Pivot points — the day's levels, calculated last night
The answer
A pivot point is a single price calculated from yesterday's high, low and close. Traders draw it on today's chart, along with support and resistance levels derived from the same 3 numbers, and use them as reference points for an intraday session.
They are not predictions. They are yesterday's range, rearranged into today's grid.
Why this costs you money
Pivot levels look official. They appear on the chart before the market opens, they have names like R1 and S2, and they are identical on every platform. That combination produces a specific and expensive error: treating a calculated number as a place where something must happen.
Here is the loss. A stock opens above the central pivot. A trader buys, with a stop just below the pivot, targeting R1. Price drifts sideways, dips 0.4% below the pivot, and stops them out. It then rises to R1 without them. They re-enter near R1 because "the level broke", and price falls back to the pivot.
Two losses in one session, both taken at levels chosen by arithmetic on yesterday's candle.
The second loss is larger and less obvious. On a day when the market gaps — an overnight event, a results announcement, a policy decision — yesterday's range is irrelevant. The pivot grid is still drawn, still looks authoritative, and is completely disconnected from the prices that are now trading. Traders place stops at S1 on a day when the market has already repriced 3% below it.
A level calculated from a period that no longer describes the market is worse than no level, because it produces confident action.
How it works
The base calculation uses only 3 numbers from the previous session.
Pivot (P) = (High + Low + Close) ÷ 3
That is it. The central pivot is the average of yesterday's high, low and close. It is sometimes called the typical price of the previous session.
From P, the classic floor-trader formulas produce 3 resistance levels and 3 support levels:
| Level | Formula |
|---|---|
| R1 | (2 × P) − Low |
| R2 | P + (High − Low) |
| R3 | High + 2 × (P − Low) |
| S1 | (2 × P) − High |
| S2 | P − (High − Low) |
| S3 | Low − 2 × (High − P) |
Look at what those formulas actually contain. Every level is P plus or minus some multiple of yesterday's range. R2 and S2 are exactly 1 full range above and below the pivot. So the entire grid is a single statement: here is yesterday's centre, and here is yesterday's range measured out from it.
The Central Pivot Range, or CPR, adds 2 more lines around the pivot:
- BC (bottom central) = (High + Low) ÷ 2
- TC (top central) = (2 × P) − BC
The distance between TC and BC is the CPR width. A narrow CPR means yesterday's close sat near the middle of yesterday's range. A wide CPR means the close was far from the middle. Indian intraday traders read a narrow CPR as a sign that a trending day may follow, and a wide CPR as a sign of a range-bound day.
The origin of the classic formulas is the open-outcry futures pits, where a trader could compute the day's levels overnight and write them on a card. The CPR variant was popularised in English by Frank Ochoa in Secrets of a Pivot Boss, and it reached Indian retail trading through that route.
What it tells you, and what it does not
It tells you where yesterday's centre was and how wide yesterday was. Those are useful facts. A trader who knows that today is trading entirely above yesterday's whole range knows something real.
It gives you a fixed, pre-known grid. That has a genuine practical value that has nothing to do with prediction: you can plan a trade before the open, with a defined entry, stop and target, instead of improvising inside the session. Most intraday losses come from improvising.
It does not tell you the market will stop at those levels. There is no mechanism. Nothing about "1 full previous range above yesterday's midpoint" creates supply.
It does not adapt to today. If yesterday was unusually quiet and today has an event, every level is too tight. If yesterday was unusually violent and today is quiet, every level is too far away and will never be reached.
It does not survive a gap. When today's open is outside yesterday's range, the grid was built from a market state that has been replaced.
And it does not work as a swing or investing tool. Weekly and monthly pivots exist, but the further out you go, the more the "previous period" is a poor description of the current one, and the fewer traders are watching the same numbers.
The decision rule
Use pivots as a map of yesterday, not as a set of instructions.
Step 1, before the open. Compare today's expected open to yesterday's range. If the open is inside the range, the grid is usable. If the open is outside the range, discard the grid for at least the first hour and use the opening range instead.
Step 2, the bias. Price holding above the central pivot after the first 30 minutes is an up-bias for the session. Below is a down-bias. This is a description, not a signal.
