How to Read Z-Scores for Crypto Trading: Complete Guide
A z-score tells you exactly how far an asset's current price is from its average, measured in standard deviations. It's one of the simplest yet most powerful tools in quantitative trading — and it's especially useful in crypto, where emotions drive prices to statistical extremes regularly.
This guide covers what z-scores are, how to calculate them, how to interpret them across multiple timeframes, and how to use them for crypto trading decisions.
What Is a Z-Score?
A z-score measures how many standard deviations a data point is from the mean. In trading, it tells you whether the current price is statistically "normal" or at an extreme.
- Z = 0: Price is exactly at the mean — perfectly average.
- Z = +1: Price is one standard deviation above the mean — moderately elevated.
- Z = +2: Price is two standard deviations above — statistically rare (only ~2.3% of observations should be here).
- Z = -2: Two standard deviations below the mean — extremely low relative to recent history.
Z-Score Interpretation Ranges
| Z-Score Range | Signal | What It Means | Frequency |
|---|---|---|---|
| Above +2.0 | Extreme Overbought | Price is statistically rare territory above the mean. High probability of mean reversion. | ~2.3% of time |
| +1.0 to +2.0 | Overbought | Price is elevated but not extreme. Could continue or revert. | ~13.6% of time |
| -1.0 to +1.0 | Neutral | Price is within normal range. No strong directional signal. | ~68.3% of time |
| -2.0 to -1.0 | Oversold | Price is depressed. Potential value opportunity if fundamentals hold. | ~13.6% of time |
| Below -2.0 | Extreme Oversold | Statistically rare low. Historically, these have been high-probability entry points. | ~2.3% of time |
Note: These percentages assume a normal distribution. Crypto returns are not normally distributed — they have fat tails. This means extreme z-scores happen more often than 2.3%, which actually makes them more useful as trading signals.
Worked Example: Bitcoin Z-Score
Example: BTC 30-Day Z-Score
Suppose BTC's average price over the last 30 days is $82,000 with a standard deviation of $3,500.
Current price: $93,000
Z = ($93,000 − $82,000) / $3,500 = $11,000 / $3,500
Z = +3.14 → Extreme Overbought
This means BTC is 3.14 standard deviations above its 30-day mean. Statistically, this is very rare and suggests elevated risk of a pullback.
Multi-Timeframe Z-Scores: Why One Window Isn't Enough
A single z-score can mislead. Consider these scenarios:
- 7-day z-score: +2.1, 90-day z-score: -0.5 — The asset had a sharp short-term spike within an overall downtrend. The 7-day signal is overbought but the longer trend is still negative. This is more likely a bear market rally than a new trend.
- 7-day z-score: +0.3, 30-day z-score: +1.5, 90-day z-score: +2.0 — Price has been grinding up consistently. The short-term z-score is calm, but the longer timeframes show accumulating overbought conditions. Trend is strong but extended.
- All timeframes negative (-1 to -2): — Broad oversold conditions across all windows. This is the highest-conviction mean-reversion signal — the asset is depressed on every timeframe.
Fred Intelligence calculates z-scores across four timeframes for each of 15 crypto assets: 7-day, 30-day, 90-day, and 365-day. This multi-timeframe view is what separates noise from signal.
Z-Scores for 15 Crypto Assets
We track z-scores for the top 15 crypto assets by relevance and liquidity:
Large cap: BTC, ETH, BNB, XRP, SOL
Mid cap: ADA, DOGE, AVAX, LINK, SUI
Emerging: ARB, OP, INJ, JUP, APT
Each asset's z-score is calculated using data aggregated from all five exchanges, weighted by volume. This cross-exchange aggregation eliminates single-exchange noise and produces a more reliable statistical measure.
How to Trade Using Z-Scores
Mean Reversion Strategy
When an asset hits extreme z-scores (above +2 or below -2), there's a statistical tendency to revert toward the mean. This is the simplest z-score strategy:
- Buy when z-score drops below -2 (extreme oversold)
- Sell/reduce when z-score rises above +2 (extreme overbought)
- Use stop-losses, since z-scores can stay extreme during trending markets
Divergence Strategy
When different timeframes disagree, it signals potential turning points:
- Short-term oversold, long-term overbought: Could be a dip-buying opportunity in an uptrend, or the start of a reversal. Use regime detection to disambiguate.
- Short-term overbought, long-term oversold: The asset is bouncing from lows. If the long-term z-score is improving, this could be early accumulation.
Portfolio Rotation
Compare z-scores across assets to identify relative value. If BTC has a 90-day z-score of +1.8 and SOL has a 90-day z-score of -0.5, SOL is relatively cheaper by this measure. Rotating into lower z-score assets is a quantitative sector rotation strategy.
Warning: Z-scores are not buy/sell signals in isolation. They measure statistical deviation, not value. An asset can be "extreme overbought" and keep going higher for weeks (see: BTC in any bull cycle). Always combine z-scores with regime context and risk management.
Try the Live Z-Score Calculator
Fred Intelligence publishes live z-scores for all 15 assets daily, calculated from data across 5 exchanges. The 30-day z-scores are free. Multi-timeframe z-scores (7d, 30d, 90d, 365d) are available to subscribers.
Live Crypto Z-Score Calculator
See which assets are statistically cheap or expensive right now.
View Live Z-Scores →Summary
- Z-scores measure how far price is from the mean in standard deviations
- Above +2 is extreme overbought; below -2 is extreme oversold
- Use multiple timeframes (7d, 30d, 90d) — one window isn't enough
- Combine with regime detection for higher-conviction signals
- Z-scores work best for mean-reversion strategies and portfolio rotation
- Cross-exchange data produces more reliable z-scores than single-exchange