The stochastic oscillator, popularised by George Lane in the 1950s, is based on a simple observation: in uptrends, closes tend to sit near the top of the recent range, and in downtrends near the bottom. When that stops being true, momentum may be fading. It reacts faster than RSI, which makes it popular in ranges and for timing pullbacks.
How it’s calculated
Fast %K = (close − lowest low of 14 periods) ÷ (highest high of 14 periods − lowest low of 14 periods) × 100
The common slow stochastic (14, 3, 3) then:
- Slow %K = 3-period simple average of fast %K
- %D = 3-period simple average of slow %K
Reference levels: 80 (overbought) and 20 (oversold). Crosses of %K over %D, especially from beyond those levels, are the classic signals.
Example
Over the last 14 days, a stock’s highest high is $60 and lowest low is $40. Today it closes at $55.
- Fast %K = (55 − 40) ÷ (60 − 40) × 100 = 75
If the previous two fast %K values were 68 and 79:
- Slow %K = (68 + 79 + 75) ÷ 3 = 74
If the last three slow %K values were 70, 78 and 74, %D = 74. The close is in the upper quarter of its range, but not yet in overbought territory.
Common mistakes
- Shorting every reading above 80. In a strong trend, the stochastic can stay overbought for long stretches.
- Using it in trends like in ranges. It works best between clear support and resistance; in trends, use oversold readings only in the trend’s direction.
- Trading every %K/%D cross. The fast settings generate many crosses; filter them with the trend or a level.
- Ignoring divergence context. Divergences matter more at major levels than in the middle of a range.
- Confusing fast and slow versions. Platform defaults differ; check which one you are reading.
For setups and combinations with MACD, see the stochastic oscillator guide.
Educational content — not financial advice.
Updated