Momentum

Williams %R Explained: The Inverted Oscillator That Spots Overbought Crypto Before the Drop

📅 June 02, 2026· 7 min read· Strategester
0 -20 -50 -80 Overbought Oversold Sell zone Buy zone ─── -20 ─── -80 Williams %R (14-period)

Most oscillators run from 0 to 100. Williams %R flips the scale — it runs from 0 to -100. That inversion trips up a lot of traders, but once you understand it, %R becomes one of the fastest overbought/oversold reads available. It measures exactly where the current closing price sits within the highest high and lowest low over a lookback period, expressed as a negative percentage.

Larry Williams introduced the indicator in 1973, and it remains widely used in crypto precisely because it reacts quickly to short-term extremes — a quality that matters when BTC can move 8% in a single session.

The Maths Behind the Inversion

The formula is deceptively simple. For a 14-period %R on a 1-hour chart:

%R = (Highest High₁₄ − Close) / (Highest High₁₄ − Lowest Low₁₄) × −100

Walk through a concrete example. Say over 14 hourly candles BTC reaches a high of $68,400 and a low of $65,200. The current close is $67,900.

%R = (68,400 − 67,900) / (68,400 − 65,200) × −100 = 500 / 3,200 × −100 = −15.6

A reading of -15.6 means price is very close to the top of its 14-candle range — which is why values near 0 signal overbought conditions, not oversold ones. Conversely, if price were near the bottom of the range, the numerator would be large and %R would approach -100 (oversold).

The Two Key Thresholds

Nearly all %R strategies use the same two reference lines: -20 and -80.

When %R climbs above -20 (i.e., a value between -20 and 0), price is in the top 20% of its recent range — overbought territory. When %R falls below -80, price is in the bottom 20% — oversold. These mirror the 80/20 levels on a standard RSI but with the sign flipped.

Key rule: On Williams %R, high numbers mean oversold, low numbers mean overbought — the opposite of RSI. A reading of -5 is extreme overbought; a reading of -95 is extreme oversold. Print this rule somewhere visible until it's second nature.

Trading Williams %R in Crypto

Zone Exits — the Classic Setup

The most common entry signal is not when %R enters the overbought or oversold zone, but when it exits it. Waiting for the exit avoids chasing momentum that could continue running for many candles.

For a long entry: wait for %R to drop below -80 (oversold), then take the trade when it crosses back above -80. On a 1-hour ETH chart this has historically coincided with short-term bounce setups after fast sell-offs. For a short: wait for %R to rise above -20 (overbought), then enter when it crosses back below -20.

A 14-period default is standard, but crypto traders frequently use a shorter 9-period %R on 15-minute charts for intraday scalping entries, and a 21-period %R on 4-hour charts for swing setups.

Divergence — the Higher-Conviction Signal

Divergence on %R works the same way as on RSI or MACD: price makes a new high but %R makes a lower high, suggesting momentum is weakening ahead of a potential reversal. On a 4-hour BTC chart, look for:

Bearish divergence: price prints a higher high but %R prints a lower high (below -20 on both peaks). This pattern often precedes sharp pullbacks. Bullish divergence: price makes a lower low but %R makes a higher low (both in the -80 to -100 zone). This is a classic accumulation signal before a recovery leg.

Practical tip: Williams %R divergence is most reliable when the second peak or trough fails to reach the extreme zone. If price makes a new high but %R only reaches -28 instead of -8 on the previous high, that's a meaningful loss of momentum — even though %R still looks "overbought" at -28.

Pairing %R with a Trend Filter

Like all oscillators, %R generates excessive false signals in strong trends. A 14-period %R on a runaway BTC bull move will sit near 0 for dozens of candles while shorters get destroyed. The fix is simple: only take %R signals that align with the higher-timeframe trend.

A common approach is to use a 200-period EMA on the same chart as your %R. If price is above the 200 EMA, only trade %R signals from the oversold side (longs). If price is below the 200 EMA, only trade %R signals from the overbought side (shorts). This single filter eliminates most of the countertrend noise.

Williams %R vs Stochastic Oscillator

The two indicators are almost mathematically identical. Stochastic %K uses the same formula as Williams %R but inverts the sign and uses the lowest low in the numerator instead of the highest high. The practical difference is that Stochastic smooths its signal line with a moving average (%D), making it slightly slower. Williams %R is the raw, unsmoothed version — faster to react, noisier on lower timeframes.

On Strategester's Mix & Backtest tab, you can combine %R-style momentum signals (via the Stoch strategy) with trend filters like Supertrend or EMA crossover to see exactly how adding a momentum oscillator changes backtest metrics on 90 days of real Bybit candle data.

Timeframe guide: 9-period %R on 15m for scalps. 14-period %R on 1h for intraday swings. 21-period %R on 4h for multi-day setups. As period increases, fewer but more reliable signals — each one reflects a wider price range and more market context.

Common Mistakes

The sign inversion trips up beginners every time. Double-check: a value of -5 means nearly overbought (close to 0), not nearly oversold. Second, avoid entering the moment %R touches -80 or -20 — in strong trends, it can stay pinned there for 20+ candles. Always wait for the cross back out of the zone. Third, %R is a short-term tool; it resets with every new bar, so it's not designed to track multi-week trends the way EMA or Ichimoku cloud does.

On Strategester, you can overlay the Stochastic strategy (the closest live equivalent) on any of the 32 markets to watch %R-style dynamics in real time as candles form.

See Williams %R live on any market

Open Strategester — free, no account needed. Live data on 32 crypto markets.

Open Strategester →
Williams %R Momentum Oscillator Overbought Oversold Crypto Indicators Technical Analysis Divergence