Methodology

Collection and data time

Market, derivatives, on-chain and news data arrive from independent sources at different intervals. Every publication records its data cut-off. A delayed source is marked as stale and is never replaced with an invented estimate.

Normalization

Tickers, units and time zones are normalized. Cross-exchange aggregates use only available observations; missing data is not interpreted as zero.

• prices and volumes are checked for valid numeric ranges;
• funding and open interest retain exchange attribution;
• news is deduplicated while preserving the primary-source link.

Interpretation and limits

Metrics describe an observed market state, not a future outcome. Correlation, historical frequency and model output do not prove causality. Each study states its assumptions, gaps and alternative explanations.