Williams %R Indicator: Complete Trading Strategy Guide

Williams %R, developed by legendary trader Larry Williams in 1973, is a momentum oscillator that measures the current closing price relative to the highest high over a lookback period. Despite being one of the simplest indicators to calculate and interpret, Williams %R provides highly effective signals for identifying overbought and oversold conditions, momentum shifts, and potential trend reversals. Its inverse relationship with the Stochastic Oscillator makes it a natural complement to any momentum-based trading framework.

This guide covers the indicator's mechanics, signal interpretation, and three proven trading strategies with specific entry, exit, and risk management rules.

Williams %R Calculation

The formula for Williams %R is:

Where:

• Highest High = Highest high over the lookback period (typically 14 periods)
• Lowest Low = Lowest low over the lookback period
• Close = Current closing price

• -100 = The close equals the lowest low of the lookback period (maximum oversold)
• 0 = The close equals the highest high of the lookback period (maximum overbought)

Relationship to the Stochastic Oscillator

Williams %R is mathematically the inverse of the Fast Stochastic %K, multiplied by -100:

• Stochastic %K measures how close the price is to the period high (0 to 100, higher = closer to high)
• Williams %R measures how far the price is from the period high (-100 to 0, closer to 0 = closer to high)

Interpreting Williams %R Levels

The Trend Context Rule

As with all oscillators, overbought does not mean "sell" and oversold does not mean "buy." In strong uptrends, Williams %R can remain above -20 for extended periods as the close consistently prints near the period's high. During these persistent overbought conditions, selling based on the %R reading alone produces consistent losses.

Trading Strategy 1: Failure Swing

The failure swing is Larry Williams's preferred signal and one of the most reliable patterns the indicator produces.

Bullish Failure Swing

1. Williams %R drops below -80 (enters oversold territory)
2. %R rises back above -80
3. %R pulls back but stays above -80 (fails to re-enter oversold territory)
4. %R then moves higher, confirming the failure swing

Bearish Failure Swing

1. Williams %R rises above -20 (enters overbought territory)
2. %R drops below -20
3. %R rises again but stays below -20 (fails to re-enter overbought territory)
4. %R drops further, confirming the failure swing

Trading Strategy 2: Divergence Trading

Divergence between Williams %R and price provides some of the highest-probability reversal signals.

Bullish Divergence

Price makes a new low, but Williams %R makes a higher low (does not confirm the price low). This indicates that despite lower prices, the closing price is higher relative to the period's range, meaning selling momentum is diminishing.

Bearish Divergence

Price makes a new high, but Williams %R makes a lower high. Despite higher prices, the close is lower relative to the period's range, indicating weakening buying momentum.

• Identify the divergence on the daily chart
• Wait for Williams %R to cross out of the overbought/oversold zone in the direction of the divergence
• Enter on the cross with a stop beyond the recent price extreme
• Target: The 50-period SMA as the initial target, then trail

Trading Strategy 3: Multi-Timeframe Momentum

This strategy uses Williams %R across two timeframes to capture entries within the prevailing trend.

Setup

1. Weekly Williams %R: Determines the trend bias

• Weekly %R above -50: Bullish bias (only take long entries)
• Weekly %R below -50: Bearish bias (only take short entries)

1. Daily Williams %R: Provides the entry signal

• In a bullish weekly bias: Enter long when daily %R crosses above -80 (exits oversold)
• In a bearish weekly bias: Enter short when daily %R crosses below -20 (exits overbought)

Exit Rules

• Profit target: When daily %R reaches the opposite extreme (-20 for longs, -80 for shorts)
• Stop-loss: 2x ATR below entry (longs) or above entry (shorts)
• Trailing stop: Move stop to breakeven when daily %R reaches -50, then trail using the Chandelier Exit

Williams %R Settings Optimization

The standard 14-period setting works well for daily charts across most markets. Alternative settings for different applications:
| Application | Period | Rationale |
|-------------|--------|-----------|
| Scalping (1-5 min) | 7-10 | Faster response for short timeframes |
| Day trading (15-60 min) | 14 | Standard setting |
| Swing trading (daily) | 14-21 | Balanced sensitivity |
| Position trading (weekly) | 14 | Standard on weekly bars |
| Trend confirmation | 28 | Smoother, fewer false signals |

Avoid over-optimizing the period setting. The standard 14-period setting is the most widely used, meaning more traders are watching the same levels, which creates self-reinforcing behavior at those readings.

Python Implementation

import pandas as pd
import numpy as np
def williams_r(df, period=14):
    highest_high = df['High'].rolling(period).max()
    lowest_low = df['Low'].rolling(period).min()
    wr = ((highest_high - df['Close']) / (highest_high - lowest_low)) * -100
    return wr
def williams_r_signals(df, period=14, ob=-20, os=-80):
    df['WR'] = williams_r(df, period)
    df['WR_prev'] = df['WR'].shift(1)
# Oversold exit (bullish signal)
    df['Buy_Signal'] = (df['WR_prev'] < os) & (df['WR'] >= os)
# Overbought exit (bearish signal)
    df['Sell_Signal'] = (df['WR_prev'] > ob) & (df['WR'] <= ob)
return df

Key Takeaways

• Williams %R measures the current close relative to the highest high over the lookback period, ranging from -100 (oversold) to 0 (overbought).
• The indicator is the mathematical inverse of the Fast Stochastic %K and provides equivalent information with an inverted scale.
• Overbought/oversold readings are conditions, not automatic signals. Always filter by the prevailing trend direction before acting.
• The failure swing pattern (inability to return to the extreme zone after exiting it) is one of the most reliable Williams %R signals.
• Divergence between price and Williams %R at key support/resistance levels provides high-probability reversal setups.
• Multi-timeframe analysis (weekly trend direction + daily entry signal) significantly improves win rates by ensuring entries align with broader momentum.

Frequently Asked Questions

How is Williams %R different from the Stochastic Oscillator?

Williams %R is the mathematical inverse of the Fast Stochastic %K. While the stochastic measures how close the price is to the period's high (0-100 scale), Williams %R measures how far the price is from the period's high (-100 to 0 scale). The stochastic adds a signal line (%D) and smoothing, which Williams %R does not include by default. In practice, many traders find Williams %R simpler to interpret because it uses a single line without signal line crossovers.

What is the best Williams %R setting for forex?

The standard 14-period setting works well for forex trading on all timeframes. For scalping on 1-minute to 5-minute charts, reducing the period to 7 or 10 can provide faster signals. For longer-term forex analysis on daily and weekly charts, the 14-period setting remains optimal. The most important factor is not the period setting but the trend filter, as taking only trend-aligned signals dramatically improves performance in directional forex markets.

Can Williams %R be used as a standalone indicator?

While Williams %R provides valuable information on its own, it performs best when combined with trend-identification tools. At minimum, pair Williams %R with a moving average (50 or 200 SMA) to determine trend direction, and only take %R signals in the direction of the trend. Using Williams %R in isolation, particularly in mean-reversion mode during strong trends, is one of the most common causes of losses with this indicator.

How do you handle Williams %R signals in choppy markets?

In rangebound or choppy markets, Williams %R tends to oscillate between overbought and oversold zones frequently, producing many signals with lower reliability. To filter these, add a volatility filter such as the ADX. When ADX is below 20 (non-trending), either avoid Williams %R signals entirely or require them to occur at clearly defined horizontal support and resistance levels. When ADX is above 25 (trending), Williams %R signals in the trend direction become much more reliable.