Williams %R: The Overlooked Oscillator That Outperforms

Williams %R is one of the most underused tools in technical analysis. Developed by legendary trader Larry Williams (winner of the 1987 World Cup Championship of Futures Trading, turning $10,000 into over $1 million in 12 months), the indicator is mathematically simple yet surprisingly effective. It's often dismissed as "just an inverted Stochastic," but its unsmoothed nature gives it a speed advantage that matters in real trading.

The Calculation

Williams %R measures where today's close sits within the recent high-low range:

%R = ((Highest High - Close) / (Highest High - Lowest Low)) × -100

For a 14-period %R, you look at the highest high and lowest low over the last 14 bars. The result ranges from 0 (close at the highest high) to -100 (close at the lowest low).

Notice the negative scale — this is opposite to most oscillators. A reading near 0 means price is near the top of its range (overbought). A reading near -100 means price is near the bottom (oversold). The thresholds are -20 (overbought) and -80 (oversold).

Why the Negative Scale Matters

The inverted scale isn't just cosmetic. It changes how you read the chart. On RSI and Stochastic, high values are at the top of the panel. On %R, overbought (0 to -20) is at the top and oversold (-80 to -100) is at the bottom. This alignment — overbought at top, oversold at bottom — actually mirrors price more intuitively than the Stochastic's orientation.

More importantly, %R is NOT smoothed. It reacts to every single bar's close. This makes it noisier than the slow Stochastic but also faster to signal changes. If you're watching for the exact moment price leaves overbought territory, %R will signal 1-3 bars before the slow Stochastic.

Trading Strategies

1. Overbought/Oversold Exits

The classic approach: buy when %R crosses above -80 from below (leaving oversold), sell when %R crosses below -20 from above (leaving overbought). In trending markets, only take signals in the trend direction.

On BTC/USDT 4H (2021-2024) with a 200 SMA trend filter (longs only above SMA):

• Signals: 62
• Win rate: 56%
• Profit factor: 1.38
• Average winner: 1.8× average loser

2. Failure Swings

A bullish failure swing: %R drops below -80, bounces above -80, pulls back but stays above -80, then rallies. This pattern indicates buyers are stepping in at oversold levels with increasing conviction. It's less frequent than a simple crossover but more reliable.

On the same BTC 4H dataset, failure swings produced only 18 signals but with a 67% win rate and 1.72 profit factor.

3. Divergence

Because %R is unsmoothed, divergences appear faster. A bullish divergence — price makes a lower low, %R makes a higher low — signals momentum is shifting before the Stochastic shows it. In our ETH/USDT daily test, %R divergence signals preceded Stochastic divergence signals by an average of 2.4 bars.

Williams %R vs. Stochastic vs. RSI

We tested all three oscillators with comparable settings on BTC/USDT 4H using the same strategy (oversold cross + 200 SMA filter):

A clear tradeoff: %R generates the most signals and reacts fastest, but has the lowest win rate per signal. RSI generates the fewest signals and is slowest, but has the highest win rate. Stochastic sits in between.

If you want speed and volume (e.g., you're swing trading and want more opportunities), %R is the best choice. If you want high-conviction signals (e.g., you're position trading and want quality over quantity), RSI is better. For balanced trading, Stochastic is the middle ground.

The Unsmoothed Advantage

%R's lack of smoothing is a feature, not a bug. In fast-moving markets — crypto during a liquidation cascade, forex during an NFP release — smoothed indicators lag. %R immediately reflects the new close relative to the recent range.

This speed advantage is most valuable for exit signals. When a trend is ending, you want to know immediately — not 3 bars later. Many traders use a slower indicator (like RSI or MACD) for entries and %R for exits, combining the best of both worlds.

Practical Tips

Use %R on the same timeframe as your primary trading chart. Because it's fast, using it on a higher timeframe defeats its purpose — you'd lose the speed advantage. The 14-period default works well for 4H charts. On the daily chart, consider 21-period to reduce the noise without losing too much responsiveness.

Avoid using %R on very low timeframes (1m, 5m) without additional filtering. The unsmoothed nature becomes pure noise below 15m on most instruments. If you do trade low timeframes, pair %R with a volume confirmation or a higher-timeframe trend filter.