Three boards undone by one.

Three sessions, then one that undoes them

Monday, Tuesday and Wednesday each close higher than the one before: ₹30.50, then ₹30.95, then ₹31.40. An ordinary short run of buying.

Thursday opens at ₹31.55 — above Wednesday's close, so the run appears to be continuing — and then falls all day to ₹29.90. That last number is below Monday's opening price. One session has covered the whole of the three before it.

That is a three-line strike. Four boards: three going one way, and a fourth that opens beyond the third and closes past the first.

Why it is called bullish when Thursday is a long red board

This will look like an error the first time you meet it, so it is worth stating plainly: the sequence above is a bullish three-line strike.

The name describes the three, not the fourth. Three rising boards make it bullish; the falling board that follows is, in the traditional reading, an interruption rather than a reversal. The mirror — three falling boards and then one long rising one — is a bearish three-line strike.

So the pattern is named for the run it interrupts and filed as a continuation: the books read the fourth board as profit-taking inside a rise that will resume. Hold that thought, because it is the part that does not survive contact with evidence.

A pattern named for the move it destroys is telling you the naming came before the testing.

Where the books and the testing disagree

Granthara's rules engine carries a note on this pattern that it carries on no other. Its recorded bias is not bullish or bearish but disputed, and the note reads: traditional books read it as a pause in a rise; tested on real charts, it often turned lower instead.

That is a direct contradiction, and it is worth understanding what kind of disagreement it is. Nobody disputes the geometry — four boards, measured the same way by everyone. What is disputed is the claim attached to the geometry, and that claim came from observation and tradition rather than from counting.

When somebody did count, the continuation reading did not hold up. So the pattern sits in a chapter called Continuation and Multi-Candle Formations while quite possibly not being a continuation pattern at all. Granthara has left it there, and labelled it, rather than quietly moving it.

What to do with a disputed pattern

Not discard it. The four boards record something real: a run of three sessions, and then a session large enough to undo all of them. That is a fact about the week, and a session that covers three others is genuinely unusual.

What you drop is the direction. You may say the rise of the last three sessions has been erased in one. You may not say therefore it resumes, because the only people who checked found otherwise.

And there is a general lesson underneath, which 04.05.04 will make properly. A pattern's name, its chapter, and its traditional meaning are three claims of very different quality. The name is a convention. The geometry is checkable. The meaning is an empirical claim that somebody has to have tested — and for most of the catalogue, either nobody did, or the result was quietly not repeated.