Volatility Cycles — How Markets Shift Between Expansion and Compression
Markets do not stay volatile or quiet indefinitely. Understanding when volatility expands — and when it contracts — is central to knowing which models apply.
Volatility is not a fixed property of a market. It expands and contracts in cycles, and those cycles shape which trading approaches work and which ones fail.
This is one of the most important structural realities in systematic trading — and one that is frequently underestimated when building or evaluating models on historical data.
What volatility cycles look like in practice
A volatility expansion phase is characterised by larger price movements per bar, wider average ranges, and often more directional behaviour. Trending conditions tend to emerge during periods of elevated volatility — price moves meaningfully in one direction before consolidating.
A volatility compression phase is the opposite: price action narrows, ranges tighten, and movement per bar shrinks. In these conditions, markets often oscillate within a limited range, with neither buyers nor sellers achieving sustained control.
The transition between these states is not sudden. Periods of compression tend to precede periods of expansion. Markets often consolidate before a breakout, and that consolidation can persist for longer than expected before volatility re-emerges.
This cyclical pattern appears across all major currency pairs and timeframes, though the duration and character of each cycle varies considerably.
Why volatility cycles affect model performance so directly
A trend-following model is designed to capture directional price movement. It performs well when volatility expands and direction is sustained. In a compressed, ranging market, the same model generates frequent false signals — entering on what looks like a breakout only to see price revert.
A range-based model has the opposite problem. It performs well in compressed, oscillating markets but suffers heavily when a genuine trend emerges and price moves well outside its expected range.
This is why market regime classification matters so much. Knowing whether a market is in a trend-dominant or range-dominant state is partly a function of understanding where it sits in its current volatility cycle.
The Average True Range is one of the primary tools used to track volatility state. When ATR is expanding, volatility is increasing and the market is entering a more dynamic phase. When ATR is contracting, compression is underway. Models calibrated using ATR-based position management naturally adapt their stop and target distances to reflect these changes — wider in expanding phases, tighter in compression.
The instability of volatility transitions
The periods most difficult for systematic models are the transitions themselves.
During the move from compression to expansion, price can behave erratically before committing to direction. False breakouts are common — volatility increases briefly, triggering signals, before reverting to a compressed state. This produces stop-outs and whipsaws that are difficult to avoid with entry logic alone.
Conversely, during the transition from expansion to compression, a model may continue to generate trend signals as momentum fades. Entries that would have worked during the peak of the trend now encounter a market that is losing its directional quality.
This is precisely the kind of environment that an effective regime filter is designed to navigate. A filter that can distinguish between genuine expansion and a transient volatility spike — or recognise when a trend is exhausting — provides meaningful protection against the most damaging transitions.
How darwintIQ tracks volatility and regime in real time
The TrendMatrix in darwintIQ provides direction and strength readings across multiple timeframes from M1 to W1. Strength is expressed on a scale of 1–5, reflecting the degree of directional conviction at each timeframe.
When multiple timeframes align on high strength values, the market is likely in an active expansion phase with a clear directional character. When strength readings are low and mixed across timeframes, compression or instability is more probable.
This multi-timeframe view helps identify where a market sits in its volatility cycle without relying on a single indicator in isolation. Models that are currently ranked highly in darwintIQ are, by definition, performing well in the current volatility regime — whatever that regime happens to be. When the regime shifts, the ranking reflects that change.
Final thoughts
Volatility cycles are not a secondary consideration in systematic trading. They are one of the core mechanisms through which market conditions change and model performance shifts. Understanding that markets oscillate between compression and expansion — and that different models suit different phases — is essential context for interpreting any trading signal or evaluation metric.