Building a repeatable chart-analysis process

In simple terms

Two people can look at the same chart and see completely different things. Even one person can see different things on different days, depending on mood, news and recent wins or losses. A process is a written set of steps that makes your analysis consistent, so that conclusions come from the chart and your rules rather than from your feelings.

This lesson builds such a process as an educational exercise. It does not suggest that trading is a good idea: SEBI's studies, described in the investing compared with trading lesson, found that most individual traders lost money over the periods studied. A process is how you would find out, honestly and cheaply, whether your own approach is any better.

A sample checklist

Write your own, but a typical one might be:

1. Timeframe and context. Which timeframe? What does the next longer timeframe show?
2. Structure. Uptrend, downtrend or range, by your written swing rule, as in the market structure lesson?
3. Levels. Where are the main support and resistance zones?
4. Indicators. Where is the price relative to its moving averages? What do the one or two indicators you use show?
5. Volume. Is recent volume unusual compared with its average? Any mechanical reasons?
6. News and events. Any results, announcements or index changes due?
7. Liquidity and costs. Is the share liquid enough, and what would a round trip cost?
8. Observations. List what you see, factually.
9. Hypotheses. State any testable ideas, each with a failure condition.

Observations and hypotheses

The most important habit is keeping two kinds of statement apart.

• Observation: a fact visible on the chart. "The price is 3% below resistance at ₹520." "RSI is 72."
• Hypothesis: a testable idea about the future, with conditions. "If the price closes above ₹520, it may rise towards ₹560. This idea is wrong if it closes back below ₹505."

Many chart commentaries mix the two: "RSI is 72, so the share is about to fall." The first half is an observation; the second is a prediction presented as a fact. Writing them separately keeps you honest about what you know and what you are guessing.

The journal

A trading journal records, before you know the outcome:

• the date, share and timeframe;
• your checklist answers;
• your observations and hypotheses;
• any decision and the reason for it;

and, afterwards:

• what actually happened;
• whether the hypothesis was right, wrong or not tested;
• costs and the result in rupees and as a share of capital.

Over dozens of entries, the journal tells you how often your hypotheses work, which is the information every trading method needs and most traders never collect.

Hindsight bias

After a move, the reasons seem obvious: "of course it fell, RSI was overbought and it was at resistance". This is hindsight bias: the tendency to believe, after the fact, that an outcome was predictable. It makes us overconfident in our ability to read charts. The journal is the cure, because it shows what you actually thought before the outcome.

Why it matters

Without a process, chart analysis drifts with emotion and memory, and there is no way to know whether it works. A checklist and journal turn impressions into a measurable record. For many people, the honest record is the most valuable result: it shows that their chart reading is less reliable than they felt, before they have lost much money finding out.

Limitations

A process makes analysis consistent; it does not make it correct. A journal records results, but small samples can mislead in either direction, as the expectancy lesson explains. And no checklist removes the costs and competition that make short-term trading hard.

Common mistakes

• Mixing observations with predictions.
• Changing the checklist after each loss.
• Writing journal entries after the outcome.
• Remembering wins and forgetting losses.

Try it yourself

1. Write a checklist of no more than ten steps for analysing a daily chart.
2. Apply it to three shares chosen at random, writing observations and hypotheses separately.
3. Fill in the outcome for each hypothesis after four weeks.
4. Note how often your hypotheses were right, and how that compares with how confident you felt.