What Is a Pivot Point? The Level You Don't Have to Draw

Nearly every level a trader puts on a chart got there by judgement. Somebody looked at the price action and decided this peak mattered and that one didn't. A pivot point skips all of it. You do a division sum on yesterday's numbers, and today's levels exist before the market opens.

That's the whole method. It takes about fifteen seconds and there's nothing left over.

So the obvious question turns up early, and this piece spends most of its time on it. If the number falls out of arithmetic that knows nothing about the market, why would the market care? The honest answer is more interesting than either yes or no.

Every other level on your chart is an opinion

Put two traders in front of the same chart and ask each of them to mark support. You'll get two lines. Both will be defensible, both will be drawn off real lows, and they'll be forty cents apart.

That isn't incompetence. A level is a judgement about which of the many places price turned deserves to be remembered, and sensible people weigh that differently. One trader anchors on the deepest low. The other picks the low that produced the strongest bounce. Neither of them is doing it wrong, and neither can prove the other isn't.

Our guide to swing highs and swing lows is the drawing half of this job — which peak counts, why support is a zone rather than a line, what happens when a broken support turns into resistance. Break of structure picks up where that leaves off. This article won't re-teach any of it, because it's about the levels you don't draw.

A pivot point contains no judgement at all. One input, one formula. Hand the same session to a thousand traders and you get one number a thousand times. Whatever else that is, it's rare.

There's one other calculated line in common use, and the difference is worth naming. A moving average is computed too, but it recalculates on every bar and slides along underneath price. A pivot level is fixed for the whole session. You work it out once, before the open, and then it sits there.

Where the number actually comes from

Take the session that just finished. It ran up to $104.40, down to $100.20, and closed at $103.50. Those three numbers are everything the formula uses. The open doesn't come into it and neither does volume.

Add them: $308.10. Divide by three: $102.70. That's the pivot, written P, and it's today's centre line.

The rest of the grid falls out of the same three numbers. First resistance is twice the pivot minus yesterday's low — $205.40 minus $100.20, so $105.20. First support is twice the pivot minus yesterday's high: $205.40 minus $104.40, so $101.00. Second resistance and second support add and subtract yesterday's range of $4.20, landing at $106.90 and $98.50. The third pair reaches further out again.

The grid is yesterday's range with a new middle

Here's the part almost nobody writes down, and it changes what you're actually looking at.

Subtract S1 from R1. On our session that's $105.20 minus $101.00, which is $4.20. Now subtract yesterday's low from yesterday's high: $104.40 minus $100.20. Also $4.20, and not by luck.

One line of algebra shows why. R1 is 2P minus the low. S1 is 2P minus the high. Subtract one from the other and the 2P cancels, leaving the high minus the low. The pivot disappears out of the sum entirely. Whatever yesterday's close did, the distance from S1 up to R1 is yesterday's range exactly.

So the first support-to-resistance band isn't a forecast of anything. It's yesterday's range, picked up and re-centred on a new middle. R2 to S2 is the same trick at double the width.

Now do something with that. The day on the chart above ranged $1.35 — a normal, quiet session. It never went near R1 or S1. It couldn't have: the rails are $4.20 apart and the day only had $1.35 in it. No amount of good behaviour by price was going to reach both.

The other case is just as mechanical. A day that does tag both rails has, by definition, covered more ground than yesterday did. The day in the first chart ranged $4.73 against yesterday's $4.20, which is what it took to reach S1 in the morning and R1 by the close.

None of that is the levels working. It's a ruler laid across a day, and the day was a certain size. Hold on to it for the section on evidence, because a fair share of every "price respected S1" story is really a day that happened to be about as big as the one before it.

The close tilts the whole thing

The second thing worth knowing is that the grid isn't symmetric, and the lopsidedness is doing more work than most people notice.

R1 sits above the pivot by exactly the distance from the pivot down to yesterday's low. S1 sits below the pivot by exactly the distance from yesterday's high down to the pivot. So where yesterday closed decides how much room today gets in each direction.

Our session closed at $103.50, near the top of its range. That drags the pivot up to $102.70, which leaves $2.50 of room up to R1 and only $1.70 down to S1. A strong close buys you a distant ceiling and a floor right underfoot.

The middle row is the one to remember. When yesterday closes dead centre, R1 and S1 land exactly on yesterday's high and low — not near them, on them. That's the grid at rest, at $104.40 and $100.20. Every other close just shoves it off centre.

And the shove is symmetrical. Close near the low, as in the top row, and the pivot drops to $101.80: now the ceiling is $1.60 away and the floor $2.60 below. Same range, opposite tilt.

So a trader reading a pivot grid is reading yesterday's range with the middle pulled toward whoever finished on top. That's a real piece of information about yesterday. It just isn't a statement about today.

Build a session below and watch both facts hold. The band between the rails stays exactly as tall as the range you make, and the middle slides as you drag the close.

What the grid claims

Everything so far is arithmetic, and all of it is provable. Here's what traders do with the result, set out as what the method asserts rather than as advice.

