Charts, indicators, and their limits

What a candle records

A candlestick summarizes four numbers for a chosen interval: the price at the start, the highest and lowest reached, and the price at the end. It is a compression of everything that happened in that window into four values.

The interval is a choice, and it changes the picture substantially. The same week rendered in one-minute candles and in daily candles produces two charts that support different narratives. Neither is more true; they answer different questions. Be suspicious of any chart presented without its interval and venue clearly stated, because both were selected by whoever is showing it to you.

Volume records how much traded in the period. It adds useful context about activity, and it does not reveal motive. A large volume bar tells you that many units changed hands, not that informed participants were buying.

Indicators are recipes applied to past prices

Every indicator is a formula over historical data. A moving average takes the mean of recent prices, which smooths short-term noise and necessarily lags. A momentum indicator such as RSI compares the size of recent gains to recent losses and expresses it on a bounded scale. Neither incorporates any information from outside the price series.

This is the key limitation, and it is structural rather than a matter of tuning. An indicator cannot know about an exploit, a listing, a regulatory action, or a large holder's intention. It can only restate the price history in a different form. When an indicator appears to anticipate something, it is because price began moving before the news was public—the indicator followed price, as it always does.

Support and resistance are interpretations of past behaviour rather than physical properties. They can be self-fulfilling to a degree, since many participants watch the same levels, but they are not barriers and they do not hold reliably.

The hindsight problem

Chart-based claims are unusually easy to make convincingly after the fact. Once the outcome is known, it is trivial to select the indicator, interval, and starting point that would have predicted it, and the resulting illustration looks compelling.

A method that means anything states its rule, its costs, its sample, and its failure cases before the test. Ask of any chart argument: what would this rule have said at every other point, not only at the ones being shown? How many times did it signal wrongly? Were fees and slippage included? An approach that cannot answer those questions is a story, not evidence.

A reasonable role for charts

None of this makes charts useless. They are a good way to see the range an asset has traded in, how violently it has moved, whether liquidity has thinned, and how the present compares to the past. Used that way, a chart is a description of conditions, and description is genuinely useful.

What it should not do is substitute for risk limits or research. A beginner's decisions—how much to hold, in what custody, with what maximum loss—should be made from the analysis in the rest of this course. If a chart is deciding position size, the position is being sized by the least informative input available.

• Note the interval and venue before reading any chart.
• Ask what information an indicator could not possibly contain.
• Require the rule to be stated before the outcome is known.
• Keep position sizing separate from chart reading.