Pivot Point Calculator

Pivot points are a map of a normal day

Every pivot calculator on the internet gives you seven prices. Almost none of them tell you the one thing that decides how to use those prices: how often price actually gets there. So we measured it on five years of our own hourly data, one New York trading day at a time.

Classic daily pivots from the prior day's high, low and close. The day runs 18:00–17:00 New York, the same boundary most forex and CFD platforms use. 1,291 trading days from 20 September 2021 to 18 September 2026; Bitcoin from July 2022, weekends included.

Read the table the way a risk manager would. Price trades back through the pivot on about three days in four. It reaches R1 or S1 on roughly 45% of days each, and one of the two on about 76%. R2 and S2 drop to one day in five or six. R3 and S3 are one-day-in-twelve-to-twenty events.

That makes the ladder a description of range, not a set of turning points. R1 and S1 mark the edges of an ordinary day. R2 and S2 mark an unusual one. Once you see it that way, two common habits look wrong: setting a target at R2 as though it were the default outcome (it misses on four days in five), and fading R1 as though it were a wall (on a trending day it is simply the first stop).

The one reading that has something behind it

The number that surprised us is the open. When NAS100 opened the day above its pivot, it also closed above the pivot 66.4% of the time; gold did the same 66.2% of the time. Opening below and closing below was weaker, at 57.3% and 58%. EURUSD leaned the other way (57.2% above, 62.1% below).

That is a lean, not a signal. A market that has drifted up over five years will show an upside bias in almost any measure you pick, and a 66% day-level tendency says nothing about where you get in or where your stop goes. What it does tell you is which side of the pivot deserves your attention first, and that is a rule you can write down and test before you trade on it.

Classic, Fibonacci, Camarilla, Woodie: what actually changes

All four start from the same three numbers. They differ in two ways: where the centre sits and how far apart the levels are.

Classic is the floor-trader original: P = (H + L + C) ÷ 3, then R1 = 2P − L and S1 = 2P − H. It is the default on most platforms, and it is the one measured above.

Fibonacci keeps the same pivot and places levels at 38.2%, 61.8% and 100% of the prior range either side. R3 and S3 are exactly one full range from P, which makes them easy to reason about.

Camarilla ignores P for the levels and hangs them off the close: close ± range × 1.1 ÷ 12, ÷ 6, ÷ 4, ÷ 2. The result is tight: H3 and L3 sit about a quarter of yesterday's range from the close, and the method was built to fade that range and treat a break of H4 or L4 as a trend day.

Woodie counts the close twice, P = (H + L + 2C) ÷ 4, so a day that finished near its high carries a higher pivot into the next one.

Pick the one your chart draws. The worst version of pivot trading is using one method in your analysis and another in your platform, because you end up testing levels you never see live.

Pivots and the levels the market already remembers

Look at what goes into the formula: yesterday's high, low and close. Two of those are levels in their own right. The previous day's high and low are where the obvious stops sit, which is why they get run so often. The liquidity sweep reference covers the mechanic, and the key levels backtester draws PDH and PDL automatically on every market.

In practice the pivot and the prior day's extremes work as one map. P and the daily open tell you whether price is in premium or discount for the day; PDH and PDL tell you where the liquidity is; R1 and S1 tell you what a normal range looks like from here. A level where two of those line up is worth more attention than any one of them alone.

Daily, weekly or monthly

The formula does not care about the timeframe. Daily pivots come from yesterday's candle and suit intraday trading; weekly pivots come from last week and suit swing trading; monthly pivots are for people who hold for weeks. Enter the finished week or month in the calculator and it works the same way.

The mistake is using a pivot from a period that has not finished yet. Today's pivot uses yesterday's numbers and stays fixed all day, which is exactly why it can be tested honestly: you knew it at the open.

Sizing a trade off a pivot

A pivot gives you a place. It does not give you a trade until the stop and size are set.

Size from the stop. A fade at R1 with the stop beyond R2 is a very different position from one with the stop a few points past R1. The position size calculator turns the distance into lots.

Check the ratio against the hit rate. If R2 is reached on under a fifth of days, a trade that only pays at R2 needs a reward-to-risk that justifies it. The risk/reward calculator gives the break-even win rate for any ratio.

Count enough days. A pivot rule tested over three weeks is fifteen observations. The sample size calculator shows how wide the error on that is.

Test the levels on your own market

The table above describes whole days. Your strategy lives inside them: a particular session, a particular reaction, a particular stop. The only way to know whether a pivot rule has an edge on the market you trade is to replay it with the levels on the chart and no view of what comes next, then count.

CRTLAB replays real price one candle at a time and draws the previous day's high and low and the daily and weekly opens for you. Put the pivot lines from this calculator on the chart, step through the session and log what happens at each one. The rules-writing guide covers how to phrase the rule so the test means something.

Frequently asked questions