Step 3, the use. Use levels as targets and stop references, not as entry triggers. Entering because price touched a calculated number is trading arithmetic. Exiting at one is managing a position, which is a different and defensible thing.
Never use pivots on a scheduled event day — a budget, a central bank decision, a results release. Yesterday does not describe today.
Try this now
This takes about 5 minutes and it settles the question with your own instrument.
- Pick 1 index or liquid stock you actually trade or watch intraday. Open a daily chart and write down the high, low and close for each of the last 20 trading days. Most apps let you hover over each candle to read them.
- For each day, compute the next day's pivot: (High + Low + Close) ÷ 3.
- Now open the next day. Note 2 things. Did price open above or below that pivot? Did price close above or below it?
- Count how many of the 20 days opened and closed on the same side of the pivot.
- Separately, count how many of the 20 days reached R1 or S1 at any point.
What you should see. Two numbers, and they usually land in a predictable place.
The same-side count in step 4 is typically somewhere around 12 to 15 out of 20 on a liquid index. That sounds impressive until you realise most days close near where they open, so a coin that always predicted "same side" would do well anyway. Write down what a simple rule of "today closes on the same side of yesterday's close as it opened" would have scored, and compare.
The R1 or S1 count in step 5 is usually high, often 15 or more out of 20. That is because R1 and S1 sit inside the previous day's range projected outward, and most days move at least that far. A level that is reached almost every day is not a target. It is a description of a normal amount of movement.
Now find the 2 or 3 days in your 20 with the largest moves. Check where price finished relative to R3 or S3. On the big days, the whole grid was left behind before lunch. The days the grid describes well are the days there was nothing to make.
Three real cases
1. NIFTY 50, 3 and 4 June 2024 — two consecutive days that destroyed the grid On 3 June 2024, after exit polls were published, the Indian index rose sharply, by roughly 3%. On 4 June 2024, when actual election results came in differently from the exit polls, it fell heavily intraday, with the index down around 6% at the close. Pivot levels for 4 June were computed from 3 June's range. They were irrelevant within the first minutes. Every support level was passed. This is the clearest available example of the general rule: pivots describe a market state, and events replace market states overnight.
2. Dow Jones Industrial Average and S&P 500, March 2020 — the month levels stopped meaning anything US indices moved by more than 5% in a single session repeatedly through March 2020, in both directions, including some of the largest single-day percentage moves in decades. Each day's pivot grid was built from a previous day that was itself extreme. R3 and S3, which are reached on perhaps a handful of days in an ordinary year, were passed regularly. The grid was not wrong. It was measuring yesterday, and yesterday had stopped being a guide.
3. The trading floor, before electronic markets — why the formula is this shape The classic pivot formulas exist in the form they do because a pit trader needed levels that could be computed by hand overnight from 3 numbers printed in a newspaper, and carried on a card. That constraint explains everything about the design: only 3 inputs, only addition and multiplication, no averaging over multiple days. It is a good design for its constraint. It was never a claim about market behaviour, and modern platforms that draw it automatically have removed the context that made it modest.
The question that resolves it
A novice sees price approach R1 and asks: will it stop here?
An expert sees the same approach and asks: how much of yesterday's range has today already covered?
The second question tells you what R1 actually represents. If today has already travelled 1.5 times yesterday's range, the grid is stretched and the levels above are being generated by a formula that has run out of relevance. If today has covered half of yesterday's range, the grid is still describing something.
That single comparison — today's range so far against yesterday's full range — tells you more than the level names ever will.
What would make this wrong
If pivot levels had real reflecting power, price would reverse at them more often than at randomly chosen levels the same distance from the open, tested across many instruments and many sessions.
That is a fair test and anyone can run it. Take your 20 days. Instead of R1, mark a level the same distance above the open but chosen at random. Count reversals at each. If the pivot does not beat the random level, the reversals you have been seeing were the ordinary behaviour of price at any level a certain distance away.
The honest limits are 3.
