Concept

Overbought/oversold

Overbought/oversold, also known as 80/20 & 70/30 conventions, is a Momentum & Oscillators concept. The Library holds 13 implementations, each one a working definition you can pull into Quant.

Top Overbought/oversold indicators

13 total

What is Overbought/oversold?

Overbought and oversold are zone labels on bounded oscillators. When a momentum gauge such as RSI rises above its upper threshold (70 in Wilder's convention) the market is called overbought; below the lower threshold (30) it is oversold. The Stochastic Oscillator uses 80/20, Williams %R uses -20/-80. The label means price is stretched relative to its own recent behavior, not that the instrument is expensive or cheap in any absolute sense.

What the label implies depends on regime. Inside a trading range, extreme readings often precede rotation back toward the middle. In a strong trend the same readings mark strength: a bounded oscillator can hold above 70 for weeks while price keeps climbing, a state traders call pinned or embedded. That is why the common trigger is not entry into the zone but the cross back out of it, and why some systems read the first push into overbought as initiative rather than exhaustion.

The convention matters because it is the shared grammar of nearly every oscillator. Thresholds decide when a divergence or failure swing counts, when a screener flags a symbol, and when a mean-reversion system arms. Misreading the regime, fading every overbought print inside a persistent trend, is the classic failure mode.

How to read overbought/oversold zones

The zones are conventions layered on a bounded scale, so reading them is mostly about context.

  1. 1Pick a bounded oscillator and note its conventional thresholds: 70/30 for RSI, 80/20 for stochastics, -20/-80 for Williams %R. Unbounded gauges like CCI use ±100 by convention instead.
  2. 2Classify the regime before reading the zone, using market structure or a higher-timeframe trend filter. In a range an extreme argues for reversion; in a trend it often argues for continuation.
  3. 3Treat entry into the zone as a condition, not a signal. The usual trigger is the cross back through the threshold, or corroborating evidence at the extreme such as divergence or a failure swing.
  4. 4Adjust thresholds to the instrument and regime. Persistent bull phases may never print RSI 30, which is why asymmetric schemes such as RSI range rules shift the operating bands instead of keeping 70/30 fixed.

How traders use it

  • As mean-reversion arming inside ranges: wait for the oscillator to re-enter the normal zone and fade back toward the middle, often paired with an S/R zone or band tag at the range boundary.
  • As a pullback timer inside trends: an oversold print in an uptrend flags a dip to consider buying rather than a top to sell, which keeps oscillator signals aligned with the prevailing direction.
  • As a strength read: some momentum systems deliberately buy the push into overbought, treating it as a momentum thrust that marks initiative, and only exit once the reading drops back out.
  • As screening grammar: scanning a universe for symbols at extremes builds watchlists and breadth-style context, since how much of a market is stretched at once is itself information.

Overbought/oversold vs neighboring reads

Centerline Regime: Both are oscillator grammar, but the centerline read asks which side of the midline (50 on RSI, zero on MACD) the gauge lives on to define bias. Overbought/oversold asks whether it reached an extreme: midline for regime, extremes for stretch.

RSI Range Rules: Range rules replace fixed 70/30 thresholds with regime-shifted operating bands, such as RSI respecting a higher floor throughout bull phases. Overbought/oversold is the static convention those rules deliberately bend to fit the regime.

Momentum Thrust: A thrust approach buys precisely when the oscillator crosses into overbought, reading the extreme as initiative rather than stretch. Same zones, inverted interpretation, which is why regime classification has to come first.

Bollinger Band Tag Reversion: A band tag measures stretch in price terms, distance from a moving average in standard deviations, while overbought/oversold measures stretch on a bounded oscillator scale. The two often fire together but can disagree after volatility shifts.

More Overbought/oversold implementations

Related concepts · Oscillator grammar (cross-cutting)

Concept family

Momentum & Oscillators

91 concepts mapped · 72 in the Library

Overbought/oversold FAQ

Is overbought a sell signal?

No. Overbought only means the oscillator is in its upper zone, which happens in every strong uptrend. In ranges it often precedes reversion; in trends it frequently accompanies continuation. Many systems wait for the cross back below the threshold, a divergence, or a structural trigger before treating the condition as actionable.

What are the standard overbought and oversold levels?

Conventions, not laws: 70/30 on RSI (Wilder's original), 80/20 on the stochastic oscillator and Stochastic RSI, -20/-80 on Williams %R, and roughly ±100 on CCI. They mark the outer portion of each scale, and traders routinely tighten, widen, or shift them by instrument, timeframe, and regime.

Why does RSI stay overbought in a strong trend?

Because bounded oscillators compress persistent one-way movement into the top of the scale. As long as average gains keep dominating average losses, RSI holds above 70; the reading reflects continued strength, not an imminent reversal. Traders call this an embedded reading, and it is why regime context comes before the label.

Should I change the default 70/30 or 80/20 thresholds?

Sometimes. Quiet instruments may rarely reach the standard zones while volatile ones live in them, and bull regimes lift RSI's effective floor. Approaches in the spirit of Constance Brown's range rules use asymmetric zones per regime. Whatever you pick, fix it before testing rather than fitting thresholds to whatever would have worked.

Do overbought and oversold work the same on every timeframe?

The math is identical but the meaning shifts. On lower timeframes extremes fire constantly and carry more noise; on higher timeframes they are rarer and slower to resolve. Many workflows read the higher timeframe for regime and the lower one for timing, so the same reading can be continuation on one chart and stretch on another.

Build Overbought/oversold your way.

Quant writes, tests, and refines it with you — then it runs on LuxAlgo charting or ports to TradingView.