Supply and demand trading marks the price areas where an imbalance between buyers and sellers was large enough to move the market sharply, and treats those areas as places where the imbalance may still exist when price returns. A demand zone is drawn at the origin of a strong move up; a supply zone at the origin of a strong move down.
The idea rests on a simple premise about order flow: a move that leaves a level almost vertically suggests that one side was overwhelmed there, and that not every order at that level was filled. If unfilled interest remains, price returning to the area may meet it again.
That premise is plausible and unprovable. You cannot see the order book that produced a move on a chart from last week, and you cannot know whether anything is still resting there. What you can do is draw the zones by a consistent rule, trade them with a defined invalidation, and measure the results. This page covers how the zones are drawn, how they differ from classical support and resistance, and where the approach is weakest — which is the drawing itself.
How a zone is drawn
The zone is not the extreme of the move. It is the small area of consolidation the move departed from — the last few candles before price accelerated away.
The usual procedure:
- Find a move that is clearly impulsive: several candles in one direction, large bodies, little overlap. Ordinary trending movement does not qualify.
- Look back to where that move began.
- Mark the last opposing or neutral candle or small cluster before the acceleration — the base.
- Draw the zone from the open or body extreme of that base to its wick extreme.
For a demand zone the base sits below the impulsive rally; for a supply zone above the impulsive decline. Zone width is a judgement call, and it is where most of the disagreement between traders lives.
Base shapes are commonly named by what came before the impulse:
| Shape | Structure | Read as |
|---|---|---|
| Drop–base–rally | Fall, pause, sharp rally | Demand |
| Rally–base–rally | Rally, pause, sharp continuation | Demand, in trend |
| Rally–base–drop | Rally, pause, sharp fall | Supply |
| Drop–base–drop | Fall, pause, sharp continuation | Supply, in trend |
The two continuation shapes are generally considered stronger in a trending market, because they are in the direction the market is already going.
Where it differs from support and resistance
The concepts overlap and are not the same.
| Support and resistance | Supply and demand | |
|---|---|---|
| What is marked | A line at a price that has turned the market before | An area at the origin of an impulsive move |
| Evidence required | Prior reactions at the level | One departure, even without a prior reaction |
| Form | A single price | A band with width |
| Repeat touches | More touches usually read as stronger | More touches usually read as weaker |
That last row is the substantive disagreement. Classical technical analysis often treats a level that has held three times as well established. Supply and demand logic treats each retest as consuming some of whatever unfilled interest remained, so a fresh zone — untouched since it formed — is considered the strongest, and a zone that has been tested twice is close to spent.
Both views cannot be right in general, and neither has been established. Pick one, apply it consistently, and let your own records tell you which describes the instruments you trade.
Trading a zone
Entry. Three approaches, in decreasing aggression: a resting limit order at the zone edge; entry after price enters the zone and produces a rejection candle such as a hammer or gravestone doji; or entry after a lower-timeframe structure break inside the zone. The limit order gets the best price and takes every zone including the ones price slices through. The confirmation approaches skip some losers and miss some winners.
Stop. Beyond the far side of the zone, plus a buffer. Not at the zone edge — price frequently trades into a zone before reacting, and a stop at the edge is inside the noise the zone is made of.
Target. Commonly the next opposing zone, or a fixed reward-to-risk multiple. The next-zone approach is more honest about structure; the fixed multiple is easier to measure across many trades.
Zone width drives everything downstream. A tight zone gives a tight stop and therefore a large position for a given risk; a wide zone gives a small one. Because you choose the width when you draw it, you are effectively choosing your position size when you draw the zone — which is a good reason to draw it by rule rather than by feel, and to compute size from the stop afterwards with the lot size calculator rather than deciding the size first.
The weakness: the drawing is discretionary
This is the honest limitation of the method and it is rarely stated plainly.
Give the same chart to five traders who all describe themselves as supply and demand traders and you will get five different sets of zones. They will disagree about which moves count as impulsive, whether the base is one candle or four, whether the zone runs to the wick or the body, and whether a zone touched once is still valid. Each will produce a chart that looks convincing in hindsight.
That flexibility is why the approach backtests so well by eye and so much less well in practice. On a finished chart the zones that worked are obvious and the ones that did not are easy not to draw. The discipline that makes the method usable is writing your rules down before you look — what counts as impulsive, how the base is bounded, how many touches kill a zone — and then applying them without adjustment.
If your rules are written down, backtesting them on historical data means something, because you are testing a rule rather than your hindsight. If they are not, you are testing your ability to remember which zones worked.
How zones fail
Price runs straight through. The most common outcome. Whatever imbalance created the move is gone, or was never the reason for it. The stop beyond the zone is what makes this survivable.
The zone is overrun and then reverses. Price trades through your stop and turns. Widening the zone would have saved the trade and would also have made every other trade smaller. There is no setting that avoids this.
The zone was drawn after the fact. The most damaging failure, and invisible while it is happening: the base was identified because price had already reacted there. Marking zones only on the hard right edge, before the reaction, is the only defence.
Trend overwhelms it. A demand zone in a persistent downtrend gets consumed. Zones read in the direction of the higher-timeframe trend are on firmer ground than zones traded against it.
Nobody can tell you what proportion of zones hold. It depends on how you draw them, which market, which timeframe, and which entry rule — a set of choices specific enough that only your own records answer it.
Making it measurable
Record each zone trade with the timeframe, the base shape, the zone width in pips or points, whether the zone was fresh or retested, whether you used a limit or waited for confirmation, and the result in R.
Those fields answer the questions the approach actually turns on: whether fresh zones outperform retested ones for you, whether your limit entries beat your confirmed ones, whether continuation zones beat reversal zones. Fips's trading journal records trades in R and groups them by setup so these comparisons are a filter rather than a spreadsheet, and account analysis shows the distribution rather than the average — with an approach this dependent on discretionary drawing, the spread of outcomes is more informative than the mean.
Position sizing deserves the same discipline. Zone width varies trade to trade, so a fixed lot size means wildly varying risk. Fixing the risk percentage and letting the size follow the stop is the correction; the arithmetic is in risk management and the calculation on the lot size calculator.
Frequently asked questions
Is supply and demand different from support and resistance?
They overlap. Support and resistance marks price levels that have produced reactions; supply and demand marks areas that impulsive moves departed from, whether or not price has returned. The sharpest practical difference is that supply and demand treats a fresh, untested zone as strongest, while classical analysis often treats a repeatedly tested level as strongest.
How wide should a zone be?
From the base's body extreme to its wick extreme is the common rule. Wider zones catch more entries and force smaller positions for the same risk; tighter zones do the opposite. What matters more than the exact rule is using the same one every time, because zone width sets your stop and therefore your position size.
Does a zone stop working after it is touched?
In the logic of the method, yes — each touch is assumed to consume some of the unfilled interest, so many traders treat a zone as spent after one or two reactions. This is a premise of the framework rather than a demonstrated fact, so it is worth tracking in your own records rather than assuming.
Which timeframe should zones be drawn on?
Higher timeframes produce fewer zones with more participants behind them. A common approach is to mark zones on a 4-hour or daily chart and refine entries on a lower one. Zones drawn on very low timeframes are numerous and mostly noise.
Can I automate it?
Partly. The impulsive-move detection can be coded with thresholds on candle size and overlap, but the judgement about which base to use and how wide to draw it resists specification — which is a useful signal about how discretionary the method really is.