Define Low Liquidity Pairs before checking claims
A “low liquidity pair” is best treated as a characterization of trading conditions rather than a fixed label. In practice, liquidity is reflected in how easily traders can transact with limited price disruption. Information about a specific pair being “low liquidity” should therefore be verified in terms of measurable proxies (for example, order-book depth, trading volume, or bid–ask spread behavior), not as a universal property.
Because market conditions change, the same pair can appear more or less liquid at different times. So, verification should start with clear definitions and measurable indicators, then move to checking whether any claim uses those indicators consistently.
Use a source hierarchy that matches what you want to verify
To verify information, use a hierarchy that starts from stable references and moves toward variable, market-specific data.
-
Stable concepts (no-time claims). Use general explanations of what liquidity means, common proxies, and why liquidity can vary. These are generally verifiable without relying on real-time prices.
-
Method definitions (how the indicator is computed). If a source says a pair is low liquidity, check what it actually measured: spread type (quoted vs effective), volume definition, time window, and whether it uses spot FX venues or aggregated feeds.
-
Market-condition evidence (time-dependent claims). For variable assertions (for example, “low liquidity during these hours”), verify using multiple time windows and independent data sources. Avoid trusting a single screenshot or one moment in time.
-
Provider and execution context (how your results might differ). Even if a pair is low liquidity “in general,” the experienced cost depends on execution quality, quote aggregation, and trading costs. Verification should therefore include provider documentation and the assumptions behind any cost estimate.
For a deeper concept overview, you can use the internal page on low liquidity pairs: low liquidity pairs.
Reproducible verification steps (no real-time data required)
Follow steps that let a reader reproduce the check with offline or archived data.
-
Write down the claim you want to verify. Example template: “Pair X is low liquidity.” Specify what “low” means by the source’s metric (spread size, depth, or volume).
-
List the indicator(s) and time window. If the source refers to spread, note whether it is a quoted spread or an effective spread proxy, and whether the window is hourly, daily, or event-based.
-
Collect data with consistent assumptions. Use the same time granularity across periods. If you compare two sources, ensure they refer to comparable instruments and time zones.
-
Compute a simple summary statistic. Examples:
- Spread proxy: average bid–ask spread (and its variability) over the window.
- Trading activity: average volume or turnover over the same window.
- Dislocation: how much the mid-price moves when spreads widen (requires caution; see limitations).
-
Cross-check with a second proxy. Liquidity can look “low” under one metric but not another. A reproducible verification includes at least two independent indicators.
-
Document the calculation and rerun it. Save your formulas, filters, and any exclusions (for example, missing data). If another person cannot rerun the steps to reach the same conclusion, the claim is not well-verified.
If you want guidance on what data is needed, use: what data is needed to assess low liquidity pairs?.
Material limitations and failure modes to check
Even well-measured liquidity proxies can fail in ways that change interpretation.
-
Stale quotes and sampling bias. A dataset may contain quotes at uneven intervals, causing apparent spreads or depth to look worse or better than reality.
-
Effective costs vs quoted metrics. Quoted spread may understate the cost you actually experience if execution quality differs from the assumed model.
-
Regime changes. Liquidity can shift due to news, market hours, or risk appetite. Historical relationships do not guarantee future results.
-
Provider-specific aggregation. Different data vendors or execution venues may compute liquidity proxies differently (for example, how they aggregate across venues or how they treat outliers).
-
Hidden variability around events. Low liquidity pairs can become more volatile or more costly around macro releases or liquidity “gaps,” meaning a single average can hide important extremes.