What “verification” means for Forex indicator information
Forex indicators are tools that transform price and/or time-related data into values (for example, a line, histogram, or indicator score). Verifying information about them means confirming three things: (1) what the indicator is and how it is computed, (2) what inputs and parameters are required, and (3) what limitations follow from the method and the data used. Verification should not assume that a past relationship will hold in the future, and it should not rely on implied guarantees.
Source hierarchy: where to start and what to trust
A practical hierarchy helps you avoid confusing marketing claims with reproducible facts.
- Primary description of the indicator: look for the original definition, formula, or formal algorithm description. If a provider uses a custom variant, the exact variant must be specified.
- Independent technical documentation: compare the indicator’s definition in multiple reputable technical references (for example, platform documentation) to see whether they describe the same computation.
- Empirical claims (if any): treat performance or “works best” statements as hypotheses. They must include assumptions, data selection rules, and evaluation methods so you can test them.
When you verify, you are mainly checking the mechanics and the test method, not the reputation of the source.
Verification mechanism: reproducible steps you can repeat
Follow the same workflow for any indicator you encounter.
1) Capture the exact definition
Write down the indicator name as given, then record: required inputs (price type such as close, open, or high/low; time frame), parameters (lookback length, smoothing method), and the calculation description (formula or step-by-step rules). If the information is missing a parameter, treat it as incomplete.
2) Check the computation on a fixed dataset
Use one historical dataset and one time frame. Then compute the indicator twice in two different tools, or recompute it manually from the formula. Assumption: both tools use the same input series and parameter values. If outputs differ, identify where the mismatch occurs (data normalization, session handling, or algorithm variant).
3) Validate interpretation rules
Many indicator “signals” are really interpretations applied to the indicator value (for example, “crossing above a threshold”). Verification requires the interpretation to be stated as an explicit rule. Define the rule precisely (threshold value, comparison direction, and whether you require confirmation on the next bar). Then test whether the rule matches the described behavior on the same dataset.
Evidence and example: a falsification mindset
Example of a verification check (no real-time data): suppose a description claims that “indicator X rises when short-term momentum increases.” To verify, choose a dataset and compute both indicator X and a momentum proxy. Then define what would disprove the claim: for instance, find multiple instances where momentum increases but indicator X does not increase (or increases in a different direction than stated). If you can’t reliably find disconfirming cases, that supports the interpretation; if you can, it limits it.
Limitations and failure modes to look for
At least one material limitation should be expected for most indicators:
- Parameter sensitivity: changing lookback length or smoothing can change indicator behavior and therefore the interpretation.
- Regime changes: relationships seen in one market condition (trend vs. range) may weaken in another.
- Costs and execution effects: indicator-based interpretations often ignore spreads, slippage, and latency; these can alter outcomes even if the indicator value behaves as described.
- Data handling differences: time zones, session breaks, and bar construction can produce different inputs, leading to different indicator outputs.
- Overfitting: rules tuned to a small historical period may fail elsewhere.
Verification or next question
If you can’t independently reconstruct the indicator’s computation, treat the information as unverified. A good next question is: Which exact formula, parameters, and input series were used? Then test that definition on a fixed dataset and explicitly define what outcome would confirm or disconfirm the interpretation. This approach keeps you focused on stable mechanics while acknowledging that market conditions and implementation details vary.