Average twenty years of EURUSD returns by month and you will get a bar chart with tall bars and short ones. It looks like information. Split the same twenty years into two ten year halves and the tall bars usually move. That instability is the first thing to understand about seasonality: with only one observation per year, twenty years of data is twenty data points, and twenty data points will produce a striking pattern by chance alone.
This does not make calendar effects worthless. Some have an identifiable cause, a set of participants who must act at a certain time for reasons unrelated to price. Those are worth knowing. The rest are decoration.
Effects with a mechanism
The clearest calendar effects in FX operate on a monthly rather than annual cycle. Month end brings portfolio rebalancing: funds holding foreign assets adjust their currency hedges to match the new value of those assets, which produces predictable FX flow into the late London afternoon. Quarter end amplifies it, and year end more so. This is why the last trading day of a month can produce a sharp move in the final hour that has nothing to do with any news.
Rollover and swap mechanics create a second, smaller calendar structure. Positions held across Wednesday typically carry triple swap to cover the weekend value date, which changes the cost of holding certain pairs on that day. The mechanics are set out in swap rates, and for carry sensitive positions the day of the week genuinely changes the arithmetic.
Liquidity seasonality is the third real one and probably the most useful. Volumes thin out in late December and, to a lesser degree, in August when European desks are on holiday. Thin markets are not calm markets. Ranges can compress for days and then move violently on order flow that would have been absorbed in March, and spreads on the less liquid crosses sit wider all month. The concept is covered in liquidity in forex, and it changes position sizing more than it changes direction.
Gold and the story problem
Gold attracts more seasonal folklore than any other instrument: wedding season demand in India, Chinese New Year buying, a supposed summer lull. Physical demand is real and it does have a calendar. Whether it moves the price of a metal whose daily turnover is dominated by futures, ETFs and central bank activity is a different question, and the honest answer is that jewellery flows are a small share of the total.
What actually moves gold is closer to macro than to calendar: real yields, the dollar, and demand for a hedge when risk appetite disappears. Those drivers are explained in the gold trading guide and in safe haven flows. If a seasonal chart and the macro picture disagree, the macro picture is the one carrying the money.
Seasonal averages hide their own dispersion. A month with an average gain can contain several years of heavy losses offset by one enormous year. Always look at how many of the individual years agreed with the average before treating it as a tendency, and never size a position on a seasonal chart alone.
Why most seasonal studies break
Three problems recur. The first is the sample size already mentioned. The second is regime change: the euro has existed since 1999, and the monetary environment of 2005, 2015 and 2025 differs enough that combining them into one average blends three different markets. The third is survivorship in the presentation. Nobody publishes the chart of the month with no pattern, so the seasonal studies that circulate are pre selected for looking convincing.
There is also a practical trap in how seasonality is measured. A study based on calendar month closes cannot be traded, because you never know the month close until it has happened. Any seasonal rule has to specify an entry date, an exit date, a stop and a size before it means anything, and once you add a stop most seasonal edges shrink considerably.
Using calendar effects properly
The realistic use is context, not signal. Knowing that late August tends to be thin tells you to expect wider spreads and to reduce size, which is a risk decision rather than a directional one. Knowing that the last London hour of a month can jump tells you not to leave a tight stop sitting under the market at 16:00 on the 31st. Knowing that the days around a major holiday produce erratic quotes tells you to be flat.
None of that requires you to believe the market goes up in November. It requires you to know when the participants change, in the same way that session structure tells you when they change within a day. Seasonality is that same idea stretched across a year, and it is at its most reliable when it is describing liquidity rather than direction.
Testing a seasonal idea without fooling yourself
If you want to check a calendar effect, three rules keep the exercise honest. State the mechanism first and the test second, because a pattern found by scanning every date will always turn up something. Split the history into halves and require the effect to appear in both. And test it with a real stop and real costs, since the ability to hold through a large adverse move is the assumption doing most of the work in a seasonal backtest.
The general discipline is the same as for any historical study, and the traps are the same ones described in backtesting basics: too many parameters, too little data, and results that depend entirely on the sample chosen. Seasonality simply makes those traps easier to fall into, because the data set is small by construction and the story always sounds plausible.
"If you cannot name the people whose behaviour creates the seasonal pattern, you have found a coincidence in a small sample, not a pattern."
— Alex Onta, Executive Director, SINGUARD
Key Takeaways
- One observation per year means most seasonal charts rest on a couple of dozen data points, which is far too few to be confident about a pattern.
- Month end, quarter end and year end rebalancing flows have a real mechanism and a real time of day, unlike most annual calendar effects.
- Thin liquidity in late December and August is the most useful seasonal fact, and it changes position size rather than direction.
- Any seasonal claim you intend to trade needs a stated mechanism, an entry and exit date, a stop and consistent behaviour in both halves of the history.
Frequently Asked Questions
Is there a reliable best month to trade forex?
No month has a dependable directional edge that survives being split across different decades and different monetary regimes. What does repeat is the liquidity pattern: quieter markets around late December and mid summer, busier ones from September through November.
Does gold really rise in Indian wedding season?
Physical jewellery demand in India does follow a calendar, but the gold price is set largely by futures, ETF flows, central bank buying, the dollar and real yields. Physical demand is one input among several and rarely the dominant one in any given week.
What is the month end fix and why does it matter?
Funds holding foreign assets rebalance their currency hedges at month end, generating FX orders concentrated into the late London afternoon of the final trading day. It can produce a sharp move with no news attached, which is worth knowing if you hold positions with tight stops into that hour.
About the Author
Alex Onta is an Executive Director at SINGUARD. He built eTrader, the terminal, the mobile apps, eTrader Broker, Copytrading, Business and Community, along with the worldwide clustered-server infrastructure it all runs on, with his brother Roman Onta helping on the design, and he leads that division today. Together with Roman he builds the Prop Firm CRM, the Broker CRM, Scalegram and CopySignals, and the two of them carry worldwide compliance, payment processing and international business structuring side by side. He lives and works in Dubai for most of the year. Meet the executive duo leading Singuard's five divisions.