A third refusal. Is that more evidence?

The same chart, one visit later

In 07.02.01 the illustrative closes rose from about ₹92 to ₹120, fell to ₹108, returned to ₹119, and eased to ₹112. The boundary chosen before the outcome was ₹108, the low between the two highs.

Price did not break it. It rose again to ₹119 and slipped back to ₹110. There are now three highs in the same area, separated by two declines. That is a possible triple top.

Read the picture in colours rather than shapes. Green is a session that closed higher, red one that closed lower. Three green runs reach the same neighbourhood, and each is followed by red. Nothing has broken the dashed line yet.

No, the three highs are not level — and no rule says how close

This drawing shows ₹120, ₹119 and ₹119. That is already tidier than anything you will meet. On a real chart the three highs scatter, sometimes by several percent, and one is usually the highest by a clear margin.

There is no published tolerance. Nobody can tell you that 2% apart qualifies and 3% does not, because the definition contains a judgement and always has. That is not a gap in your knowledge; it is a property of the pattern.

Two things follow, and they are the practical half of this page. If the third high is meaningfully above the other two, stop calling it a triple top — the shape is something else, and 07.02.03 takes the case where the middle one is taller. And because the tolerance is yours to choose, choose it before you see what happened next. A reader who decides afterwards how level counts as level will find triple tops wherever the outcome was convenient.

Why the lower low is the boundary

With two highs there was one low between them and no decision. With three there are two, and choosing the wrong one is the commonest way this pattern gets fudged.

Take ₹112, the shallower one, and the pattern completes sooner — a close at ₹111 would do it. Take ₹108 and the same close proves nothing. Both are defensible until you see the outcome, and only one of them is defensible afterwards.

So the convention here is the stricter line: the low that would be broken last. Write it down when you first name the shape, and do not revisit it once prices arrive.

Three tests tell you the area was visited three times. They do not tell you it will hold a fourth.

You can only see it once it is over, and that cannot be fixed

Here is the question most books avoid. While the pattern is forming, can you identify it?

No. After the first high there is a high. After the second there is a possible double top. The third high is only a third high once price has turned away from it — until then it is simply the latest attempt, and it might be the breakout that ends the discussion.

So a triple top is recognised in the rear-view mirror, always. What you can do in advance is narrower and more honest: notice after the second high that an area is being tested, write down the boundary, and watch. The naming happens later; the decision about the line happens first. That order is the only thing protecting you from fitting a pattern to an outcome you have already seen.

07.05.05 treats this as hindsight bias. It is worth meeting here, because a triple top is where a beginner first feels it.

On a real chart this looks nothing like the drawing

Everything above is drawn clean so the definition is learnable. A real chart is not clean, and the gap between them is where most mistakes live.

Expect all of this. The highs will be ragged rather than level. The two declines will be different sizes, so the two lows will not sit near each other. There will be sessions that poke above the area and close back inside, which a closes-only line cannot even show. The whole thing may take eight weeks or eight months, and on a short timeframe the same shape forms and dissolves constantly.

Three habits carry over regardless. Fix the timeframe first — a triple top on a daily chart is not one on a five-minute chart. Write the boundary down as a price before the next session exists. And check there was a prior rise, which is 07.01.02's requirement, because without one you are looking at a range rather than a reversal.

Learning on tidy drawings and then meeting a real chart is like learning a language from a textbook and then hearing it spoken. The grammar transfers. The fluency has to be built separately, and this book will say so rather than pretend the drawing was the thing.

What Rahul should do next, which is less than he wants

Worth saying plainly, because the honest answer is unsatisfying and a page that leaves it out feels evasive.

Nothing yet. Not because the shape is worthless, but because nothing on this page can tell you where to buy or sell, how much to risk, or where you would be wrong. A chart pattern is an observation about past prices; a trade needs an entry, an invalidation and a size, and those are Part 20 and Part 21.

What he can do now is write two sentences: an area near ₹120 has been tested three times; the boundary is ₹108, decided today. Then watch whether a completed close goes through it. That record is worth more than an opinion, because in a month he can check what actually followed rather than what he remembers thinking.

He can also hold the other argument in mind, the one the books rarely print. Each test consumes whatever was refusing price there. Some readers treat repeated tests as a sign the area is weakening, not strengthening. Both stories fit this chart, which should tell you how much the chart alone settles.

Triple bottom, and the shape it is often confused with

The mirror runs the same way. After a fall, three lows in the same area separated by two rallies make a possible triple bottom, and the boundary is the higher of the two intervening highs — again the one that would be broken last.

But three highs at one level and three lows at another is also the description of a range: price moving between a ceiling and a floor, going nowhere. The drawing does not distinguish them. What distinguishes them is the move that came before.

Without an earlier rise or fall there is nothing to reverse, and calling the shape a triple top borrows a meaning the chart has not earned. A range is a perfectly good thing to observe — it is just not a reversal.