Holding a Trade Over the Weekend Is a Decision You Are Already Making

Every calendar-first routine has a blind spot, and it is the largest scheduled event of the week. Friday's close is on every calendar in the world. It happens at a known time, it happens every week, and it is the one event where the standard advice — know what is coming, decide in advance what you will do — is almost never applied. Positions are simply carried into it, because carrying them is what happens if you do nothing.

That is the whole point of this piece. Holding over a weekend is not a neutral act you can drift into. It is a decision with a specific risk profile, and if you did not make it deliberately on Friday, you made it by default — which is the same decision with none of the preparation.

The market does not close, only your access does

This is the sentence worth sitting with. When the forex and metals desks shut on Friday evening, the world does not pause. Elections are held. Central bankers speak at weekend conferences. Wars start and stop. Data is released in economies whose weekends do not line up with the one your broker observes. Sanctions are announced on Saturdays specifically because markets are shut.

None of that stops mattering. It accumulates. And on Sunday evening the market does not resume where it left off — it opens at whatever price now reflects everything that arrived while nobody could act. The gap is not a glitch or an unfairness. It is the market's first opportunity to answer a question that has been building for two days.

So the honest way to describe an open position on a Friday night is this: you are holding a directional bet on two days of news you have not read yet, with no ability to change your mind. That framing sounds dramatic until you notice it is simply a description of the mechanics.

Why your stop is not the protection you think it is

Most traders size a position off the stop. Risk per trade is the distance to the stop multiplied by the size, and that arithmetic is the foundation of nearly every risk framework worth using. It is also the arithmetic that quietly stops being true over a weekend, and this is the single most expensive misunderstanding in the whole subject.

A stop is an instruction to exit at a price if the market trades there. It is not a guarantee that you will get that price — it is a trigger. During normal trading that distinction rarely matters, because price moves through levels continuously enough that the trigger and the fill are close together. Over a gap, price does not move through anything. It disappears at one level on Friday and reappears somewhere else on Sunday.

If it reappears past your stop, the stop has not been jumped or ignored. It has been triggered correctly, and then filled at the first price actually available — which may be well beyond the level you chose. The order worked exactly as designed. What failed was the assumption underneath your position size: that the distance to the stop was the size of your loss.

The practical consequence is simple and unwelcome. On a weekend hold, the number you calculated as your risk is a floor, not a ceiling. Everything else in this piece follows from that.

The Sunday open is the thinnest market of the week

There is a second mechanism stacked on top of the first, and it is the one that turns a manageable gap into an unnecessarily expensive one.

When trading resumes, it resumes thin. The first prints are made by a small number of participants, most of the usual liquidity has not arrived, and spreads are wider than at any other point in the week — reliably wider than around scheduled releases, which is saying something. If you have read the piece on what the spread does around a release, this is the same phenomenon with a longer fuse and a bigger footprint.

That matters in three ways. A stop triggered into the open is filled into that spread. A position you decide to close in a panic on Sunday evening is closed into it too. And any level sitting close to Friday's close can be touched by nothing more than the spread widening around it, without the market having gone anywhere at all.

So the risk of a weekend hold is not only the size of the gap. It is the gap plus the cost of transacting in the worst liquidity of the week, at the exact moment you are most likely to want to transact.

Measure your own open — nobody else's numbers apply

You will find plenty of confident figures about typical weekend gaps. Ignore them, including any you might expect here. Brokers open at different times, price from different liquidity pools, and apply different spread policies to the first minutes of the week. The only version of this that can inform a decision is the one you measured on the account you actually trade.

It takes a few seconds each week and one column of a spreadsheet. For each weekend, record:

• The instrument, and the last price on Friday before your platform stops updating.
• The exact time your platform's price starts moving again on Sunday — brokers differ by more than you would guess, and knowing yours to the minute is worth having.
• The first price you see, and the difference from Friday's last.
• The spread at that first print, and the spread at your instrument's normal weekday baseline.
• How long until the spread comes back to that baseline — five minutes, twenty, an hour.
• Anything that happened over the weekend which plausibly explains the move, or a blank if nothing did.

After six or eight weekends you have something no article can give you: your own broker's behaviour, in the instruments you trade, including the quiet weekends. The quiet ones matter as much as the dramatic ones, because they are what set your expectations the rest of the time.

One warning about how to read that log. The number to plan around is not the average gap — it is the largest one you have recorded, and even that is only the largest so far. An average weekend costs nothing and teaches nothing. The distribution here has a long tail, and the tail is the entire subject.

The Friday decision, written down

The point of all this is not to hold nothing over a weekend. That rule has its own cost — it forces you out of trades that were working, on a schedule that has nothing to do with your thesis, and it will make some of your best trades into small ones. The point is that the decision gets made on purpose, before the close, with three questions answered.

