Gapping and weekend risk

A price chart drawn from trades is a record of levels at which somebody actually transacted. When a market reopens at a price several levels away from the one it closed at, no transaction took place at any level in between, and the chart shows a blank space. That space is a gap, and it is not a drawing artefact. Nothing traded there.

The mechanism is the same whether the interruption lasts a weekend, an overnight session break or a few seconds around a scheduled release. Information arrives while the market is closed or while nobody is willing to quote, opinions about the correct price adjust, and the first price at which somebody is willing to transact again is somewhere other than the last one. The market does not travel to the new level. It appears at it.

Why a closed market reopens elsewhere 

Markets close. News does not. Company announcements are conventionally released outside trading hours precisely so that the market has time to digest them, central banks and governments act at weekends when the alternative would be acting into an open market, and geopolitical events keep no schedule at all. Every hour a market is shut is an hour in which the fair price can move while the quoted price cannot.

The size of a reopening gap therefore tends to scale with the length of the closure and with how much happened during it. A currency market closed from Friday evening to Sunday evening carries a shorter exposure than a single stock market closed from Friday afternoon to Monday morning, which is one reason the two behave differently at the open. An instrument that trades almost continuously gaps rarely and by small amounts; an instrument with a long daily break gaps routinely.

Gaps also occur inside an open session. A scheduled economic release can move a price faster than quotes can be refreshed, and in the seconds around it the book can empty entirely, so the next price printed is well away from the last. A disorderly move in one instrument can do the same to correlated instruments that had no news of their own.

What happens to orders resting inside the jump 

This is the part that surprises people, and it follows from a single fact: an order rests at a price, and a price that never trades cannot fill an order at it. When a market opens beyond a resting stop order, the stop's trigger condition is satisfied, because the price is now past the level. The order is therefore released into the market, and the market's first available price is on the far side of the gap. It fills there.

A stop order is an instruction to transact once a level is passed, not a promise to transact at that level. The distinction is invisible in a continuous market, where the next available price after a trigger is usually a tick away, and it is the whole story in a gapped one, where the next available price can be a long way away. Every stop resting inside the jump triggers at once, and all of them are filled at the reopening price.

Limit orders resting inside a gap are treated differently, and the difference is worth stating because it is the mirror image. A limit order carries a price boundary that must be respected, so a buy limit inside the jump can only be filled at its level or better. When the market opens beyond it in the favourable direction, it fills at the better opening price. When the market opens beyond it in the other direction, it does not fill at all and continues to rest.

The one order type that does guarantee a level 

A guaranteed stop is a separate instrument rather than a setting on an ordinary stop. Where it is offered, the firm undertakes to close the position at the stated level regardless of where the market actually trades, absorbing the difference itself. Because that undertaking is a real liability for the firm, it is priced: guaranteed stops conventionally carry a premium, are available on a restricted list of instruments, and may require a minimum distance from the current price.

Separately, many regulators require firms to operate negative balance protection for retail clients, which limits an account's liability to the funds in it after a catastrophic move. That is an account level protection applied after the fact, not a per position guarantee, and it does not prevent the position from closing at the gapped price. The two mechanisms address different points in the same sequence and are not substitutes for one another.

Where gaps cluster 

Gapping is not evenly distributed across the calendar, and its concentration points are all structural. The reopening after the weekend is the longest closure in the week for continuously traded instruments. The daily reopening after a session break matters most for index and commodity contracts that pause each day. Company results and dividend dates are published outside trading hours by convention. Scheduled macroeconomic releases and central bank decisions land at published times, and the seconds around them are the thinnest of the session.

All of those are on a calendar that is published in advance, which is the useful property of the list. The gaps that cannot be anticipated are a smaller category than the gaps that can, and the distinction is between an event whose timing is known and one whose direction and size are not. Neither is knowable in advance in the second sense, and no order type changes that.