Kagi

Point & Figure defines a reversal with an arbitrary box size - drop three boxes and the trend is declared over. But what if the market's own rhythm does not fit your box? Suppose a stock habitually pulls back ₹15 before rallying again, and you have set your reversal at ₹10. You will be thrown out of a perfectly healthy trend, over and over, by pullbacks the stock was always going to make.

So here is the question. Instead of imposing a number on the market, could a chart define a reversal from the market's own behaviour - from the actual swing highs and swing lows price keeps making - rather than from a formula you picked in advance?

The Kagi chart - the word means "key", after an old L-shaped key - was developed in Japan in the 1870s to track the young Japanese stock market. Like Renko and Point & Figure, it throws away time. But instead of bricks or Xs, a Kagi chart draws a single continuous line that snakes left to right, and its cleverness lives not in the line's shape but in its thickness.

The line is always in one of two states:

1. Yang - a thick line, meaning demand is currently in control.
2. Yin - a thin line, meaning supply is currently in control.

The idea is to encode the balance of buyers and sellers directly into how the line looks, so that a single glance tells you who is winning right now. Not who will win - who is winning, according to the structure price has actually built.

There are really two rules working at once, and it helps to keep them separate.

First, the line follows price to each new extreme and only turns when price reverses by a set amount - the reversal amount (say ₹4). Small wiggles smaller than that never register; the line keeps heading the way it was going. This is the noise filter, and it works much like Renko's brick or P&F's box.

Second - and this is the part unique to Kagi - the line changes thickness based on swing points:

• It becomes yang (thick) the moment price breaks above the previous swing high.
• It becomes yin (thin) the moment price breaks below the previous swing low.

So thickness is about structure (are we making higher peaks or lower troughs?), while turning is about the reversal amount (has price moved enough to bother drawing a turn?). Every price here is an illustrative example, not a real quote. illustrative

On those candles the stock climbs to about ₹130, then rolls over and eventually undercuts an earlier trough. A Kagi chart of the same path stays thick (yang) all the way up - through the small pullbacks that never break a prior low - and thins to yin only at that final point where a genuine swing low is taken out. The moment of thinning, not the day of the high, is what Kagi is built to flag.

Below is a fixed example with a ₹4 reversal. The line runs vertically to each new extreme and only turns after price reverses ₹4; it thickens (yang) when it breaks the prior peak and thins (yin) when it breaks the prior trough. Follow the thickness, not just the direction.

Watch how the line can be heading down while still thick - falling toward a low it has not yet broken. That is the whole point of separating direction from thickness: price can retreat a long way and the structure can still, technically, be intact. Whether that is reassuring or dangerous is exactly the tension we come to below.

Where it helps - a mechanical exit for a long-term hold. This is Kagi's best use. You buy when the line turns yang (breaks a prior high) and decide you will only sell when it turns yin. For two years the stock rises, pulls back, rises, pulls back - and because it never breaks a prior low, the line stays thick and yang the whole way. You are given a clear, unemotional rule that keeps you in a good trend and stops you panic-selling every dip. The line holds yang as long as the structure of higher lows survives, which is the trend quietly persisting.

Where it helps - knowing when to leave a stock alone. In a tight, choppy range, the Kagi line just bounces around without breaking highs or lows, staying yin. That is a useful "not yet" signal: there is no structure to trade until something actually breaks.

Where it hurts - a runaway, one-way trend. Kagi's reliance on the previous swing low is a virtue in normal trends and a curse in violent ones. If a stock rockets up without ever pausing to form a swing low, the last low can be left far, far below the current price - and then a savage fall can wipe out most of your gains before the line finally thins. The next section spells out how brutal that can be.

It cannot tell you when. Like Renko and P&F, Kagi has no clock. A single horizontal connector could bridge three days or three months - the chart cannot say which, so timing questions have no answer here.

It cannot tell you how far a break will go. When the line turns yang by clearing a prior high, that break may fail on the very next swing, plunging back and thinning to yin. A fresh yang line is a shift in the odds, not a promise the move continues. Kagi describes what just happened to the structure; it forecasts nothing about the next leg.

It cannot protect a runaway profit. Because it waits for a prior swing low to break, a stock that ran from ₹100 to ₹300 with no intervening low can crash all the way to ₹150 and still keep the line yang - the last real low is still down at ₹100. You could watch half your gains evaporate while the chart calmly refuses to give you permission to sell. A chart that only exits at an old, distant swing low can hand back most of a winner before it acts.

The first way people are fooled is the runaway-trend problem just described: they trust Kagi as a trailing stop without noticing how far behind the "previous low" has fallen, and they take a far bigger hit than they ever intended before the line thins.

The second, subtler trap is the thickness itself. A fresh yang line feels like a promise, and it is not.

Before you trust a fresh yang line, ask the reader's question: what would change my mind? If the thick line holds above the broken peak and then prints a higher swing low, the strength reading gets stronger. If it thins back to yin below that peak within a swing or two, the failed-breakout reading wins.

Kagi fixes the arbitrary-box problem by leaning on the market's own swing highs and lows. But it stumbles in huge, uninterrupted trends, where the "previous low" can be left far below. What if we wanted a chart that tightened its reversal rule as the trend matured - pulling its safety net closer using only the most recent blocks, so a turn could trigger faster during a steep rally? That brings us to the Three-Line Break chart.