Last August, a trader messaged us. His EA had been printing money: +8% monthly for five straight months. Then June hit -6%, July -4%. He was convinced the strategy had broken.
We looked at the dates and said: your EA isn't broken. It's summer.
The forex market doesn't run uniformly all year. It has high seasons and dead zones, explosive months and months where the price barely moves. Running an EA with the same position size and parameters year-round is like wearing the same outfit through all four seasons. Eventually the weather catches you.
Why markets have seasons
Three structural forces drive forex seasonality.
Institutional capital flows. The world's largest banks and funds rebalance portfolios at quarter ends. The last week of March, June, September, and December produces surges in volume and directional movement.
Holiday calendars. Christmas, Thanksgiving, Japanese Golden Week, European summer vacations. When major market participants leave, liquidity drops, spreads widen, and price becomes erratic. The technical signals your EA relies on degrade to noise.
Fiscal year cycles. US tax filing in April, Japan's fiscal year ending in March, European fund performance reviews in December. Each triggers large position adjustments that look nothing like normal market behavior.
The same EA can earn 15% in September and lose 5% in July. Not because the strategy broke. Because the market changed shape.
Month by month breakdown
January: cautious start. First week has leftover holiday thin liquidity. From week two, institutions start positioning for the new year. Volatility medium-high. EURUSD 10-year average monthly range: ~370 pips. Run at 70–80% normal position. Watch and wait week one, normal operation from week two.
February: business as usual. Continuation of January's trends. No major holiday disruptions. Volatility moderate (~380 pips). Normal position. Nothing special to adjust.
March: quarter-end gold rush. Q1 closes. End-of-quarter rebalancing drives large flows. Japan's fiscal year ends in March, making yen pairs especially volatile. EURUSD average range: 420–450 pips. Trend EAs can run 110–120% position. Watch the last 3 days closely — quarter-end flows can be extreme.
April: tax season confusion. US tax filing deadline mid-April creates unpredictable dollar flows. False breakouts and whipsaw are common. EURUSD range contracts to ~340 pips. Reduce to 60–70%. Mean reversion strategies can maintain normal size. Trend EAs should tread carefully.
May: the summer slowdown begins. "Sell in May and go away" has statistical backing. Both equity and currency volatility decline. EURUSD range: ~330 pips. Reduce to 70%. If the first half is acceptable, start scaling down in the second half.
June: transition into the dead zone. First two weeks are tolerable. From mid-June, European traders begin summer vacations. Liquidity noticeably drops. Spreads widen. EURUSD range: ~310 pips. Reduce to 50%. Widen stop losses by 10–15 pips to handle wider spreads. Three consecutive losing days? Pause.
July: worst month of the year. Northern hemisphere summer vacations are in full swing. London and New York trading desks are understaffed. Liquidity hits its annual low. Price movement is directionless noise. EURUSD range: ~290 pips, mostly choppy. Pause or reduce to 30%. Many professional traders take July off. Your EA should too.
August: still dead, with one exception. Similar to July, possibly worse. All of Europe is on holiday. But late August brings the Jackson Hole symposium (usually last week of August), which can produce sudden big moves. Maintain pause or 30% for the first three weeks. Can briefly restore to 50% for Jackson Hole week with tight stops.
September: best month of the year. Everyone returns from vacation. Institutions start positioning for Q4. Capital flows normalize. Volatility surges, trends are clear and sustained. EURUSD 10-year average September range: 450+ pips, highest of any month. Run at 100–120%. September is your primary profit month. Make sure your EA has updated parameters and fresh backtest data before the first week.
October: high volatility, strong trends. Continues September's momentum. Many major market events in history happened in October: 2008 financial crisis, 2022 GBP crash. Volatility remains high with clear directional movement. EURUSD range: 420–460 pips. Maintain 100–120%. Consider widening take profit targets — autumn trends run long.
November: sprint then stop. First three weeks are strong. After US Thanksgiving (fourth Thursday), the market effectively shuts down for a week. 100% position before Thanksgiving. Starting Wednesday of Thanksgiving week, reduce to 30% or pause. Restart the following Monday.
December: two-phase month. First two weeks, fund managers make final year-end adjustments — decent trading. Around December 20, the market enters "Christmas mode": liquidity plummets, spreads widen dramatically, price jumps without logic. 70% position until December 15. 30% after the 15th. Full shutdown after the 20th. Restart second week of January.
