Gap Risk: When Prices Jump Past Your Stop-Loss Level

Key Takeaways

• Gap risk occurs when new information arrives while markets are closed or liquidity providers pull back, so the next available price is far from the last traded price with nothing in between.
• A biotech stock with a 5% stop-loss can open 35% lower after an FDA rejection, the stop triggers but fills at the opening print, not the stop level.
• A normal-distribution VaR calibrated on ordinary daily returns might report a 99% one-day VaR of 6% on a biotech holding; a gap event can produce a 35% loss in one session.
• Reducing positions before known binary events (earnings, FDA decisions, central bank meetings) and using OTM options for protection are the main operational responses to gap risk.

What It Is

Most quantitative risk models assume prices move along a continuous path. In that world, if your stop is at 95 and you are long at 100, the worst immediate outcome is being filled near 95. Gap risk is the breakdown of that assumption.

A gap occurs when new information arrives while the market is closed, when a liquidity provider pulls back in stress, or when a circuit breaker triggers. The next available price can be far from the last traded price.

Common triggers include overnight earnings releases, geopolitical events, macroeconomic surprises, central bank decisions, and weekend news for instruments that do not trade 24 hours.

The Intuition

A continuous process can be hedged continuously. A discontinuous process cannot. Option pricing theory, stop-loss order strategies, and delta hedging all work best when prices drift smoothly. When a price prints at 100, then at 85 on the next tick with nothing in between, anything based on the assumption of small moves fails at once.

This is why gap risk sits next to jump risk and tail risk in the toolkit. All three describe the fat-tail behaviour that normal distributions cannot capture. Gap risk is the specific case where the discontinuity comes from market structure, not just return distribution.

How It Works

A pure gap model adds a jump term to the standard diffusion:

dS/S = mu dt + sigma dW + J dN

Where dW is the usual Brownian motion, dN is a Poisson process counting jump arrivals, and J is the jump size distribution. The Merton jump-diffusion model is the classic case, with J drawn from a lognormal centred on an expected jump size.

Risk measurement for gap uses three main techniques:

• Overnight return tail analysis. Separate the return series into intraday and overnight components. Compute VaR and expected shortfall on overnight alone. It is almost always wider than intraday.
• Event-conditional scenarios. For each expected event (earnings, central bank meeting, OPEC), a pre-computed gap scenario is applied to the book.
• FRTB Default Risk Charge. For trading books with jump-to-default exposure, Basel's Fundamental Review of the Trading Book imposes a specific capital charge on instantaneous credit jumps, separate from ordinary market risk.

Mitigants include reducing positions into known event windows, buying out-of-the-money options, hedging with variance or jump-sensitive derivatives, and keeping sufficient liquidity to absorb a mark-to-market shock without a forced sale.

Worked Example

Consider a fund holding 10 million of shares in a biotech name the evening before an FDA decision. Implied volatility on the at-the-money straddle is pricing an 18 percent expected move. The fund has a 5 percent stop-loss rule.

• Scenario A: decision is positive. Stock opens up 22 percent. No loss.
• Scenario B: decision is negative. Stock opens down 35 percent. The 5 percent stop cannot fill anywhere near 5 percent below prior close because there is no trade between the prior close and the new opening print. Realised loss: 35 percent, or 3.5 million, not the 500 thousand the stop would suggest.

A VaR model calibrated on ordinary daily returns might report 99 percent one-day VaR of about 6 percent. The actual outcome in scenario B is five times that figure. Either hedging the event with a put spread, reducing position size, or both, is the operational response most institutional desks take into binary events.

Frequently Asked Questions

Q: What is gap risk in simple terms? Gap risk is what happens when prices skip from one level to another without any trades in between. Your stop-loss order fires at the market after the gap, not at the stop level. For binary events like earnings or drug trials, the overnight gap can be many times the normal daily range.

Q: How does gap risk affect investment decisions? Quantitative risk models calibrated on continuous intraday data understate overnight and event-driven risk. Institutional desks routinely reduce or hedge positions before known binary events and maintain explicit event-risk budgets separate from daily VaR.

Q: What is a real-world example of gap risk? A fund holds $10M in a biotech stock before an FDA decision. The stock gaps 35% lower at the open on a rejection. The 5% stop-loss fires near -35%, not -5%, producing a $3.5M loss rather than the planned $500K maximum. The gap bypassed the stop completely.

Q: How can investors protect against gap risk? Before binary events, either reduce the position size to the level you are comfortable losing in a worst-case gap, or hedge with a put spread that pays off specifically in the gap scenario. Pricing the protection against the at-the-money straddle implied move gives a market-based estimate of the expected gap size.

Q: How is gap risk different from jump risk? They describe the same phenomenon from different perspectives. Gap risk is the market-structure view, the next trade printed far from the last because of a liquidity break or market closure. Jump risk is the probability-model view, a Poisson-distributed discontinuous shock added to standard Brownian motion. Both point to the same failure of continuous-price assumptions.

Common Mistakes

1. Trusting stop-loss orders through events. A stop is a market order once triggered. In a gap, it fills at the next available price, which can be far from the stop level. Investors sometimes discover this only after the fact.

2. VaR without overnight decomposition. A 24-hour VaR computed on close-to-close returns mixes in everything, but a 10-hour overnight window around an event has a completely different return distribution. Treating it as equivalent misstates risk.

3. Assuming weekends are quiet. For equities, commodities, and US Treasuries, weekend news (sovereign action, conflict, regulatory announcement) can gap Monday open sharply. Friday close positioning should account for this.

4. Over-relying on circuit breakers. Exchange-level price limits or trading halts do cap the immediate move on the listed exchange, but the underlying risk does not disappear. Related products, off-exchange venues, and the reopening print can still produce the gap.

5. Ignoring correlation gaps. Single-name gap risk is obvious. A gap that hits multiple correlated names at once (macro surprise, sector news) can collapse diversification assumptions at the same moment as each individual position gaps.

Sources

1. Federal Reserve. "Economic Research on Market Microstructure." https://www.federalreserve.gov/econres.htm
2. Basel Committee on Banking Supervision. "Minimum capital requirements for market risk (FRTB)." BCBS d457. https://www.bis.org/bcbs/publ/d457.htm
3. CFA Institute. "Derivatives and Risk Management." https://www.cfainstitute.org/en/membership/professional-development/refresher-readings
4. CME Group. "Price Limits and Circuit Breakers." https://www.cmegroup.com/markets/price-limits.html

Disclaimer

This article is educational content only and is not financial advice. Nothing here is a recommendation to buy, sell, or hold any security. Consult a licensed advisor before making investment decisions.