First, the self-fulfilling argument is real and this article does not dismiss it. If enough traders place orders at the same computed number, activity clusters there. That is a genuine effect. It is also fragile: it works when the market is quiet enough for retail order flow to matter, and disappears the moment a large participant or an event enters.
Second, the discipline value is real and separate from the prediction value. A trader with pre-defined levels makes fewer improvised decisions. If pivots give you that structure, they are earning their place even if the levels themselves have no power. Be honest with yourself about which of the 2 benefits you are actually getting.
Third, this article treats standard and CPR pivots together. Fibonacci pivots, Camarilla pivots and Woodie pivots use different multipliers on the same 3 inputs. The fact that so many variants exist, all claiming reliability, is itself information about how much reliability there is.
In India
Pivot points and the Central Pivot Range are extremely widely used by Indian intraday traders. Broker platforms and free charting sites display them by default, most commonly with identical formulas.
Three Indian features matter.
Scale of shared use. The CPR in particular has become close to a standard tool in Indian retail intraday trading, taught in a large number of courses and displayed on the most-used platforms. When a very large number of traders draw the same 3 lines from the same formula on the same instrument, orders cluster at those lines. This gives Indian pivot levels a measure of self-fulfilling weight that is genuinely stronger than in most markets. It is not magic. It is coordination.
That weight has a limit. It is strongest on NIFTY, BANK NIFTY and the most liquid single stocks, on ordinary days, in the middle of the session. It is weakest at the open, on event days, and in stocks where a single institutional order dominates the volume.
Price bands. A stock that closed at a circuit limit yesterday produces a truncated range, so today's pivot grid is compressed. All 6 levels sit closer to the pivot than the stock's real behaviour warrants.
Weekly index option expiries. On expiry days, index price behaviour around round strike prices reflects options positioning. Those strikes are not pivot levels, and when the 2 sets of levels are close together, traders routinely attribute an options-driven move to a pivot.
In the United States
The classic floor-trader pivot formulas originate in US futures pits and remain in use, particularly on index futures and liquid large-cap equities.
Three US features matter.
A long pre-market session. US equities trade extensively before the 9:30 am regular open. By the time the official session begins, price may already be far from yesterday's close, and the pivot grid built on regular-session data does not include any of that movement. Many US traders therefore compute pivots from the futures session instead, which produces different numbers. Decide which you use and be consistent.
Earnings after the close. A US stock that reports at 4:05 pm can open the next day 10% away. The pivot grid for that day is computed from a session that occurred before the information existed.
Less uniformity of practice. US traders use a wider mix of methods — opening range, VWAP, previous day's high and low, market profile levels. Because practice is more fragmented, the clustering effect that gives Indian CPR levels their weight is weaker. More US attention goes to VWAP, which updates continuously through the session rather than being fixed overnight.
Where they differ, and what that tells you
The honest difference is not in the formula. It is identical in both markets. The difference is how many people are looking at the same number.
In India, a very large share of intraday retail traders draw the same CPR from the same default settings. That concentration creates a real, observable tendency for activity to cluster at those lines on ordinary days.
In the United States, intraday attention is spread across VWAP, opening range, prior day levels and several pivot variants. No single grid captures the same share of attention.
What that tells you is 2 things, and the second is the important one.
First, a level's power can come from agreement rather than from mathematics. That is worth knowing generally, because it means the useful question about any level is "how many people are watching this?" rather than "is the formula sound?"
Second, and less comfortable: a level that works because everybody watches it is also a level everybody's stop sits under. Concentrated retail orders at a known price are visible to participants who can act on that information. A crowded support level is a pool of predictable stop orders. In India this shows up as sharp moves through a widely-watched level followed by an immediate recovery. If you place your stop exactly at the level everybody uses, you are placing it where it is easiest to find.
The practical correction is small and it costs nothing: put your stop a distance beyond the level, sized by that instrument's ATR, rather than exactly on it. Article 16 shows how to choose that distance.
Carry this
- Every pivot level is yesterday's centre, plus or minus a multiple of yesterday's range.
- On a gap day, discard the grid. It describes a market that has been replaced.
- Where pivots work, they often work because many people watch them. Put your stop beyond the level, not on it.