Quick reference table
| Month | Volatility | EURUSD avg range | Recommended position | Note |
|---|---|---|---|---|
| Jan | Medium-high | 370 | 70–80% | Cautious start |
| Feb | Medium | 380 | 100% | Normal |
| Mar | High | 435 | 110–120% | Quarter-end opportunities |
| Apr | Medium-low | 340 | 60–70% | Tax season caution |
| May | Medium-low | 330 | 70% | Begin scaling back |
| Jun | Low | 310 | 50% | Defensive mode |
| Jul | Very low | 290 | Pause/30% | Rest |
| Aug | Very low | 285 | Pause/30% | Rest (except Jackson Hole) |
| Sep | Very high | 455 | 100–120% | Best month — full aggression |
| Oct | High | 440 | 100–120% | Trend following paradise |
| Nov | High→low | 400 | 100%→30% | Sprint then stop at Thanksgiving |
| Dec | Medium→dead | 340 | 70%→shutdown | Exit by Dec 20 |
Seasonal parameter adjustments
Reducing position size alone isn't enough. The EA's parameters should shift with the seasons.
Summer (June–August): reduce lots to 30–50%. Widen stop losses by 15–20 pips because spreads are wider and tight stops get triggered by noise. Lower take profit targets because sustained trends are rare. If your EA supports time filtering, consider disabling the Asian session entirely — summer Asian liquidity is especially thin. The most aggressive option: pause for two months. Historical backtest data shows that most trend following EAs have negative expectancy in July and August.
Autumn (September–mid-November): lots back to 100–120%. Stop losses can be tightened by 5–10 pips because liquidity is good and slippage is minimal. Increase take profit targets by 20–30%. Autumn trends are long and smooth. If your EA has position-adding logic, autumn is the best time to enable it.
Year-end (mid-December–early January): gradual reduction starting December 15. Lots drop to 50%, then 30%, then shutdown. The market between Christmas and New Year's lacks any predictable behavior. We've watched traders give back an entire year's profit in the last week of December because of a flash crash during zero-liquidity conditions.
Key dates every EA trader must track
Monthly: NFP (first Friday). Pause EA 30 minutes before, resume 60 minutes after.
Central bank rate decisions: Fed FOMC (~8/year), ECB (~8/year), BOJ (~8/year), BOE (~8/year). Two hours of elevated volatility around each decision.
Major holidays (reduce or pause around these):
- New Year's (Jan 1)
- Easter (March/April, EU/US markets closed)
- Japan Golden Week (late April–early May)
- US Memorial Day (last Monday May)
- US Independence Day (July 4)
- Jackson Hole (late August)
- US Labor Day (first Monday September)
- US Thanksgiving (fourth Thursday November)
- Christmas (Dec 24–26)
- New Year's Eve (Dec 31)
Liquidity drops significantly one day before and after each holiday. Mark them all on your trading calendar.
What the data says
Ten years of EURUSD data (2015–2024): September and October average the highest monthly ranges at roughly 455 and 440 pips respectively. July and August are the lowest at about 290 and 285 pips.
Autumn volatility is more than 1.5x summer volatility. For EAs that profit from directional movement, that translates directly to 1.5x the income opportunity.
More importantly, autumn volatility is directional. Moves last days or weeks. Summer volatility is noise: up 30 pips today, down 35 tomorrow, up 20 the day after. The problem for trend EAs in summer isn't less profit. It's actual losses from repeated false signals.
FAQ
My EA uses a grid/mean reversion strategy. Is summer better for it?
In theory, range strategies like quiet markets. In practice, summer's problem isn't just low volatility — it's low liquidity. Low liquidity means wider spreads and worse fills on every order. Grid strategies with frequent entries get eaten alive by execution costs. And summer occasionally produces liquidity-driven flash crashes with zero warning. Grid and Martingale EAs are especially vulnerable to these. Even range strategies should reduce in summer.
If I increase positions in September-October, what about black swans?
Increased position doesn't mean abandoned risk management. Keep per-trade risk at 1–2%. September and October have higher volatility, which means larger absolute swings. But statistically, the positive expectancy in these months is significantly higher than any other period. Don't skip the best months because you're afraid of one bad day.
Are seasonal patterns reliable every year?
The broad pattern is stable: summer slow, autumn strong, Christmas dead. This has held for 20+ years because it's driven by structural factors (institutional calendar, holiday schedules) that don't change. Specific months can deviate — March 2020 was extreme due to COVID, September 2022 was unusual due to the UK pension crisis. Treat seasonal patterns as the default setting, then adjust when exceptional events override them.
Which month should a beginner start testing?
September. High volatility, clear trends, good liquidity. Your EA gets the fairest chance to show its true performance. Starting in July and watching the EA lose money might make you abandon a perfectly good strategy that just needs a better market environment. Read our $100 challenge article for a practical testing framework.
Is there a simple rule I can just remember?
Yes: Full aggression September through mid-November. Rest or light position June through August. Normal everything else. This alone avoids most of the year's worst trading conditions. Every EA in the FXTool marketplace includes seasonal performance notes so you can verify these patterns against each strategy's specific data.
About the author: The FXTool team builds and tests MetaTrader trading tools daily. We run every EA we sell on live accounts and publish the results. This guide reflects what we've learned from building 50+ EAs and working with thousands of retail traders.
Forex trading involves significant risk and may result in total loss of capital. This article is for educational purposes only and is not investment advice. Understand the risks and consider your financial situation before trading.