Which Open Do You Use in a Market That Never Closes
Ask what time the currency market opens and the honest answer is that it does not, in the sense that matters. Trading passes from region to region without interruption from Sunday evening until Friday close. There is no bell, no auction, no moment when everyone arrives together. An opening range rule needs a starting point regardless, and since the market will not supply one, the trader has to.
The Usual Candidates
Several starting points are in common use. The Tokyo open marks the beginning of meaningful Asian activity. The London open is the largest single transition in the currency day, bringing the biggest dealing centre online. The New York open adds the second largest. The daily rollover, the point at which one trading day is administratively replaced by the next, is used by some purely because it is unambiguous.
These are not equivalent, and choosing between them is not arbitrary. Each anchors the range to a different set of participants, and the resulting levels carry the character of whoever was trading while they formed.
What Makes a Substitute Open Work
The value of an exchange open is not the clock time. It is that a large number of participants act within a short window, which produces both a genuine discovery process and levels that many people are watching. A substitute open is useful to the extent that it reproduces those two things.
By that standard the London open is the strongest candidate for most major pairs, because the change in participation is abrupt and large. The rollover is the weakest, because although it is precise, nothing happens there. It marks an accounting boundary, and a range formed around an accounting boundary is a range formed while very little was going on.
The Pair Changes the Answer
A range anchored to London means something different depending on what is being traded. For pairs where European currencies are involved, the London session is where the primary interest sits and the open is a genuine event. For pairs dominated by Asian participation, London is a secondary session and its open is less consequential than the Tokyo one.
Commodity linked currencies follow their own pattern again, with meaningful activity tied to their domestic hours and to the times when the underlying commodities trade actively. Applying one anchor across every pair on the screen imposes a single market structure on markets that do not share one, and the pairs where it fits worst will look as though the rule is failing when the anchor is what is wrong.
The Clocks Do Not Stay Aligned
Session opens are local events, and the regions change their clocks on different dates. For several weeks each year the gap between London and New York differs from the rest of the year, and Asian centres that do not observe seasonal changes shift relative to both.
A rule written in a fixed clock time will therefore point at the wrong moment for part of the year, and the failure is quiet. It does not produce an error, it produces a range formed some way off the intended transition, which is enough to change what the range contains. Defining the anchor relative to the local session rather than to a fixed hour in your own time zone avoids the whole problem.
Pick One and Leave It Alone
The temptation with several plausible anchors is to use whichever one produced the better looking range on the day. That is not a strategy, it is a choice made after seeing the outcome, and it will produce a record that cannot be evaluated.
The workable approach is to select an anchor per pair on structural grounds, write it down, and hold it long enough to generate a real sample. If a comparison between anchors is wanted, it has to run across the same sessions with both defined in advance. Anything else is choosing the range after the fact, which will always look excellent and will never be repeatable.