is a judgement call in real time.

Support and resistance

Support is a price area where buying has previously been sufficient to halt a decline. Resistance is where selling has previously halted an advance.

The behavioural explanation is straightforward enough: people remember the price at which they bought or wished they had, and act on that memory. When a resistance level is decisively broken it is often said to become support, and vice versa.

The honest caveat is that these levels are areas rather than precise prices, and identifying them involves judgement. Two competent analysts will draw different lines on the same chart.

Volume

Volume is the number of shares traded in a period. It is the closest thing technical analysis has to corroborating evidence, because it is a count rather than an interpretation.

A price move on unusually heavy volume reflects genuine participation. The same move on very light volume may reflect little more than a few trades in a quiet market. In thinly traded shares this matters enormously: a dramatic-looking chart can be the product of a handful of transactions, and it will not be repeatable when you try to trade against it.

Moving averages

A moving average plots the average closing price over a defined number of periods, smoothing short-term noise to make direction more visible. A 50-period and a 200-period average are the most commonly watched.

Their essential property is that they lag. A moving average is a summary of what has already happened, and it will always turn after the price does. This is not a flaw to be corrected — it is what an average is — but it means moving-average signals arrive late by construction.

Crossovers between a shorter and a longer average are widely watched and widely reported. They generate a large number of false signals in a market that is moving sideways, which is a substantial fraction of the time.

RSI and MACD

The Relative Strength Index compares the size of recent gains to recent losses and expresses it between 0 and 100. Readings above 70 are conventionally called overbought and below 30 oversold.

The word "overbought" causes real damage to beginners. It does not mean a share is about to fall. In a strong trend the RSI can remain above 70 for weeks while the price continues to rise, and selling on that basis alone has cost people a great deal.

The Moving Average Convergence Divergence compares two moving averages of price and plots the relationship along with a signal line. It is used to gauge momentum and its changes. Being built from moving averages, it inherits their lag.

A general point applies to every indicator: they are all calculated from price and volume. An indicator cannot contain information that is not already in the price history. Adding six indicators to a chart does not add six sources of evidence — it adds six rearrangements of the same evidence, and the resulting appearance of confirmation is illusory.

Chart patterns

Patterns are recurring shapes to which conventional interpretations are attached: head and shoulders, double tops and bottoms, triangles, flags.

They are worth knowing because they are widely watched, and they are worth treating sceptically for two reasons. Human beings are extremely good at finding patterns in random data — this is a well-documented cognitive bias, not a criticism of any individual. And pattern identification is subjective enough that the same chart supports several readings, which makes claims about their reliability difficult to test.

Timeframes

The same security produces entirely different pictures on different timeframes. A share in a clear downtrend on a daily chart may be in an uptrend on a weekly one.

Neither is more correct. They answer different questions. What causes trouble is switching timeframes to find one that supports a conclusion already reached — which is a very easy thing to do without noticing you are doing it.

The serious objections

These are not minor caveats and any honest treatment has to state them.

The academic evidence is not supportive. A large body of research finds that price movements are close to unpredictable from past prices alone, and that trading rules which appear profitable in historical testing frequently fail out of sample.

It is unusually vulnerable to hindsight. Every pattern is obvious on a chart of the past. The test of a method is what it says about the right-hand edge, before the outcome is known, and that is far harder than any illustration suggests.

It costs money to be wrong. Technical approaches typically generate frequent trading, and every trade carries commission, taxes and the bid-ask spread. A method that is right slightly more often than chance can still lose money after costs.

Thin markets distort it. In illiquid shares, chart formations can reflect a few trades rather than any collective judgement, and are correspondingly meaningless.

Using it responsibly

Given all of that, what is defensible? A few things.

Charts are an efficient way to see what has happened to a price and over what period — that is description, and it is reliable. Volume genuinely distinguishes a well-participated move from a thin one. Knowing where widely watched levels sit tells you where other participants may act, whether or not the theory behind those levels is sound.

What is not defensible is treating an indicator as a forecast, trading on a pattern without regard to what the company is, or increasing position size because a chart looks convincing. Whatever method leads you to a decision, the size of the position and the point at which you accept you were wrong are what determine the damage — and those belong to risk management, not to analysis.

Frequently asked questions

A balanced position

Charts are a useful way to see what has happened and how much participation was behind it. They are a poor basis for confident claims about what happens next, and the academic evidence on that point is not favourable.

An investor who uses charts to observe and fundamentals to decide, while letting position sizing determine the consequences of being wrong, is on considerably safer ground than one who treats an indicator as a signal.

Continue learning

• Fundamental Analysis — valuing the business rather than reading the price.
• Risk Management — what actually determines the cost of being wrong.
• Investment Strategies — the recognised approaches and what each is trying to do.
• Financial Glossary — definitions for the terms used above.