Stochastic
Stochastic oscillator — %K / %D, overbought/oversold.
What it is
Stochastic oscillator — %K / %D, overbought/oversold.
The stochastic oscillator, developed by George Lane in the 1950s, measures momentum by asking where the current close sits within the high-to-low range of the look-back window. %K is the raw measure; %D is a simple moving average of %K used as the signal line. The canonical setting is 14, 3, 3 — 14-bar look-back, 3-bar SMA of %K to smooth it, then 3-bar SMA of %K to get %D.
Reads are similar to RSI: above 80 = overbought, below 20 = oversold. The classic signal is a %K / %D crossover near the extremes, plus divergence with price.
Unlike RSI, stochastic measures position-in-range rather than ratio-of-gains-to-losses, so the two oscillators behave differently in choppy markets — stochastic is faster and noisier.
%K = 100 · (C − L_N) / (H_N − L_N) %D = SMA(%K, 3)
Read the full Stochastic oscillator definition in the glossary →
Live chart
BTC/USDT on Binance with this indicator pre-loaded. Powered by TradingView.
Chart by TradingView. Built-in study shown for illustration; the Noon Barbari engine computes its own values.
Parameters
| Parameter | Default | Range |
|---|---|---|
| Period | 14 | 2 – 200 |
| Smooth K | 1 | 1 – 50 |
| Smooth D | 3 | 1 – 50 |
Output fields
The named values this indicator exposes to your entry and exit rules.
Related strategies
Ready-made, runnable templates that use this indicator. Open one to inspect or backtest it.
Learn more on the blog
Deep-dive articles explaining this indicator and how to trade it.
How to backtest a Stochastic Oscillator strategy on crypto
Turn a Stochastic %K/%D crossover idea into complete, testable rules, run it on real crypto history, and tell whether the edge is real or curve-fit.
The stochastic oscillator: a momentum tool, not a reversal alarm
The stochastic oscillator tells you where price closed within its recent range. Used as a standalone overbought/oversold signal, it loses money. Used as a filter, it works.
Backtest this indicator
Drop this indicator into a rule-set, run it over years of BTC/USDT data, and see whether the edge is real or just curve-fit — no credit card required.