Z-Score Spread Trading Guide

Z-score tells you how far the current spread is from its normal range. A 0.5% spread may be normal for one coin and extreme for another, so z-score puts every spread in the context of its own history.

• It scores how unusual the current spread is.
• Normal is defined per pair, not across all coins.
• A high absolute z-score means far from normal.

Z-score needs a history of spreads to define an average and a normal range. Without enough history, the score is unreliable, so new or thin pairs can produce misleading values.

• The average spread is the center of normal.
• The recent range sets how wide normal is.
• Thin history makes the z-score less trustworthy.

A trader may look for opportunities when the spread is unusually wide and consider exiting as it returns closer to the average. This is a mean-reversion idea, and it does not guarantee the spread will revert.

• A wide z-score can flag a potential entry.
• A move back toward the mean can flag an exit.
• Mean reversion is a tendency, not a promise.

Spreads can stay abnormal during news, low liquidity, listings, delistings or exchange issues. A high z-score during a real structural change is not a mean-reversion signal, so combine it with liquidity and fees.

• News and events can keep a spread wide for a long time.
• Low liquidity distorts both the spread and the score.
• Always combine z-score with depth and cost checks.