Guide

What Is Crypto Market Regime Detection? A Data-Driven Guide

Every crypto market exists in one of four states at any given time. These aren't arbitrary labels — they're statistically identifiable regimes that determine how prices behave, how volume flows, and what strategies work. Understanding which regime the market is in right now is arguably the single most important input for any trading or investment decision.

This guide explains what market regimes are, how they're detected using real exchange data, and how you can use regime signals in your own analysis.

The Four Market Regimes

Market regime theory draws from Wyckoff analysis and modern quantitative finance. The concept is simple: markets cycle through four distinct phases, each with different statistical properties.

Regime	Characteristics	Typical Duration	What to Watch
Accumulation	Low volatility, range-bound, smart money buying	Weeks to months	Volume spikes on dips, narrowing range
Markup	Trending up, increasing volume, breakouts	Weeks to months	Higher highs, higher lows, expanding volume
Distribution	High volatility, range-bound at top, smart money selling	Weeks	Volume spikes on rallies, failing breakouts
Markdown	Trending down, capitulation volume, fear rising	Weeks to months	Lower lows, lower highs, sentiment collapse

How Regime Detection Works

Modern regime detection uses multiple data inputs to classify the current market state. At Fred Intelligence, our model processes data from five major exchanges (Binance, Kraken, Bybit, OKX, and Coinbase) daily. Here's what goes into it:

1. Price Momentum (Trend Strength)

We calculate moving averages across multiple timeframes (7-day, 30-day, 90-day) and measure the relationship between them. When short-term MAs are above long-term MAs and the gap is widening, that's markup. When they're converging, the regime may be shifting.

2. Volatility Analysis

Each regime has a different volatility signature. Accumulation and markup tend toward moderate, declining volatility. Distribution shows increasing volatility with large intraday ranges. Markdown often features volatility spikes (capitulation events). We measure rolling standard deviation and compare it to historical percentiles.

3. Volume Profile

Volume tells you who is participating. In accumulation, volume is typically low overall but spikes on down moves (smart money absorbing supply). In distribution, volume spikes on up moves as holders distribute to new buyers. We analyze volume across all five exchanges to avoid single-exchange anomalies.

4. Cross-Exchange Consistency

A true regime should be visible across exchanges. If Binance shows accumulation signals but Coinbase shows distribution, the signal is weak. Our model weights cross-exchange agreement as a confidence multiplier.

5. Sentiment Overlay

We incorporate the Fear and Greed Index and prediction market data as confirming indicators. Extreme fear during accumulation is a high-confidence signal. Extreme greed during distribution is similarly informative.

Statistical Methods Behind the Model

Several quantitative approaches are used in regime detection. Here's what the research supports:

• Hidden Markov Models (HMMs): The most common approach. HMMs model the market as having hidden states (regimes) that produce observable outputs (returns, volume). The model learns transition probabilities between states.
• Z-Score Composites: Z-scores across multiple timeframes create a composite signal. An asset with extreme negative z-scores across 7d, 30d, and 90d windows is likely in markdown or late accumulation.
• Volatility Clustering: GARCH models identify periods of high and low volatility clustering, which map directly to regime characteristics.
• Trend Decomposition: Separating price into trend, seasonal, and residual components helps identify whether the dominant force is momentum (markup/markdown) or mean-reversion (accumulation/distribution).

Why Multi-Exchange Data Matters

Single-exchange analysis is the biggest mistake in crypto regime detection. Here's why:

• Volume manipulation: Some exchanges inflate volume. Cross-referencing eliminates fake signals.
• Regional differences: Binance reflects Asian trading patterns, Coinbase reflects US institutional flows. A true regime signal should appear across both.
• Liquidity depth: Order book depth varies dramatically. Bybit and OKX derivatives data adds a layer that spot-only analysis misses.
• Funding rates: Perpetual swap funding rates on Bybit and OKX signal leveraged positioning that influences regime transitions.

Fred Intelligence aggregates data from Binance, Kraken, Bybit, OKX, and Coinbase — no paid APIs, no middleman data providers. Direct exchange data means no lag, no filtering, no third-party bias.

How to Use Regime Signals

Once you know the current regime, your strategy should adapt:

Regime	Strategy Implications
Accumulation	DCA into positions. Look for assets with extreme oversold z-scores. Reduce cash allocation gradually. Timeframe: patient, multi-week entries.
Markup	Hold positions. Add on pullbacks that hold higher lows. Avoid selling into strength. Let winners run. This is where profits are made.
Distribution	Tighten stops. Take partial profits on overbought z-scores. Increase cash allocation. Watch for failed breakouts as exit signals.
Markdown	Preserve capital. Minimal exposure. Monitor for accumulation signals. This regime rewards patience, not action.

Current Regime: Check Live Data

Fred Intelligence publishes the current regime status daily, based on data from all five exchanges. You can check it live using our free tool:

For multi-timeframe z-scores that complement regime analysis, see our Z-Score Calculator.

Further Reading

• Wyckoff, R. — "Studies in Tape Reading" (original regime theory)
• Hamilton, J. (1989) — "A New Approach to the Economic Analysis of Nonstationary Time Series" (Hidden Markov Models for regime switching)
• Ang & Bekaert (2002) — "Regime Switches in Interest Rates" (applied regime models)