Technical analysis: what charts can and cannot tell you

What it is, and what it assumes

Technical analysis studies the chart. It ignores profit, debt and management, and looks only at how the price has moved and how many shares changed hands doing it.

The reasoning behind it is not unreasonable. Prices are set by people. People behave in partly repeatable ways under greed and fear. So patterns in past price behaviour might carry some information about future behaviour.

The important limit follows immediately: technical analysis can only tell you about behaviour, never about value. A chart cannot tell you whether a company is profitable, whether its debt is manageable, or whether its auditor resigned. Anyone using charts alone to make long term investment decisions is deliberately ignoring most of the available information.

The ideas worth knowing

Support and resistance

Support is a price level where buying has repeatedly appeared. Resistance is a level where selling has repeatedly appeared.

These are not barriers in any physical sense. They are records of where people previously decided to act, and the reason they sometimes hold is that people remember those levels and act there again. That is also why they stop working once enough participants change.

Trend

A sequence of higher highs and higher lows is an uptrend. The reverse is a downtrend. Neither is a prediction. It is a description of what has been happening, and trends end without warning, usually at the moment they feel most established.

Moving averages

A moving average is simply the average closing price over a recent period, recalculated each day. A 50 day average responds quickly to changes. A 200 day average is slower and smoother.

They are useful for one honest purpose: removing daily noise so you can see the general direction. Every moving average is, by construction, behind the current price. It describes the past accurately and predicts nothing.

Volume

Volume is how many shares traded. It is arguably the most underrated element in technical analysis, because it tells you how many people were involved in a move.

A 6% rise on ten times the usual volume means a great many participants acted. The same 6% on almost no volume may just mean a couple of large orders met a thin order book. Volume is also practical information: a thinly traded stock is one you may struggle to sell at a fair price.

Where the scepticism is justified

Technical analysis attracts both uncritical belief and blanket dismissal. The accurate position is in between, and it helps to know exactly which criticisms have force.

• We see patterns in randomness. Human beings find shapes in noise reliably. Generate a random price chart and you will spot heads and shoulders, triangles and double bottoms in it. That a pattern is visible tells you almost nothing about whether it means anything.
• Hindsight makes everything look obvious. Any chart annotated after the fact shows patterns that worked. The ones that failed are simply not circled. Almost all teaching material has this bias built into it.
• More indicators is not more insight. Most indicators are transformations of the same price and volume data. Six of them agreeing is not six pieces of evidence, it is one piece of evidence displayed six times.
• If a pattern works reliably, it gets traded away. Participants with better systems will act on it before you can, until the advantage disappears. Genuine edges do not survive being widely known.
• It says nothing about the business. A chart cannot show you rising debt, a qualified auditor report, or a lost major customer.

Where it is genuinely useful

Having stated the limits, there are real uses, and they are narrower and more practical than usually advertised.

• Deciding where to place a stop loss. If you are trading, a chart gives you a defensible level at which your idea is demonstrably wrong. This is risk control, not prediction, and it is the strongest use.
• Judging liquidity. Whether you can enter and exit a position without moving the price yourself is a practical question a chart answers well.
• Timing an entry you had already decided on. If your analysis says you want to own something, a chart may help you avoid buying into a sharp move. It should never be the reason for the decision itself.
• Recognising when you are early. A share you consider undervalued that keeps falling on rising volume tells you the market disagrees for now. That is information, even if you still think you are right.

What links all four: technical analysis is being used to manage execution and risk, once the decision about what to own has been made elsewhere.

A plain conclusion

AlphaVik states views rather than sitting on fences, so here is one.

For a long term investor, technical analysis is not necessary. If you own a broad index fund through a monthly SIP, charts have nothing useful to add. Your outcome depends on the businesses and on your own behaviour, and no chart improves either.

For someone swing trading, the risk control parts are genuinely valuable. Knowing where your idea is wrong, and sizing accordingly, is most of what separates surviving traders from the majority who do not.

The failure mode to avoid is using charts as a substitute for understanding what you own. A share is a piece of a business. If your entire reason for holding it is the shape of a line, you have no way to judge a fall, because you never had a reason that a fall could contradict.