Direct answer
To assess Williams %R, gather the price data it mathematically depends on (high, low, and a reference “current” price), the exact lookback window used for the calculation, and information about how the data is sourced and updated. Then apply quality checks: confirm consistent timeframes, verify that the indicator is computed with the same formula and settings, and evaluate limitations such as sensitivity to the chosen window and the fact that historical relationships do not establish future results.
Mechanism and definition
Williams %R (often written as Williams Percent Range) is a momentum oscillator that compares where a chosen current price sits relative to the highest high and lowest low over a lookback period.
What data you need is therefore specific and minimal:
- A time series of highs and lows for the instrument you are analyzing.
- A reference price for each bar (commonly the latest close, but different charting implementations may allow other choices).
- A lookback length (the window size, for example the number of bars used to find the highest high and lowest low).
From these inputs, you can compute the oscillator value for each bar in the same way the charting tool does. For assessment, treat the indicator mechanics as stable: the calculation is deterministic given the same inputs and settings. What varies is the market data and the implementation choices that determine what counts as “current,” which price series is used, and which timeframe is applied.
Evidence or example checks (without assuming outcomes)
Use a simple, verifiable workflow based on inputs and provenance:
- Choose a timeframe and keep it fixed. Williams %R depends on the bar structure. A daily series and an hourly series can produce different behavior because the highs, lows, and closes differ.
- Confirm the lookback window. If two sources use different window lengths, their values will differ even with identical raw price data.
- Check the reference price used by the implementation. If your tool uses bar close, ensure you are comparing against bar-close series, not intrabar values.
- Verify calculation consistency. Recompute one or two points manually from the same high/low range and the same “current” price used by your chart. This is a strong quality check because it detects mismatched settings or data preprocessing.
Material failure mode: if the “current” value is interpreted differently (for example, using a streaming last price versus a bar close), the oscillator can shift as new ticks arrive. This can make two charts disagree even when both claim to use Williams %R, simply because one is effectively using a different timing convention.
Limitations and risks
Several limitations affect interpretation:
- Timeframe sensitivity: Changing the timeframe or lookback window changes the highest-high/lowest-low range, which can alter the oscillator’s shape and apparent “extremes.”
- Implementation variability: Different charting systems may use slightly different definitions for the reference price (even if they broadly call it “current price”), or may apply different data handling.
- No predictive guarantee from history: Even if Williams %R historically correlated with certain outcomes in a backtest, that relationship does not establish future results.
- External conditions: Costs, execution, and market regime changes can affect what actually happens, but Williams %R by itself does not include those factors.
Rode vlaggen (red flags) to watch
- The lookback window is not stated, or you cannot confirm it.
- The source of high/low data is unclear (for instance, whether it is adjusted, resampled, or derived).
- You compare values across different timeframes without normalizing settings.
Klaarcriterium (done-criteria) for assessment
You have enough information to assess Williams %R independently when you can specify: (1) the timeframe, (2) the lookback length, (3) the exact price series used for high, low, and the reference “current” price, and (4) a reproducible way to validate at least one computed point.
Verification or next question
A useful next question is: What exact settings does your charting source use (lookback window and reference price definition), and do those settings match the data you downloaded or are viewing? If you can answer that, you can verify the indicator computation rather than relying on claims about interpretation.