One month breaks the pattern
When a single calendar month averages −2.0 % over 38 years and closes positive in only 39 % of years, that is no longer random noise. That month is September for the DAX — the worst DAX month in the index's official history since 1988. We analysed every calendar month since the DAX launched. The result is strikingly clear.
What "worst month" means here
We measure the average monthly return and the win rate (share of years with a positive month) for each of the twelve calendar months. The data is the DAX performance index (^GDAXI) since trading began in 1988 — 38 complete annual cycles.
Important: seasonality describes recurring patterns of the past, not a guarantee for the future. It shows when the market historically had a tailwind or headwind — not what happens tomorrow.
Key findings at a glance
- Weakest month: September — avg −2.0 %, only 39 % of years positive
- Second weakest: August — avg −1.9 %
- Strongest month: April — avg +2.7 %, 72 % positive
- Highest win rate: December — 74 % of years positive
- August and September are the only two months with a negative average
The DAX yearly pattern in numbers
The chart below shows the average return per calendar month over the last 38 years — the current month is highlighted.
The full monthly table makes the pattern tangible:
| Month | Avg return | Win rate |
|---|---|---|
| January | +0.8 % | 59 % |
| February | +1.1 % | 56 % |
| March | +0.4 % | 51 % |
| April | +2.7 % | 72 % |
| May | +1.0 % | 64 % |
| June | −0.0 % | 49 % |
| July | +1.6 % | 66 % |
| August | −1.9 % | 47 % |
| September | −2.0 % | 39 % |
| October | +2.0 % | 68 % |
| November | +2.6 % | 66 % |
| December | +2.2 % | 74 % |
Two phases stand out: a late-summer weakness (August + September) and year-end strength (October to December). In between, April is the single bright spot of the spring.
How reliable is the September effect?
A low average alone is not enough — what matters is consistency across years. The heatmap shows every individual September return since 1988. Notably, September closed negative in 7 of the last 10 years. Whether the September weakness is also statistically significant over the maximum period (t-test, p-value, effect size) is covered in detail in The DAX in September: the significance test explained.
What investors can read into it — and what not
Seasonality is a probability context, not a signal. Three takeaways:
- For long-term investors, the late-summer dip matters mostly psychologically: it explains why the market often feels sluggish in August/September — a reason to stay calm rather than sell nervously.
- For active traders, the asymmetry is interesting: months with win rates of 39 % versus 74 % carry very different risk-reward profiles.
- Important: any single year can fall completely outside the pattern. The −2 % is an average of strong up years and deep down years.
Methodology & transparency
We use normalized returns based on adjusted DAX closing prices since 1988 — not absolute point levels. How we verify data and compute seasonality is laid out openly on our methodology page. The full risk disclosure is in the legal section.
Conclusion
September is the statistically weakest DAX month since 1988 — just ahead of August. Against that stands a strong April and a reliable fourth quarter. These patterns are not a roadmap, but valuable context for your own sense of timing. Explore the interactive monthly cycle for any ticker yourself at seasonalpha.ai.
Frequently asked questions
Which is the worst month for the DAX?
September: since 1988 the DAX lost about 2 % on average in this month and closed positive in only 39 % of years — the weakest of all twelve months.
Which month is historically the best?
April, with an average of around +2.7 % and a 72 % win rate. November and December also rank among the strongest months with high consistency.
Is the September effect unique to the DAX?
No, late-summer weakness appears across many equity markets. The exact shape differs by index, though — which is why it pays to look at the actual data rather than rules of thumb.
Can I rely on this seasonality?
No. Seasonality describes averages over many years. Individual years can deviate sharply, and past patterns do not guarantee future results.