How FlowMetrics works

What you can configure

• Coin: select the token you want to analyze.
• Whale threshold (USD): defines what counts as a “whale transfer”. Example: 100,000 USD means only transfers worth ≥ $100k are counted.
• Date range: limits the charts and calculations to the selected period.

What the summary cards mean

• Spot price (CoinGecko): reference market price.
• DEX price: on-chain swap price (useful to spot thin liquidity/outliers).
• Netflow signal: a descriptive bias based on recent netflow position (context only).
• Correlation (USD netflow → price): shows whether netflow tends to lead price (lag-based).

Important: correlation describes a relationship—not a guaranteed “predictive” rule.

How to read the price panel

• CoinGecko vs DEX: if DEX deviates strongly, liquidity may be thin or there may be an outlier.
• Daily close: smoothing helps you compare flows to price movement patterns.

If the DEX line is jumpy while CoinGecko is smoother, treat DEX spikes as context, not truth.

How to read whale flows

• Inflow: large transfers into wallets (exchange → wallet) or into a labeled destination.
• Outflow: large transfers to exchanges (wallet → exchange).
• Netflow: inflow minus outflow; positive means more “buy-side” accumulation (contextual).
• Other: internal / unlabeled / neutral transfers (depends on address labeling).

Lag-based correlation (0–6 days)

FlowMetrics checks correlations across multiple lags to answer: “Do flows tend to move before price?”

• Lag 0: same-day relationship
• Lag 1: flows today vs price tomorrow
• Lag 2…6: longer lead times

A stable correlation across time windows is more meaningful than a single spike.

What FlowMetrics is not

• Not financial advice and not a guaranteed signal.
• Address labeling can be incomplete (some flows may be classified as “other”).
• Thin liquidity can distort DEX prices, especially for small-cap tokens.
• Correlation does not prove causation.