WhitmanTrading

How to Trade a Supply and Demand Zone

To trade a supply and demand zone, find where price left an area fast and draw the zone from the last candle before that move. Wait for price to return to it once, and place the stop beyond the far edge before entering. The size then comes from that stop distance.

Most supply and demand teaching draws a box around a sideways patch and calls it a zone. That is a range, and a range is the one thing a zone is not. This procedure starts from the departure instead.

Before you start

A clean chart on two timeframes. The slower one decides which side you are allowed to take; the faster one is where you draw. Nothing else on the chart — indicators come later or not at all.

One instrument you already watch. Zones are read against how a market normally moves, and you cannot judge “left fast” on something you have never seen before.

A stop distance you can measure before entering. The far edge of the zone plus a margin. You need this in step one because it decides the size, and therefore whether the trade is takeable at all.

The steps

1. Set the direction on the slow chart first

A long-horizon view of the same price series.
The slow chart decides which side you may take at all. Illustrative chart - not real market data.

Higher highs and higher lows means you take demand zones only. The reverse means supply only. This step costs nothing and removes half the zones on the chart before you draw one.

2. Find where price left, not where it stayed

A chart of ordinary price bars with the departure point marked.
A zone is where price left fast, not where it sat. Illustrative chart - not real market data.

Scan for moves that went somewhere fast — several bars in one direction with little overlap between them. The zone sits at the start of that move. Ignore every quiet patch that led nowhere.

3. Draw the zone from the last candle before the move

The same window with the zone drawn from the origin candle.
Draw it from the last candle before the move. Illustrative chart - not real market data.

One candle, not the whole base. Take its high and its low, and that is the box. A zone drawn around twenty bars of chop is a range with a new name, and price will be inside it constantly.

4. Check that anyone was actually there

A candlestick chart with a volume histogram beneath it.
Participation says whether anyone was actually there. Illustrative chart - not real market data.

The departure should carry more volume than the bars around it. A fast move on nothing is a thin market with no one in it, and there are no unfilled orders left behind by absence.

5. Mark it and then leave it alone

A long-horizon view showing untouched zones.
The first touch is the only one the zone was built for. Illustrative chart - not real market data.

An untouched zone is the only kind the reasoning applies to. Once price has returned and traded through it, whatever was unfilled there is gone. Delete zones after their first test.

6. Wait for price to come back to it

A window of price bars approaching a marked level.
Wait for price to come to you, not the other way round. Illustrative chart - not real market data.

You do not chase price into the move. You wait at the origin. Most zones are never revisited, and that is the procedure working rather than failing.

7. Place the stop beyond the far edge before you enter

Price bars with entry and stop levels drawn as horizontal lines.
The stop goes beyond the zone, not inside it. Illustrative chart - not real market data.

Below the low of a demand zone, above the high of a supply zone, plus a margin for noise. A stop placed inside the box sits inside the range the box was drawn to contain.

8. Size from that distance, then take it or skip it

A candlestick chart annotated with the round-trip cost.
Every retest you take costs 2% of a typical bar. Illustrative chart - not real market data.

Divide what you are willing to lose by the stop distance. If the resulting size is absurd, the zone is too wide, and the answer is to skip it rather than to move the stop closer.

How to tell it worked

The test is not whether the trade won. One result tells you nothing, because the shared price series closes higher 10 bars later 54% of the time. That is a coin flip with a slight lean, so a single winning trade is indistinguishable from luck.

Score your last 20 trades against these four counts. How many came from a departure rather than a consolidation. How many were first returns. How many had the stop set before entry. And how many were sized from the stop distance rather than from confidence. Anything under 20 out of 20 on the last two is a procedure problem, not a market problem.

A section of the price series with marked decision points.
Count the zones you skipped, not the ones you took. Illustrative chart - not real market data.

Then count your skip rate over those same 20. Fewer than 10 skipped means the drawing is too loose — on a normal chart most marked zones are never revisited at all. A method producing a trade every session is describing something other than an imbalance.

Why the first touch is the whole idea

The reasoning behind a zone is unfilled orders. Price left fast because one side was overwhelmed and not everyone who wanted to transact got to. The claim is that some of that interest is still sitting there.

That claim can only be true once. When price returns and trades through the area, whatever was left is filled. A second visit is a level with a memory, which is a different and much weaker idea — and treating the two as equivalent is how a chart ends up with forty boxes on it.

Which is why the zones get deleted after their first test. Not because the level stops mattering, but because the specific reason for waiting there has been used up.

The original data

Of the 24,971 unique videos in research/search-study-corpus.jsonl, 18 have an instruction-shaped title mentioning supply and demand, at a median of 89,784 views across 14 channels, with a maximum of 644,762. That median is the second highest of any instructional subject measured in the corpus. The counts come from site/rank_howto.py, which deduplicates by video id before counting anything.

A rising stretch of price bars cut short at the decision bar.
Price is back at the zone. Buy it? Illustrative chart - not real market data.

Eighteen videos against a median of 89,784 is the shape of a real gap. Compare it with TradingView tutorials: 562 instructional videos at a median of 2,575. The crowded subjects are crowded, and this one — with roughly a thirty-fifth of the supply — pulls thirty-five times the audience per video.

The answer to the question on that chart is that the zone does not decide it; the drawing does. If it came from a departure and this is its first return, the plan already exists and the stop is already chosen. If you are asking the question at the moment price arrives, the work was not done beforehand — and that is the actual failure, not the trade.

When it fails

A sideways, range-bound series of price bars.
In a range every bar looks like a zone. Illustrative chart - not real market data.

In a range the method eats itself. Every small push looks like a departure, so every bar becomes a zone, and price returns to all of them constantly. Direction runs on the shared series average 2.01 bars with the longest at 11 — most moves stop almost at once, and in a range each of those brief pushes is a candidate. A chart with a dozen live zones on it is telling you the conditions are wrong, not that there are a dozen opportunities.

A candlestick series containing several opening gaps, with the largest marked.
And a gap can jump the zone entirely. Illustrative chart - not real market data.

The second failure is a gap straight through. Price opens past the zone, the entry never triggers or fills far away, and the stop is already behind. No zone survives a move that happens between bars.

A third is drawing from the base instead of the departure. It produces wide boxes that price is always inside, which makes the stop enormous and the size tiny.

A fourth is trading the second and third touch. The unfilled-order reasoning is spent after the first, and what is left is an ordinary level being treated as a special one.

A fifth is putting the stop inside the box. The box exists because price moves within that area, so a stop there will be touched by the very behaviour you drew it around.

And a sixth is taking a zone against the slow chart. Demand in a market falling on the higher timeframe is a zone in front of a train, which is why that was step one.

Supply and demand explains what the zone is claimed to represent and why the imbalance matters more than the shape. Demand zone covers the buy side on its own, including how it differs from a support line. And support and resistance is the simpler idea underneath, which is worth being fluent in before adding zones on top of it.

What I actually do

The mistake I made for a long time was drawing zones around the area price consolidated in, which is the opposite of the idea. The consolidation is where buyers and sellers agreed. The zone is the last candle before they stopped agreeing. Once I drew from that candle instead of the base, the number of zones on my chart dropped by about two thirds and the ones left were the only ones worth waiting at.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.