Indicator library · Levels
Pivot Points From a Completed Session
Support and resistance projected from three numbers: yesterday’s high, low and close. Nothing about them adapts once today begins, which is both the whole of their appeal and the reason a touch is not a signal.
The classic formula
One completed session in, seven levels out.
- Pivot: (high + low + close) ÷ 3, which is the same arithmetic as the typical price.
- R1 / S1: 2 × pivot − low and 2 × pivot − high.
- R2 / S2: pivot + range and pivot − range.
- R3 / S3: the range projected again from the previous session’s extremes.
Every one of those is fixed before the session opens and none of them moves in response to anything price does. That is unusual among the methods on this site, where almost everything is a smoothed function of recent data, and it is why pivots are described as a prior: they say where the previous session’s arithmetic projects, and they know nothing else.
Four formulas on one session
From a session with a high of 105.4, a low of 102.6 and a close of 104.8, a range of 2.8, the four common conventions produce these levels, computed at build time:
| Formula | Pivot | R1 | S1 | R2 | S2 |
|---|---|---|---|---|---|
| Classic | 104.27 | 105.93 | 103.13 | 107.07 | 101.47 |
| Woodie | 104.40 | 106.20 | 103.40 | 107.20 | 101.60 |
| Fibonacci | 104.27 | 105.34 | 103.20 | 106.00 | 102.54 |
| Camarilla | 104.80 | 105.06 | 104.54 | 105.31 | 104.29 |
The four pivots span 0.53 points and the four R1 levels span 1.14, on a session whose entire range was 2.8. That is the practical argument for picking one convention and staying with it: the disagreement between formulas is a material fraction of the move being projected, so a level quoted without its formula is not a level anyone else can check.
It is also the argument for the classic set specifically. Nothing about it is arithmetically superior. It is the default on almost every platform, which means it is the set other participants are watching, and the coordination is the only mechanism any of these levels has.
The session boundary decides the numbers
Every level on this page comes from "the previous session", and on a great many instruments that phrase is a decision rather than a fact. Futures, currencies and crypto trade around the clock; where the day ends is set by your data provider, and two feeds that draw the line an hour apart will report different highs, lows and closes for the same market.
The consequence is not subtle. A different close feeds a different pivot, which feeds every level derived from it, so two traders using the same formula on the same instrument can be watching levels a full point apart and both be right. On an instrument with a formal exchange session the problem does not arise, which is worth knowing, because it means pivot points are on their firmest ground exactly where they were invented, on floor-traded markets with an unambiguous close.
Reading them honestly
A projected level is a place, not a prediction. What it is genuinely useful for is deciding where to look before the session starts, and the reading comes from what happens when price gets there.
Three things are worth having in view at that moment. Whether volume expands into the level or dries up as price approaches it. Whether the level holds on a second test rather than the first. And whether the wider market is doing the same thing, because an instrument stalling at its own R1 on a day when everything stalled has told you about the market rather than about the level.
None of that is available from the arithmetic, and all of it is free. The levels' job is finished the moment they are drawn.
Where it misleads
| Situation | What goes wrong |
|---|---|
| Levels carried forward | They describe one completed session and expire when the next closes. Yesterday's pivots on a weekly chart are arithmetic about nothing. |
| Formulas mixed | Four conventions disagree by a material fraction of the range. A level quoted without its formula cannot be checked. |
| 24-hour instruments | The session boundary is a data-provider decision, so two feeds produce different levels from the same market. |
| Too many lines | Seven or more levels per formula means price will touch several. Hindsight finds the one that worked. |
| Touch read as a signal | Reaching a level says the prior arithmetic projected that far and nothing about what happens next. |
| Thin instruments | The mechanism is other participants watching the same defaults. Where few are, the levels are only arithmetic. |
What volume adds
Pivot levels are computed from three prices and contain no participation at all, which leaves the most important question about any touch unanswered: was anyone there? A level reached on the heaviest volume of the week, holding into the close, is a market that was genuinely contested at that price. The same level reached on the thinnest session of the month is a market that drifted there because nothing was in the way.
That is why the calculators on this site are described as producing priors rather than signals. The arithmetic says where the question will be asked; the volume, and the breadth of the market around it, are what answer it.
Frequently asked questions
How is a pivot point calculated?
In the classic formula the pivot is the previous session’s high, low and close averaged, the same arithmetic as the typical price. The first resistance is twice the pivot minus the low, the first support twice the pivot minus the high, and the second pair adds and subtracts the whole previous range from the pivot. Everything comes from three numbers and nothing adapts once the session begins.
Why are there four different formulas?
Because different authors weighted the inputs differently and all four survived. Woodie’s pivot double-weights the close, on the argument that the settlement price matters most. The Fibonacci variant keeps the classic pivot and spaces the levels at 38.2 and 61.8 per cent of the range instead of projecting the whole of it. Camarilla drops the pivot entirely and works from the close with a set of multipliers. They are not refinements of one another; they are four conventions.
Which formula should I use?
The classic one, unless you have a specific reason otherwise, and for the same reason RSI 14 is worth keeping: it is the default on almost every platform, so it is the set of levels a great many other participants are looking at. That coordination is the only real mechanism behind any of these levels, and it is lost the moment you pick a variant nobody else plots.
Do pivot points work?
They behave, which is a weaker and more defensible claim. Nothing in market structure privileges twice the pivot minus yesterday’s low. What is real is that the same three numbers are on every platform, computed identically, so orders cluster near the resulting levels, a coordination effect with a mundane explanation. It follows that the effect is strongest on heavily traded instruments where many participants watch the same defaults, and close to absent on thin ones.
How long is a pivot level valid?
For the next session only. They are derived from a completed session and are replaced when the next one closes, which makes carrying yesterday’s levels into next week the most common way the tool is misused. Weekly and monthly pivots exist and are computed the same way from a completed week or month; they are separate levels and should not be mixed on one chart without labels.
What is the session-boundary problem?
On anything trading around the clock — futures, currencies, crypto — "the previous session" is a decision made by your data provider rather than a fact about the market. Two feeds that close the day an hour apart produce different highs, lows and closes, and therefore different pivots from the same market. On an instrument with a formal exchange session this does not arise; on a 24-hour instrument it is the single largest source of disagreement between two people’s levels.
Is a touch of R1 a sell signal?
No. Price reaching a projected level says only that the previous session’s arithmetic reached that far. What the level is good for is telling you where a question will be asked, and the answer comes from what happens there: whether volume expands or dries up, whether the level holds on a retest, whether the wider market is doing the same thing. The level marks the place; something else has to supply the reading.
Why do so many levels get plotted?
Because each formula produces a pivot plus three levels either side, and Camarilla produces four. Plot two formulas and there are more than a dozen lines on the chart; across a session price will touch several of them, and hindsight will always find the one that "worked". Deciding in advance which single level matters, and writing it down, is the discipline that makes the tool worth anything.