Stochastic RSI

An oscillator that applies the stochastic formula to RSI values instead of price, producing a faster but noisier 0-100 (or 0-1) momentum reading.

What it is

Stochastic RSI (StochRSI) is an indicator of an indicator. Developed by Tushar Chande and Stanley Kroll and published in 1994, it takes the familiar RSI calculation and then runs the stochastic oscillator formula on the RSI values themselves, rather than on price. The result measures where the current RSI sits relative to its own recent range: a reading near the top means RSI is at the high end of its recent range, and a reading near the bottom means RSI is at the low end. The stated motivation was that plain RSI can spend long stretches between 30 and 70 without reaching its classic extremes; StochRSI re-scales RSI so it reaches its own extremes far more often. That makes it more reactive, but 'faster' is not the same as 'better' — the double transformation also amplifies noise, and StochRSI frequently pins at 0 or 100 for several bars at a time. Like every oscillator, it is descriptive: it summarizes recent momentum behaviour and predicts nothing. This is educational material only, not financial advice, and all trading carries risk of loss.

How it works

The calculation is a two-step transformation. Step one: compute a standard RSI over a chosen lookback (commonly 14 bars), which produces a 0-100 momentum series based on Wilder-smoothed average gains and losses. Step two: apply the stochastic formula to that RSI series over a second lookback (also commonly 14): StochRSI = (RSI − lowest RSI over N bars) / (highest RSI over N bars − lowest RSI over N bars). The raw output runs from 0 to 1, though many platforms scale it to 0-100. Most implementations then smooth the raw value with a short moving average to produce a %K line, and smooth %K again to produce a %D signal line (3 and 3 are typical). The double transformation explains the indicator's signature behaviour: because the stochastic step measures RSI against its own recent range, any time RSI makes a new N-bar high or low — which happens often — StochRSI hits exactly 1 or 0 and stays pinned there while the streak continues. That pinning is a mechanical consequence of the formula, not an unusually strong signal. It also means StochRSI readings depend on two lookback windows plus two smoothing lengths, so small settings changes can noticeably reshape the line.

How traders read it

• Range position: StochRSI shows where the current RSI sits within its own recent range — a high reading means RSI is near its N-bar high, a low reading means RSI is near its N-bar low. It describes RSI's position, not price's future.
• Common bands: 0.8/0.2 (or 80/20) are the usual reference levels rather than RSI's 70/30, because StochRSI reaches its extremes far more often by construction.
• Pinned readings: values sitting flat at 0 or 100 simply mean RSI keeps making new lows or highs within the lookback window — common in trends and not, by itself, evidence of an imminent reversal.
• %K/%D crosses: some traders watch the smoothed %K crossing its %D line as a momentum-shift observation; on this indicator such crosses are frequent and many occur in noise.
• Versus plain RSI: StochRSI is more sensitive and cycles between extremes more often, so it produces more readings — and more false ones. Faster is a trade-off, not an upgrade.
• Any reading is best combined with price structure, the higher-timeframe trend, and a defined risk plan rather than acted on alone.

Common settings

Typical defaults are RSI length 14, stochastic length 14, %K smoothing 3, and %D smoothing 3, applied to the close. Some traders shorten both lookbacks for an even faster line or lengthen them to calm the pinning behaviour. Because there are four interacting parameters, StochRSI is more settings-sensitive than plain RSI — the same chart can look very different under modest parameter changes, so treat any preferred settings as a choice to understand, not a discovered truth.

Strengths

• Reaches its extremes far more often than plain RSI, giving a reading even in markets where RSI stays mid-range for long stretches.
• Bounded 0-1 (or 0-100) scale makes levels easy to reference consistently across instruments and timeframes.
• The %K/%D structure is familiar to anyone who has used a standard stochastic oscillator.
• The construction is fully transparent: two well-documented formulas composed together, easy to verify against a platform's built-in.
• Useful for illustrating, in an educational sense, how transforming an indicator changes its statistical behaviour.

Pitfalls to watch

• Double transformation amplifies noise: StochRSI is two derivations away from price, so it reacts to changes in a derivative of price, producing many fast, low-information swings.
• Pinning at 0/100 is mechanical: whenever RSI makes a new lookback high or low the formula outputs exactly 1 or 0, so extreme readings are common and frequently persist through strong trends.
• Overbought/oversold readings are not reversal signals: in a trending market StochRSI can sit at an extreme for many bars while price keeps going.
• Settings sensitivity: four parameters (two lookbacks, two smoothings) mean small changes reshape the line and every crossover, inviting curve-fitting.
• Still lagging: despite being 'faster' than RSI, it is built from smoothed averages of past prices and confirms moves after they have begun.
• No predictive power: like RSI itself, StochRSI describes recent momentum only — it does not forecast price and does not reduce the risk of loss.

Pine v6 example

//@version=6
indicator("Stochastic RSI Example", overlay = false)

rsiLen   = input.int(14, "RSI Length", minval = 1)
stochLen = input.int(14, "Stochastic Length", minval = 1)
kSmooth  = input.int(3, "%K Smoothing", minval = 1)
dSmooth  = input.int(3, "%D Smoothing", minval = 1)
src      = input.source(close, "Source")

rsiValue = ta.rsi(src, rsiLen)
rawStoch = ta.stoch(rsiValue, rsiValue, rsiValue, stochLen)
k = ta.sma(rawStoch, kSmooth)
d = ta.sma(k, dSmooth)

plot(k, "%K", color = color.blue)
plot(d, "%D", color = color.orange)
hline(80, "Upper", color = color.gray)
hline(20, "Lower", color = color.gray)

Pro tip: When comparing StochRSI to plain RSI, put both in separate panes on the same chart and watch how often each reaches its extremes: StochRSI will hit 0 or 100 many times for every RSI visit to 30/70, purely because of the re-scaling step. That exercise makes the trade-off concrete — more readings means more noise, not more information. Treat pinned extremes as a description of a streak in RSI, not a countdown to reversal, and pair any reading with price structure and defined risk. Educational context only; no indicator is predictive, and trading involves risk of loss.

Educational only — not financial advice, not a recommendation to trade. No indicator is predictive; trading involves substantial risk of loss.