The Weeks the CME Clock Slips in Sweden

There is a Monday in March, and another one at the end of October, when a Swedish trader opens the platform at the usual time and finds that the American market has been running for an hour without him. No email warned you. Nothing on the chart is broken. The Stockholm session is exactly where it always is, the Swedish clock on the wall is correct, and yet the S&P and Nasdaq futures have already printed their opening range and moved on.

What happened is that Sweden and the United States do not change their clocks on the same weekend, and for a few weeks a year the gap between Stockholm and Chicago is six hours instead of seven. Every session boundary that you think of in Swedish time arrives one hour earlier. It is completely predictable, it has been happening for a quarter of a century, and it still catches people every spring.

What actually changed that Monday

Nothing at the exchange. This is the part worth getting straight before anything else, because the instinct is to go looking for a fault in the data feed or the broker.

CME defines every one of its sessions in Chicago time. The equity index cash open at 08:30, the daily settlement window at 15:00, the close at 16:00 and the one hour maintenance break that follows it, the Sunday evening reopen at 17:00: all of that is written in Central Time and none of it moves, ever. The full grid sits on the CME Group trading hours page and it reads the same in February as it does in April.

Swedish time does not move either, in the sense that matters. Sweden runs on Central European Time, UTC+1 in winter and UTC+2 in summer, and the national time scale is maintained and distributed by RISE, which holds the riksmätplats för tid och frekvens, the national laboratory for time and frequency. Its job is to distribute svensk normaltid, and it does it accurately, including through the old public service that still answers when you dial it, Fröken Ur.

Both clocks are right. What changes is the distance between them, and it changes because the two legal calendars that govern the switch were written independently and land on different Sundays.

The two legal calendars

In the European Union the rule is Directive 2000/84/EC. Summer time starts on the last Sunday in March and ends on the last Sunday in October. The directive fixes the moment at 01:00 UTC, which is why every member state jumps at the same instant rather than each at its own two in the morning. For Sweden that means the change happens at 02:00 in winter and 03:00 in summer local terms, in step with Germany, Spain and Finland.

In the United States the rule descends from the Uniform Time Act as amended in 2005. Daylight saving time starts on the second Sunday in March and ends on the first Sunday in November, at 02:00 local time in each zone. The dates and the legislative history are published by NIST.

The Americans therefore go forward first, in March, and come back last, in November. Sweden goes forward later and comes back earlier. In between, on both ends, there is a hole. In that hole the offset is six hours instead of seven and the whole American trading day sits one notch earlier on your Swedish clock.

The Stockholm clock against the Chicago clock

Here is the practical version. Left column is Chicago, the reference that never moves. Middle column is what your Swedish clock shows for roughly forty nine weeks of the year. Right column is what it shows during the mismatch.

Chicago time	Normal Swedish time	Slip weeks
Sunday reopen, 17:00	00:00 Monday	23:00 Sunday
US macro release, 07:30	14:30	13:30
Equity index cash open, 08:30	15:30	14:30
Crude oil pit open, 08:00	15:00	14:00
FOMC statement, 13:00	20:00	19:00
Equity index settlement, 15:00	22:00	21:00
Daily close, 16:00	23:00	22:00
Maintenance break ends, 17:00	00:00	23:00

Read the middle column once and you understand why the normal state feels so natural to a Swede: the American cash open at half past three and the week starting at exactly midnight are easy numbers to live with. Read the right column and you understand why the slip weeks feel wrong in a way that is hard to name. Midnight stops being the boundary. The week now starts on Sunday evening.

The dates you need through 2028

Two windows per year. The spring one opens when the US moves and closes when Sweden moves. The autumn one opens when Sweden moves back and closes when the US moves back.

Year	US forward	Sweden forward	Spring slip	Sweden back	US back	Autumn slip
2026	Sun 8 March	Sun 29 March	9 to 27 March, three weeks	Sun 25 October	Sun 1 November	26 to 30 October, one week
2027	Sun 14 March	Sun 28 March	15 to 26 March, two weeks	Sun 31 October	Sun 7 November	1 to 5 November, one week
2028	Sun 12 March	Sun 26 March	13 to 24 March, two weeks	Sun 29 October	Sun 5 November	30 October to 3 November, one week

Windows are given as trading weeks, Monday to Friday. If you trade the Sunday evening reopen, the shift begins on the Sunday itself: on 8 March 2026 the new week starts at 23:00 Swedish time, not at midnight.

The arithmetic, so you can work out any year yourself

You do not need to look up a table for 2031. The pattern is fixed by the two rules.

The autumn window is always one week exactly. Sweden goes back on the last Sunday in October and the United States on the first Sunday in November, and the first Sunday in November is always the Sunday immediately after the last Sunday in October. Five trading days, no exceptions, every year.

The spring window is either two weeks or three, and it depends on how many Sundays March happens to contain. The US moves on the second Sunday; Sweden on the last. When March holds four Sundays, the last one is the fourth and sits fourteen days after the second: two weeks. When March holds five Sundays, which happens whenever 1 March falls on a Friday, Saturday or Sunday, the last one sits twenty one days after the second: three weeks. In 2026, 1 March is a Sunday, so that year gets the long version, three full weeks of a 14:30 open right in the middle of the Swedish spring.