First: does this trade need the weekend? Some do. A position built on a multi-week move is not improved by being closed and reopened every Friday at a cost. A short-term trade that has not resolved by Friday afternoon is usually a different question — it is often a trade whose premise has already expired, being held because closing it means booking something.

Second: can I survive the worst open I have on file, at this size? Not the average one. Take the largest gap in your log, apply it to the current position in the wrong direction, and look at the number. If that number is one you could not accept, the position is too big to hold through the close — which is a size decision, not a reason to abandon the trade. Reducing to a size that survives the tail is the middle option most people skip.

Third: is there anything scheduled? This is the one that is genuinely checkable and genuinely neglected. Weekend elections and referendums, OPEC meetings, central bank symposiums where officials speak on a Saturday, political deadlines that fall on a Sunday, scheduled talks between governments. None of that is secret. It sits on the same calendars you already check for Tuesday's data, and Friday afternoon is the moment to look at the weekend rows rather than skipping past them.

Three things not to do

Do not widen the stop before the close. It feels protective and it is the opposite: it does not reduce the chance of a gap through your level, it only increases what you lose if one happens. If the position is uncomfortable, the lever is size, not distance.

Do not assume a guaranteed stop makes the problem disappear without reading what it costs and when it applies. Products of that kind do exist and they do genuinely cap the fill, which is a real service. They also carry a premium, and the terms usually reserve the right to withdraw or reprice the guarantee in exactly the conditions you bought it for. Read the specific terms on your account rather than the marketing sentence.

And do not act in the first minutes of the open unless the position genuinely cannot wait. If the gap has already happened, it has already happened — closing into the widest spread of the week, in the thinnest liquidity of the week, at the moment of maximum emotional pressure, is a decision that has to earn its place rather than being the reflex.

Common questions

How big is a typical weekend gap in gold?

Refusing to put a number on it, and the refusal is the answer rather than a dodge. Any figure would be an average across brokers, instruments and periods, and an average is the wrong statistic for a risk you only care about in the tail. Most weekends the gap is small enough to be indistinguishable from an ordinary opening wobble. The ones that matter are rare and are driven by something specific that happened, which is why the useful question is never 'how big is typical' but 'what is the largest my own account has shown me, and could I take that at this size'.

Should I just close everything on Friday?

It is a legitimate rule and plenty of profitable traders use it, but it is not free and it should not be adopted because it sounds prudent. You pay the round-trip cost every week, you exit trades on a timetable unrelated to why you entered them, and you give up the moves that happen over the weekend in your favour — which are as common as the ones against you. Adopt it if your strategy is genuinely intraday, or if you have found from your own log that you do not sleep otherwise. Do not adopt it as a substitute for sizing.

My broker's Sunday open always looks fine. Is this overblown?

That experience is normal and it is exactly what makes the risk hard to respect. The overwhelming majority of weekends pass without incident, which trains everyone to treat the close as a non-event, which is why the rare weekend finds so many people carrying more than they would have chosen. The reasonable response is not alarm — it is to keep the log so that when a bad open does arrive you already know what your account does in one, and to keep the size at a level where a quiet weekend and a loud one are both survivable.

Does this apply to crypto, since it never closes?

The gap mechanism largely does not, because there is no scheduled halt to gap across. The underlying problem does, in a different shape: the market moves at hours when you are not watching, and a large move that a forex trader meets as a Sunday gap, a crypto trader meets on waking up. The stop-versus-fill distinction still applies during fast moves and thin books, and it applies at the same size. The difference worth naming is that in forex you know precisely when the exposure window opens and closes, so it can be planned for; in crypto it is continuous, so the plan has to be in the position size rather than in the calendar.

What about holidays and other long closes?

The same mechanism with less warning. Public holidays produce shortened sessions, thin books either side, and occasionally multi-day closes in a specific market while related markets keep trading — which is the situation most likely to produce a large reopening move. They also arrive irregularly enough that nobody has a routine for them. Add the market-holiday rows to the same Friday check, and treat the day before a long close the way you treat a Friday.

If I hold, should I take partial profit before the close?

This is the practical version of the sizing answer, and it is often the right shape. Reducing to a size that survives your worst logged open lets a trade keep running without the weekend being able to do real damage, and it removes most of the pressure to act badly on Sunday evening. What is worth avoiding is deciding the amount in the last minutes on Friday, when the reasoning is dominated by the current open profit rather than by the risk. Decide the rule once, write it down, and apply it the same way whether the trade is up or down — which is the same standard the journal piece asks for, applied to a different decision.