• The pivot is the dividing line. Above it is treated as the stronger half of the day, below it the weaker half, and where price sits relative to P is the first read of the session.
• R1 and S1 are where the first reaction is expected. Price arriving there is supposed to slow, stall or turn.
• A clean break through R1 or S1 is read as a trend day — the market has decided to go somewhere. A rejection at the rails is read as a range day, with the levels holding it in.
• R2, S2 and beyond are treated as the day's outer edges, reached only when something genuinely unusual is going on.

That's the standard reading and you'll find it on every platform's help page. It's also entirely a set of claims. Nothing in the last two sections supports any of them. A division sum can produce a level, but it can't produce a reason for anybody to trade there.

So does it work?

Start with what's honest. No study in a major finance journal tests pivot points as a named strategy. Papers that mention them do exist in smaller journals, and the ones we could read don't test whether trading them makes money — they compare how stable the formulas are, which is a different question and a duller one.

So what follows is reasoning, not a result. Here's the reasoning.

There is nothing in (high + low + close) ÷ 3 that a market maker respects. The arithmetic knows nothing about supply, order books, earnings or who's short. If pivot points do anything at all, it isn't because $102.70 is a special price. It's because a lot of people worked out $102.70 before the open and are watching it.

The nearest real evidence for that idea comes from an unusual study. In 2000 Carol Osler published a paper in the Federal Reserve Bank of New York's Economic Policy Review that tested the support and resistance levels six firms — banks and news services — published to their clients every day. She checked them against one-minute quotes for the mark, the yen and the pound against the dollar, from January 1996 to March 1998, New York hours. The levels did mark where intraday trends stopped, more often than randomly chosen levels did, for all six firms. The effect lasted about five business days.

Read that carefully, because it's narrower than it looks. Those weren't pivot points, and the word doesn't appear in the paper. The firms' analysts chose their levels by eye — recent highs and lows, the 50 percent rule, Fibonacci retracements, round numbers — so what got tested was human judgement published in advance, not a formula. It's spot foreign exchange rather than stocks. And Osler doesn't say why it happened: she lists self-fulfilling prophecy as one candidate explanation among several, in a single sentence, and leaves it untested. She leaves profitability to future research too, which is a polite way of saying nobody has shown you could make money from it.

What survives is still worth having. Levels everybody could see in advance marked where trends paused, in one market, over two years. Pivot points are levels everybody can see in advance. That's an analogy, and it's the strongest one available — but an analogy is what it is, and anyone selling it to you as proof is selling you a paper that doesn't exist.

Then there's the deflator from earlier. R1 to S1 spans yesterday's range exactly. So price wandering up to R1 and stopping often means today covered roughly the ground yesterday covered, which is what most days do. On our session that's $4.20. Some proportion of every "it reacted beautifully at R1" story is a day being an ordinary size.

One last framing, from an unlikely direction. FINRA's advertising rule tells member firms their communications may not predict or project performance, and then carves out an exception for a hypothetical illustration of mathematical principles, provided it doesn't forecast an investment's performance. Neither FINRA nor the SEC has ever mentioned pivot points, and that rule governs firms talking to the public rather than you at your desk. Even so, it draws exactly the right line for this tool. You can show the arithmetic. The arithmetic isn't a forecast.

Five formulas, one claim

Here's the fact that ought to settle the argument, and it takes about a minute to check. Run the five common pivot methods on the one session — high $104.40, low $100.20, close $103.50 — and ask each of them for the first resistance above the pivot.

Five formulas, five answers, one session. If the arithmetic itself were doing the work, four of these would have to be wrong. They can't all be finding the level that matters, because they don't agree on where it is.

That closes the loop. If pivot points work at all, it's because people watch them — and that hands the advantage to the most-watched version rather than the cleverest one. The plain (high + low + close) ÷ 3 is the default on nearly every charting platform, so it has the most eyes on it. That isn't a mathematical property. It's a headcount.

Three notes for anyone checking these against their own platform, because the variants disagree in ways that look like errors and aren't. The Woodie pivot above doubles the previous session's close; several platforms double today's open instead, which produces a different number. Both forms are documented and neither is wrong. Our Camarilla computes four levels each side, while some platforms document five. And on the R3 and S3 in the earlier table: TradingView lists that formula as Traditional and keeps the name Classic for a different R3, so if your platform's third rail doesn't match ours, check which one it's drawing.

Where it falls apart

Four situations where the grid stops being useful. Better to know them now than to find them with money on.

Gaps. The grid comes from a session that's already finished. Let today open above R1 at $105.20 and price starts with the whole ceiling behind it, before anyone has traded. The levels aren't wrong. They're spent.

News. An earnings release, a rate decision, a takeover — price goes where the news sends it, and $102.70 has no opinion on any of it. The whole grid assumes today looks something like yesterday. Some days flatly don't.

Thin trade. The only argument for these levels is that a crowd is watching them. In a market with no crowd, nobody's watching, and the grid is seven numbers on a screen and nothing else. That bites hardest in small stocks and in quiet hours.

Timeframe. Daily pivots come from yesterday's session and belong on a chart of today. Plenty of traders read them on a five-minute chart, which is fine, but the levels don't get more precise because you zoomed in. Weekly and monthly pivots work identically off the previous week's or month's high, low and close, and they move correspondingly less often.