Direct answer
To verify information about Bollinger Range, you need to confirm three things independently: (1) the definition and exact formula being referenced, (2) the inputs and parameter choices used in any example, and (3) the arithmetic outcome when you reproduce the indicator with the same data and assumptions. Because market conditions change and provider settings can differ, verification should focus on the mechanics and reproducibility rather than promises about future performance.
Mechanism and definition
Bollinger Range is usually described in terms of Bollinger Bands, which are commonly built from a moving average and a measure of dispersion. A typical structure is:
- Choose a lookback window length (often denoted as N).
- Compute a moving average of the selected price series over that window.
- Compute a dispersion measure, typically the standard deviation over the same window.
- Construct upper and lower bands by adding/subtracting a multiple of the dispersion from the moving average.
Information you find online may vary in details. For verification, treat the following as explicit assumptions you must match to the source:
- Price input: close price, typical price, or another field.
- Window length: the number of periods used.
- Deviation multiplier: how many standard deviations are applied.
- Moving average type: simple vs. exponential (if the source specifies one).
“Bollinger Range” in practice often refers to the distance between the upper and lower bands (or a related width measure). If a page claims a specific “range” definition, verify it by locating the exact calculation it uses and ensuring you can map it to the bands it describes.
Evidence and reproducible verification steps
Below is a reproducible process that does not require real-time data. It works with any historical price series you can obtain consistently.
Step 1: Freeze the definition
Write down the exact formula from the source you want to verify, including:
- N (lookback length)
- deviation multiplier
- moving average type
- which price series is used
- how “range” is defined (band width vs. other wording)
If the source omits a parameter, you cannot fully verify its numeric results; instead, you can only verify the general mechanics.
Step 2: Recompute with the same inputs
Using a spreadsheet or a programming environment, compute:
- the moving average for each period where the window is complete,
- the standard deviation over the same window,
- the upper and lower bands,
- the “range” measure as defined by the source.
Step 3: Cross-check with at least one independent method
To reduce implementation errors, compute the same indicator twice—e.g., once in Excel-like formulas and once in a script, or using two independent libraries/tools. You are verifying consistency of the arithmetic under the same assumptions.
Step 4: Validate with the source’s example (if present)
If a page provides a worked example (numbers for a date range), repeat it. The verification is successful only if your outputs match the page’s outputs within reasonable rounding rules you apply consistently.
Step 5: Separate “what is computed” from “what it means”
After verifying the computed series, independently evaluate any interpretation claims by checking whether they depend on unstated conditions. For example, claims framed as general rules may fail when you change market regime, data frequency, or parameter settings.
Limitations and material failure modes
Several limitations can cause misinformation or misunderstandings:
- Parameter mismatch: different N, deviation multiplier, or moving average type will produce different bands.
- Inconsistent price inputs: using close instead of another price field changes the computed result.
- Rounding and standard deviation conventions: sample vs. population standard deviation choices can shift values.
- Missing or misaligned data: corporate actions, holiday gaps, or differing time zones can distort windows.
- Low volatility periods: dispersion measures can compress so “range” becomes less informative for distinguishing conditions.
Also, even if the indicator is computed correctly, any relationship between indicator behavior and future outcomes is conditional. Outcomes vary with market conditions, trading frictions (spreads/fees), execution quality, and jurisdiction rules. Therefore, verification of the computation is not the same as verification of predictive performance.
Verification or next question
When you encounter a claim about Bollinger Range, the most reliable next question is: “Which exact formula and parameter choices are used, and can I reproduce the numeric series from a clearly specified input price dataset?