The question is usually asked as though there is a right day and a wrong day. There sometimes is, but you cannot know which is which in advance, and for most amounts the difference is smaller than the cost of the counter you walk into. It is worth separating the parts of the question that have answers from the part that does not.
First, size the thing you are worrying about
Before anything else, put a number on the risk. Take the pair's high and low over the past year and apply both ends to your amount.
Suppose a pair traded between 1.13 and 1.20 over a year — roughly a 6% span. On £500 that is about £30 between the best and worst day of an entire year, and you will not hit either. On £50,000 it is £3,000, and the question becomes serious.
That single calculation resolves most cases. If the spread of outcomes across a whole year is smaller than the effort of thinking about it, stop thinking about it and change the money.
What timing cannot do
Picking the direction. Not because it is hard, but because tomorrow's rate moves on information that does not exist yet — and everything that is known is already in today's price. That argument is set out in full in can tomorrow's rate be known, and nothing on this page contradicts it.
The practical form of the warning: waiting for a better rate is a position. Choosing not to convert today is a bet that the rate improves, taken with the same information as everyone else and usually with a deadline attached.
What timing can do, in a small way
Three timing effects are real, and all of them are about liquidity rather than direction.
Time of day. Spreads are tightest when major markets overlap — the London afternoon, which coincides with the New York morning, is the deepest few hours of the trading day. Quotes late on a Friday or during the Asian session on a European pair are, at the margin, worse. For an interbank trade this is worth something. At a bureau counter with a rate board that changes twice a day, it is worth nothing.
Around scheduled events. Rate decisions, inflation releases and major political votes produce the fastest moves of the year. If you have a large conversion and no reason to do it in that particular hour, do it in a different hour. This is risk avoidance, not prediction.
Weekends and holidays. Markets are shut, so rates are stale and providers widen their margins to cover the gap until reopening. A large transfer initiated on a Saturday is usually converted on worse terms than the same transfer on Tuesday.
What helps far more than timing
For the great majority of ordinary conversions, these four decisions move more money than the date:
- Where you change it. The gap between a competitive provider and an airport counter routinely exceeds the entire annual range of the pair. This is the biggest lever there is.
- Refusing conversion at the terminal. When a card machine abroad offers to charge you in your home currency, saying yes costs several percent in one keystroke — see dynamic currency conversion.
- Asking for the final amount, not the rate. A better headline rate with a commission attached often loses to a worse one without. Along with the rest of the common mistakes, this is where the money actually goes.
- Not converting twice. If you will need the currency again, holding it beats converting out and back, which pays the spread a second time for nothing.
Splitting, and what it is for
If an amount is large and the date is flexible, converting in two or three parts over a few weeks is a reasonable habit. It will not beat the market — on average it lands on the average — and that is exactly the point. It removes the possibility of having put everything through on the single worst day, at the cost of removing the best day too.
Splitting is insurance against regret, not a strategy for gain. Sold as the latter it is nonsense; used as the former it is sensible.
The short version
For anything up to a few hundred, the date is noise. Change it where it is convenient and check you are not being charged twice.
For anything large, the date is worth a little care and the venue is worth a lot. Look at the past year's movement to know where today sits, avoid the hours around a central bank announcement, split it if the deadline allows, and spend the saved effort on comparing final amounts rather than rates.
Last updated:
Related articles
Volatility, and why a range beats a forecast
Volatility is a measure of how far a rate moves, not where it goes. How to read an annual high and low, and what it tells you that no prediction can.
Fixed, floating and managed — what each regime means for you
A currency's regime decides whether its rate is a market outcome or a policy statement, and that changes what is worth checking before you change money.
Eight mistakes when exchanging money, and what each one costs
Accepting conversion at the terminal, changing money at the airport, comparing the wrong column — the ordinary errors that quietly add up to a few percent.
Frequently asked questions
How an exchange rate is actually set, what a spread costs on a real transaction, what happens when your income and your debt are in different currencies, and what to check before paying or being paid abroad. They explain mechanisms rather than forecast rates, and each one carries the date it was last revised at the top.
No. We publish no forecasts, price targets or trading calls, and no article recommends buying or selling a currency. Exchange rates over short horizons are close to unpredictable, and a site that pretends otherwise is selling something. What the articles do instead is explain the mechanism, so that you can read the news for yourself.
They are written by the kursdanas.rs editorial team. We accept no commissioned or sponsored articles, and advertising has no influence on what is published or on the order of anything in our tables. Where an article states a figure it names the source, and where we get something wrong we correct it and change the revision date shown on the page.