Totalled up, three or four weeks a year run on a different local schedule. Call it seven per cent of your trading days, which is more sessions than every Swedish public holiday put together.

What the slip does to a Swedish trading day

The mechanics are the same everywhere in Central European Time. Where they bite is specific to how the day is arranged here.

The Stockholm cash market runs until half past five in the afternoon, later than most of the continent. In the normal state the American open at 15:30 arrives with roughly two hours of Swedish session still to go, and the second half of the Stockholm afternoon is largely a reaction to what New York decided to do. During the slip weeks the American open lands at 14:30, three hours before the Swedish close, which lengthens the overlap and changes the character of the afternoon: more of the Stockholm session is spent inside US hours, and the correlation you are used to seeing late in the day starts earlier.

Then there is the ordinary Swedish working day. If you are trading around a job, the normal 15:30 open is manageable because it sits near the end of it. A 14:30 open does not. Three weeks of that in 2026, in March, a month with a heavy macro calendar. It is worth deciding in advance whether you will trade those weeks at all or simply sit them out, and that decision is better made in February than at 14:31 on a Monday.

Light matters here more than it does further south. In late March, Stockholm is gaining daylight at close to six minutes a day, and the spring slip sits right in the middle of that climb, so the afternoon you trade in late March looks nothing like the one you traded in January. The autumn window is the opposite and worse: Sweden has just gone back to normaltid, the sun is setting in the late afternoon, and the American open at 14:30 is one of the last hours of real daylight you get. By the time the cash session closes at 21:00 it has been dark for hours. Some people handle that fine. Others discover in their first November that trading a foreign session in a Swedish winter is a different job from trading it in June.

The part nobody notices: what it does to a backtest

Missing an open costs you one morning. A time filter that quietly means something different for three weeks a year costs you an entire research conclusion, and you will never see it on the chart.

Strategies, indicators and alerts in NinjaTrader evaluate their time conditions against the time zone the chart is drawn in. So if your chart is in Swedish time and your rule says no entries before 15:30, then during every slip week that rule blocks the first hour of the American session, which is the hour with by far the most volume and the widest range. Nothing errors. The strategy simply sits out the best part of fifteen or twenty sessions a year and you attribute the result to the market.

The mirror image is just as bad. A rule that says flatten everything by 22:00 Swedish time normally flattens you an hour before the Globex close; during the slip weeks it flattens you an hour after the daily settlement, in the thin tape that follows it, at prices that have nothing to do with the ones your backtest assumed.

Over a multi year sample this is a small, systematic and almost invisible distortion: roughly seven per cent of your sessions ran under different rules from the other ninety three, for reasons that have nothing to do with price. Anchor the chart and the Trading Hours template to Chicago and the filter means exactly the same thing in every week of the sample, which is the only condition under which the backtest is telling you about the market rather than about the calendar.

Run the platform on Chicago time

The durable fix is not a better memory. It is to stop thinking about the American session in Swedish time at all.

NinjaTrader 8 has two settings that people mix up, and they do different jobs.

The display time zone lives under Tools, Options, General. It decides what the timestamps on your charts, grids and order windows say. Put it on Central Time and the cash open is at 08:30 in March, in July and in November, permanently, with nothing to adjust twice a year. Nothing is lost when you change it: the platform stores timestamps in UTC internally and converts them for display, so your historical data is untouched and you can switch back whenever you want.

A few things around it. Windows itself has to be on the correct zone with automatic summer time adjustment enabled, because the platform converts from the system clock. Pick one time zone for your economic calendar and stay with it: half in Chicago time and half in Swedish time is how people arrive an hour early to a number. And if you keep a second clock visible, make it Chicago rather than New York, because CME session and settlement times are quoted in Central Time and the hour between the two American zones is its own separate trap.

The five minute routine, twice a year

• Enter the four change dates for the next three years into your calendar now, from the table above, with an alert on the Friday before each.
• On that Friday, write the new local open time on a note and stick it to the monitor where you cannot avoid seeing it.
• Shift alarms, routines and any scheduled task an hour earlier for the window, and remember to shift them back.
• Check the Trading Hours template on every chart you trade, not just the first one you open.
• Assume the first affected Monday is worse than usual for a while after the open, partly because a good share of European participants are making this exact mistake at the same moment.

It is fair to ask how long any of this will last. The European Union has spent years debating whether to abolish seasonal clock changes, a discussion that has been particularly loud in the Nordic countries, and the American Congress has repeatedly considered making its summer time permanent. Neither has changed the law. Until one of them does, the drift is a permanent feature of trading CME products from Sweden, and the cheapest insurance against it is a platform clock that has never heard of Stockholm.

us market open time change sweden

Educational content, not investment advice. Trading futures involves substantial risk of loss and is not suitable for every investor. Any reference to Algomyth results is hypothetical, simulated over historical data, and shown in full on the performance page; past performance does not guarantee future results.