Seasonality in Forex, Patterns Worth Observing, Not Worshipping
Some traders notice that certain months, weeks, or times of day show recurring tendencies. This is called seasonality, and while it can offer clues, it should be treated cautiously.
Examples of observed patterns:
- Time of day: volatility often peaks during London and New York sessions
- Day of week: Mondays may be slower, midweek more active, Fridays prone to profit-taking
- Month-end flows: rebalancing by large funds can move currencies
- Quarter-end and year-end: liquidity changes and portfolio adjustments
- Holiday periods: thin liquidity can produce erratic moves
- Seasonal demand: some currencies see recurring flows related to tourism or trade cycles
How to use it responsibly:
1. Collect your own data. Review several years on your pair, if possible, and note recurring behaviors.
2. Test patterns with objective rules.
3. Use seasonality as a minor filter, not a trigger.
4. Combine with structure, liquidity, and risk management.
Dangers:
- Small samples and cherry-picked examples mislead
- Patterns fade when many traders trade them
- Market conditions change
Treat seasonality as a hypothesis to test, not a promise.
If a pattern can't survive a clear backtest, don't risk real money on it.
Curiosity is healthy. Blind belief isn't.
If you test one seasonal idea this month, record the date, the result, and the conditions, then compare it to the next occurrence before trusting it.
Seasonality in Forex, Patterns Worth Observing, Not Worshipping
Re: Seasonality in Forex, Patterns Worth Observing, Not Worshipping
I'd add a warning about small samples. A pattern that showed up in five of the last seven Decembers sounds convincing, but seven data points can't separate a real tendency from chance. With dozens of months and weekdays to look at, some will always look special just by luck. If you test seasonal ideas, write the rule down first and then check it on years you haven't looked at yet. If it only works in the years where you found it, that's a sign it was noise. Use seasonality to lean slightly, for example to be a bit more careful in late December when liquidity thins out, never as the main reason for a trade.
It’s Fairman 
Re: Seasonality in Forex, Patterns Worth Observing, Not Worshipping
One that's often mentioned is month end, when rebalancing flows can push the dollar around in the final days. I've noticed odd moves near the London fix at month end, but it's not consistent enough for me to trade it. I just stay a bit more careful. The fix is at 4pm London, so 4pm Lagos during BST and 5pm Lagos from late October to March. In the 15 minutes or so around it, big orders get executed and pairs like EURUSD and GBPUSD can jump 20 or 30 pips and then drift back. If I'm in a trade into that window on the last day of the month, I expect extra noise.
It’s Fairman 
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: Seasonality in Forex, Patterns Worth Observing, Not Worshipping
Fairman's small sample warning is the right first filter. Five out of seven Decembers looks neat until you remember how many month and weekday slices you scanned to find it.
Month end around the London fix is the seasonal idea I watch without trading. In the fifteen minutes around 16:00 London, EURUSD and GBPUSD can jump twenty or thirty pips and then drift back, which is a reason to flatten early rather than a reason to enter.
Write the rule before you look at history. If the idea only works on the years you already stared at, treat it as noise. Hold out two or three recent years as a blind check.
Seasonality is a lean, not a signal. A mild bullish December bias might let you keep a runner one day longer, not open a fresh position into a clear lower timeframe short.
Fairman, on the fix moves you have seen near month end, do you cut size for the whole final day or only for the last hour? The last hour is where most of the noise sits.
Month end around the London fix is the seasonal idea I watch without trading. In the fifteen minutes around 16:00 London, EURUSD and GBPUSD can jump twenty or thirty pips and then drift back, which is a reason to flatten early rather than a reason to enter.
Write the rule before you look at history. If the idea only works on the years you already stared at, treat it as noise. Hold out two or three recent years as a blind check.
Seasonality is a lean, not a signal. A mild bullish December bias might let you keep a runner one day longer, not open a fresh position into a clear lower timeframe short.
Fairman, on the fix moves you have seen near month end, do you cut size for the whole final day or only for the last hour? The last hour is where most of the noise sits.
-
FXS Prop Desk
- Posts: 1689
- Joined: Wed Oct 07, 2026 9:52 am
Re: Seasonality in Forex, Patterns Worth Observing, Not Worshipping
One reason to be cautious is that seasonal patterns found by testing many months or days are subject to the multiple comparisons problem. With twelve months and dozens of pairs, some patterns will appear strong by chance alone. Checking that a pattern holds in separate periods and has a plausible explanation, such as fiscal year-end flows, makes it more credible.
FXS Official Team | The Idea Lab & Strategy
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: Seasonality in Forex, Patterns Worth Observing, Not Worshipping
FXS Prop Desk's multiple comparisons point can be put in numbers. Test 12 calendar months on 10 pairs and that is 120 separate checks, so at the usual 5 percent bar about six of them should look significant by luck alone.
That is why the plausible story matters as much as the statistic. Japanese fiscal year end in March gives yen flows a reason to exist, while a strong pattern in AUDCHF every August usually has no reason beyond the sample.
Time of day patterns carry a quieter trap. If the study uses broker server time, the London and New York opens move by an hour twice a year, and a pattern that seems to fade in spring is sometimes just a clock shift.
Measuring everything in London or New York local time removes that problem. It also makes a short window like the fix comparable across all twelve months.
A useful acceptance rule is that a seasonal idea has to survive both a held out period and a cost check. A December lean worth 3 pips on average means little if the thin holiday spread takes 2 of them.
That is why the plausible story matters as much as the statistic. Japanese fiscal year end in March gives yen flows a reason to exist, while a strong pattern in AUDCHF every August usually has no reason beyond the sample.
Time of day patterns carry a quieter trap. If the study uses broker server time, the London and New York opens move by an hour twice a year, and a pattern that seems to fade in spring is sometimes just a clock shift.
Measuring everything in London or New York local time removes that problem. It also makes a short window like the fix comparable across all twelve months.
A useful acceptance rule is that a seasonal idea has to survive both a held out period and a cost check. A December lean worth 3 pips on average means little if the thin holiday spread takes 2 of them.
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: Seasonality in Forex, Patterns Worth Observing, Not Worshipping
Statistical studies of currency returns reveal mild but persistent patterns by hour of day. Certain currencies tend to strengthen or weaken in specific windows, often linked to the business hours of the home market and to the flows of corporate and institutional participants.
For example, research has suggested tendencies for the dollar to behave differently during U.S. trading hours than during European hours. Such patterns are small, vary by period, and can disappear when too many participants act on them.
Testing seasonality requires large samples and honest treatment of costs. A pattern worth ten percent of a pip per trade vanishes after a half-pip spread. Scalpers interested in the subject should compute average return by hour in UTC across several years, separate by pair, and test stability across subperiods. A pattern that holds in only one year is likely noise. Even without a tradable edge, knowing how a pair tends to behave by hour helps set realistic expectations for volatility and direction during specific sessions.
For example, research has suggested tendencies for the dollar to behave differently during U.S. trading hours than during European hours. Such patterns are small, vary by period, and can disappear when too many participants act on them.
Testing seasonality requires large samples and honest treatment of costs. A pattern worth ten percent of a pip per trade vanishes after a half-pip spread. Scalpers interested in the subject should compute average return by hour in UTC across several years, separate by pair, and test stability across subperiods. A pattern that holds in only one year is likely noise. Even without a tradable edge, knowing how a pair tends to behave by hour helps set realistic expectations for volatility and direction during specific sessions.