Point & Figure

Renko was brilliant at deleting time, but it had a stubborn flaw: it prints a new brick the moment price moves down by one brick, even if that dip is a trivial breath inside a huge uptrend. If you are riding a multi-year rise, you do not want to be spooked out of your position by a single red brick that means almost nothing.

So the question sharpens. You want a chart that ignores small pullbacks completely - not just softens them - and only raises its hand when a genuine reversal is underway. You want the filter to be about the size of the counter-move, not a fixed one-step rule. How would you build that?

Long before computers, Western traders tracked prices on graph paper with nothing but Xs and Os. This is Point & Figure charting - arguably the oldest charting method in the West, and still quietly used today.

A Point & Figure chart draws rising prices as a column of Xs and falling prices as a column of Os. Its whole purpose is to separate continuation from reversal - and, crucially, the asymmetry it enforces is between those two, not between up and down. While a column of Xs is rising, it extends on every fresh high but ignores small pullbacks until they reach a set threshold. While a column of Os is falling, it does the exact mirror: it extends on every fresh low and ignores small bounces until they hit the same threshold.

That symmetry matters, because beginners often assume P&F is somehow bullish. It is not. It has no preference for up or down. It simply refuses to switch direction until the counter-move is large enough to count. It favours whichever way the current column is already going - and only that.

P&F has two settings, not one:

1. The box size - how far price must move to add a new mark to the current column (say ₹2).
2. The reversal amount - how many boxes price must move against the column before the chart admits a change and starts a new column (usually 3).

The rules then run like this. Suppose you are in an X column (going up) with a ₹2 box and a 3-box reversal. Every ₹2 higher stacks another X on top. But if price drops ₹2, nothing happens. Drops ₹4 - still nothing. The chart simply ignores the fall. Only when price drops a full 3 boxes (₹6) does the chart finally accept the reversal: it steps one column to the right and prints three Os going down.

Look at those raw candles and you can already sense what P&F will do with them: price sags back toward ₹100 twice and turns each time, then climbs away. A Point & Figure chart of the same history snaps those two turns into O columns that both bottom at the ₹100 box - drawing a clean, obvious floor that the wicks and gaps of the candlestick chart smear across a few slightly different prices. That level-clarity is P&F's signature strength. Every price here is an illustrative example, not a real quote. illustrative

Below is a fixed example: one price series drawn as P&F with a ₹2 box and a 3-box reversal. X columns are rising boxes, O columns are falling - and small counter-moves simply never appear at all. Watch how much of the wiggling in a normal chart just vanishes.

The thing to feel here is how quiet the chart is. Long stretches of the underlying price produce a single column, because nothing crossed a threshold. That silence is the filter working - but it is also information being deliberately withheld, and the later sections are about the cost of that bargain.

P&F earns its keep in a couple of very specific jobs, and is actively misleading in others.

Where it helps - finding levels. This is what P&F does better than almost anything. A daily candlestick chart of a bank over the last year might be a messy tangle of wicks. Switch to P&F and you may see five separate O columns that all stopped at the exact same box - a clean, repeated floor of support drawn without any guesswork about wicks. Because time is squeezed out, every test of a level lands at the same horizontal row, so the level jumps out at you.

Where it helps - sitting through a false scare. A stock drops sharply on a bad headline, but not enough to trigger a 3-box reversal. Panicky traders on a candlestick chart sell into the noise; the P&F chart does not even flinch, keeping you in the X column and in the trade. When the scare was genuinely just noise, that discipline is worth a great deal.

Where it hurts - anything about the live price or timing. Because P&F ignores small pullbacks and squeezes out time, the last printed X can sit far above where the stock is actually trading right now, and a single column can hide anything from three frantic days to two grinding years. If you need to act on the current price, P&F is quietly lying to you by omission - and the next sections show how.

It cannot tell you when. Because P&F ignores time, one column of Xs could be three days of frantic buying or two years of slow grind. The chart has no way to say which, so any question that starts with "when" or "how fast" is unanswerable here.

It cannot tell you about the near-misses. A stock can fall 90% of the way toward a reversal threshold, then shoot back to new highs, and the chart will show one unbroken column of Xs. You would never know the buyers almost lost control - the filter erased the scare along with the noise. A P&F support level that has held five times is odds in your favour, never a wall that cannot be breached.

It cannot tell you it is safe to skip the business. P&F is a beautifully clean map of price, and a clean map tempts you to forget it is only a map. The columns say nothing about earnings, debt, or whether the company is any good. The Xs and Os describe where price has been; they cannot tell you whether the company behind them is worth owning at all.

The most common and most expensive trap is misreading the current price.

Because P&F ignores small pullbacks, the last drawn X on your screen might sit at ₹106 while the live price has quietly slid all the way to ₹101. The chart will not show that drop until price reaches ₹100 and completes the 3-box reversal. If you are using a P&F chart to manage a tight stop-loss, this delay will hurt you - and if price gaps down through several boxes on news, the O column that finally prints can be far below where you imagined your exit was. A reversal box confirms the fall only after it has happened, and a gap can open well past it.

The deeper trap is trusting the filter itself. The same box-and-reversal rule that removes noise can also remove a real early warning.

Before you trust an unbroken column, ask the reader's question: what would change my mind? If the live price is sitting far below the last plotted box and volume is climbing on the way down, the "still in an X column" read deserves real doubt - whatever the chart happens to print.

Point & Figure is the ultimate noise filter, but it leans on an arbitrary box size to decide what counts. What if we did not want arbitrary boxes at all? What if a reversal were defined not by a set number of rupees, but by whether price broke the previous swing low the market itself had made? That brings us back to Japan, and the Kagi chart.