The stochastic oscillator is a momentum indicator that asks a narrow question: where does the current price sit within its recent range? If a market has traded between 1.2500 and 1.2600 over the lookback and price is right at the top, the stochastic reads high; near the bottom, it reads low. It is drawn as two lines bouncing between 0 and 100 in a panel below the chart, and traders use it to judge whether a move is stretched and losing steam.
What it measures
The idea, from George Lane in the 1950s, is that momentum turns before price. In an uptrend, closes tend to cluster near the top of each bar's range; when they start slipping towards the bottom even as price holds up, the buying is tiring. The oscillator captures this by placing the latest close on a 0-to-100 scale against the highest high and lowest low of the lookback — 14 periods is the common default.
%K and %D
Two lines:
- %K — the faster, rawer line. It plots where the close sits in the range right now.
- %D — a moving average of %K, usually over three periods. Slower and smoother.
%D is the signal line. When %K crosses %D, that crossover is the trigger many systems act on — much like the fast and slow lines of a moving-average cross, but bounded to a fixed range.
Overbought and oversold
Two thresholds matter:
- Above 80 — overbought. Price is near the top of its recent range.
- Below 20 — oversold. Price is near the bottom.
The tempting reading is "sell above 80, buy below 20". In a range that often works. In a trend it is a good way to lose money. A strong uptrend can pin the stochastic above 80 for days, and shorting each "overbought" reading means fighting the trend the whole way up. Overbought is not a sell signal on its own — it is a description of where price is, nothing more.
"Overbought" does not mean "about to fall". A market can stay overbought far longer than you can stay solvent shorting it. The signal is only worth acting on in the right context — usually a range, or alongside the trend rather than against it.
Divergence
The setup many traders rate most highly is divergence — when the oscillator and price disagree. If price grinds to a new high but the stochastic makes a lower high, momentum is fading under the surface even as price ticks up. That is bearish divergence. The mirror — price at a new low while the stochastic makes a higher low — is bullish divergence, a hint that selling is running out.
Say GBP/USD pushes to a fresh high but the stochastic prints a clearly lower peak. You read weakening momentum and short 0.3 lots (about £2.30 per pip) with a 40-pip stop — roughly £92 at risk. If the divergence resolves into a 90-pip drop, that is about £207 back. If price simply carries on up, you are out for £92, because divergence warns of fading momentum without promising a reversal will follow.
A caution
The stochastic is a range-and-momentum tool, not a trend detector. Its worst failures come from using overbought and oversold as automatic sell and buy signals in a trending market. Divergence is more useful but early — it can print several times before price finally turns. Pair it with a read on the trend and treat it as one voice, not the verdict.
The Karnek note
Karnek does not plot the stochastic or flag divergence — it monitors your account, not your charts. What it shows is whether a momentum EA built on these signals actually earns on your live terminal, across enough trades to judge. It is read-only and can never place a trade.
Written and reviewed by the Karnek Research team. Last updated August 2026.
Educational content only - not financial advice. Past performance does not predict future results. Trading carries significant risk.