How much of a stock's volatility comes from gaps?

It varies widely by name and period. Overnight returns often contribute a meaningful share of total variance for single stocks, especially around earnings, which is why measuring the split per instrument beats assuming a number.

Do bigger gaps mean the whole day will be more volatile?

Often, but not reliably. Large gaps are associated with elevated intraday ranges on average, yet plenty of large gaps resolve into quiet sessions once the repricing is done.

Are gaps more likely to fill in calm or volatile regimes?

Small gaps in calm regimes fill more often; large gaps in stressed or trending regimes more frequently extend, behaving like breakaway or runaway gaps. Regime context is the useful conditioning variable.

Does this matter for crypto and other 24/7 markets?

Much less, since there is no daily close to gap over. It still appears around exchange outages and in gapping derivatives such as CME crypto futures over weekends.

Gap-volatility Relation

Top Gap-volatility Relation indicator

The top custom implementation, built on the original standard Gap-volatility Relation formula.

1 total

Want to trade Gap-volatility Relation? Put the implementation below on a live chart in Vela, and Quant will help you build a strategy on it and backtest it.

Open in Vela

Gap-volatility Relation

Indicator

What is the gap-volatility relation?

The gap-volatility relation is the two-way link between opening gaps and volatility: gaps are both a component of an instrument's total volatility and a symptom of its current volatility regime. Whenever a market closes, information keeps arriving, and the next open reprices it all at once as an opening gap. That overnight jump carries real variance, often a substantial fraction of total variance for single stocks, yet it is invisible to any measure built only from intraday ranges.

This is why the relation matters mechanically for measurement. Range-based tools such as the Parkinson estimator see only movement inside the session, so they understate risk on gapping instruments (true-range measures partially correct for this by spanning the prior close), while close-to-close volatility captures gaps but mixes them indistinguishably with intraday movement. The Yang-Zhang estimator was designed around exactly this decomposition, estimating overnight and open-to-close variance separately and combining them.

The relation also runs the other way as a regime signal. Gap frequency and gap size tend to expand and contract with the broader volatility environment: quiet regimes produce small, frequently filled gaps, while stressed regimes produce large gaps that more often extend rather than fill. Traders watch the character of recent gaps as an early, tangible read on regime shifts and on event-driven volatility around earnings and macro releases.

How it's calculated

The standard decomposition separates overnight from intraday returns:

r_overnight = ln(open_t / close_(t-1))

r_intraday = ln(close_t / open_t)

r_total = r_overnight + r_intraday

sigma_overnight^2 = (1 / (n - 1)) * sum (r_overnight - mean_overnight)^2

sigma_intraday^2 = (1 / (n - 1)) * sum (r_intraday - mean_intraday)^2

overnight share = sigma_overnight^2 / (sigma_overnight^2 + sigma_intraday^2)

open_t, close_t: open and close of day t

n: number of days in the estimation window

mean_overnight, mean_intraday: window means of the respective returns

The additive variance split ignores any covariance between overnight and intraday returns; Yang-Zhang handles the combination formally.

For near-24-hour markets the overnight term shrinks toward zero and the relation matters mainly around weekend closes and exchange halts.

How traders use it

Choosing estimators honestly: traders check an instrument's overnight variance share before trusting range-based volatility numbers; a high share argues for close-inclusive or Yang-Zhang estimates.
Sizing around holds: overnight and weekend position sizing is scaled to gap risk rather than intraday risk, since stops cannot protect through a weekend gap.
Reading regime from gap character: a run of larger, unfilled gaps is treated as evidence of regime escalation, consistent with broader range expansion in other volatility measures.
Conditioning gap trades: the odds of a gap fill versus continuation shift with the volatility backdrop, so gap-fade strategies are typically throttled in high-volatility regimes.
With honest limits: overnight variance shares drift over time and differ sharply between asset types, so any fixed assumption about gap risk should be re-measured rather than carried across instruments.

Gap-volatility relation vs adjacent concepts

Event-Driven Volatility: Event-driven volatility is about scheduled catalysts producing predictable variance spikes. The gap-volatility relation is the structural accounting of where overnight variance lives, whatever its cause.

Weekend / Overnight Volatility Profile: The overnight profile describes how variance accrues across closed and thin hours. The gap-volatility relation focuses on the measurement and regime consequences of the jump printed at the open.

Breakaway Gap: Gap taxonomy classifies individual gaps by their role in a pattern. The gap-volatility relation is statistical, treating gaps in aggregate as a variance component and a regime indicator.

Gap-volatility Relation FAQ

Turn Gap-volatility Relation into a trading strategy.

Open the implementation from this page in Vela. Quant, our coding agent, builds the strategy with you and backtests it on real data.

Build a strategy in Vela

Browse this familyVolatility61 concepts mapped · 61 in the Library

Previous conceptFibonacci Bollinger Bands Next conceptGARCH-family Clustering