There is a moment in the year when the statistics suggest paying closer attention. In 2026, that moment may be very close.

Not because a date can predict what Wall Street will do, and not because there is any law that says markets must fall after mid-August. The point is different: several cyclical patterns that are normally studied separately happen to overlap this year, right between the second half of August and the fall.

August seasonality, the decennial cycle, the US presidential cycle and the historical behavior of September that follows all point to the same window of time — one that deserves attention.

The question, though, is not only about the coming weeks. If a correction were to begin, could it simply be another pullback within an uptrend, or the start of something far more significant? And, above all, how would the reading of a possible decline change when you move from a horizon of a few months to one of ten or twenty years?

The core of the analysis is to distinguish between two time horizons. In the short and medium term, the second half of August and the September–October 2026 period concentrate several cyclical and seasonal recurrences that have historically been delicate. Over the long term, however, the statistics tell a different story: even the deepest bear phases have so far been temporary segments within much broader cycles, each followed by new bull phases. It is precisely the contrast between these two horizons that makes the current market phase so interesting.

Why the second half of August 2026 could be different

Let’s start with the United States, because Wall Street remains the main reference point for international equity markets. A historical study of the Dow Jones compared the US index with the FTSE, the Italian market, the DAX and the CAC 40. Despite differing absolute performances, the major international exchanges have shown a high correlation with US market moves over time. A similar dynamic emerges when comparing the S&P 500 and the Nasdaq.

This does not mean that every exchange rises or falls at the same time and to the same degree. It does mean that a major directional move that starts on Wall Street tends to be reflected in the other main markets as well. That is why what could happen in the United States over the coming weeks takes on international relevance.

Taken as a whole, August does not stand out historically as a particularly negative month. In the Dow Jones study, using data from 1929 through August 2007, August posted an average return of +0.74% and was positive in 60.76% of cases, versus 39.24% of negative readings. The standard deviation of 6.13%, however, signals a very wide dispersion of results.

The monthly figure, though, tells only part of the story. When August is broken down by individual trading day, a much more interesting dynamic emerges. After a weak initial phase, the average curve tends to recover in the middle of the month and reaches an area of relative strength around the 18th trading day. Then the behavior changes.

In the final trading days of August, the historical curve tends to deteriorate, giving back part of the earlier move. The average sequence can therefore be summarized as:

  • initial weakness
  • recovery
  • an area of strength in the second half of the month
  • deterioration toward the end of August

And it is precisely the last part of the sequence that takes on special meaning in 2026. This year, in fact, simultaneously occupies two very specific positions within the major historical cycles: it is the sixth year of the decennial cycle and the second year of the US presidential cycle. The combination is labeled 6–2.

When a year ending in 6 coincides with the second year of a presidential term, the historical curve tends to lose strength in the second part of the year and, on average, locates a low between September and October. A further study within the analyzed data had specifically identified September–October for 2026 as the likely window for the annual low, following a phase of weakness expected in the middle of the year.

The overlap of these different cyclical patterns is hard to ignore. That temporal overlap is exactly what makes this phase of 2026 so noteworthy.

If a decline began, how long could it last?

Right after August comes one of the statistically most difficult months of the year. In the historical sample considered, September produced an average return of -1.48% and was positive in only 38.46% of cases, versus 61.54% of negative readings. The behavior within the month is also telling: from the sixth trading day onward, average historical performance tends to turn negative.

This creates a rather particular sequence:

  • second half of August — possible loss of strength
  • end of August — deterioration of the usual seasonal curve
  • September — a historically unfavorable month
  • September–October — the area where the 6–2 cycle tends to reach its low

None of these statistics prove that markets must fall. What they do is identify a window of time in which the possible appearance of technical signs of weakness would carry more meaning than in other periods of the year. At that point, a second question would inevitably arise: if a pullback began, would it be just a correction, or could it mark the start of a long bear market? In theory, both possibilities remain open.

Even the great bear markets began with a first drop that, in its early stages, could be read as a normal correction. It would therefore be methodologically wrong to rule out that a possible downward move could develop over a much longer period. Long-run statistics, however, introduce a different element.

In the historical material analyzed, it is observed that, looking at a full decade, in nearly 80% of cases bear phases were concentrated in the equivalent of roughly three years. In other words, declines can be deep, violent and sometimes prolonged, but historically they have occupied a minority portion of the great equity cycles.

The year 2020 offers an extreme example. The Dow Jones lost 38.4% in less than a month, in one of the fastest drops in its history. Yet from the March 2020 low to January 2022, the index went from 18,213.65 to 36,925.65, more than doubling. Its previous highs had already been recovered by November 2020. This does not mean, of course, that every future correction must last the same amount of time or be followed by an equally rapid recovery. The statistical point is different: a decline, even a very significant one, and a permanent change in the long-term trajectory of the markets are not the same thing.

Is it time to get out of the market? What 10 and 20 years of history tell us

And here we reach probably the most important question. If a negative phase were really to begin between the second half of August and the fall, how would its interpretation change for someone watching the markets over a multi-year horizon? Here it is important not to conflate two completely different dimensions.

A statistic concerning August, September or October may be relevant for studying market behavior over the coming months. It says far less about what might happen over ten or twenty years. In the historical material analyzed, in fact, a significant figure emerges: regardless of the entry point, holding an equity investment for a period of between 7 and 10 years, the historical frequency of a positive result is put at around 80%.

Extending the horizon to 15–20 years, in the same sample the percentage exceeds 95%. These percentages are naturally no guarantee for the future. A historical sample describes what has happened, not what will necessarily happen. But they serve to put the initial question in the right perspective.

Historical statistics show that a short- or medium-term correction and the performance of an equity investment over multi-year horizons are two phenomena that must be analyzed separately. And this holds even in the most severe scenario. If any move that started between August and September were to turn into a longer bear phase, the history of equity markets still shows an alternation between bear markets and subsequent bull phases, with the latter having on average lasted longer.

The point, then, is not to decide today whether or not it is «time to get out of the market». Such a conclusion would require individual considerations about goals, risk and time horizon that lie entirely outside a general statistical analysis.

The interesting issue is another one: how does the reading of a decline change when the horizon of observation shifts from a few months to ten or twenty years? In the short term, the 2026 combination deserves attention — the second half of August, seasonal deterioration toward month-end, a historically weak September, and the 6–2 cycle that tends to seek a low between September and October.

Over the long term, though, the statistics tell a different story: the great bear phases have historically been segments — even very painful ones — within longer cycles from which new bull phases subsequently emerged.

And perhaps this is the real thing to watch in the coming weeks. Not only whether Wall Street will start to fall, but whether a possible decline will be deep and persistent enough to genuinely alter the long-term structure of international markets. Until that happens, an unfavorable cyclical window of a few weeks or months and a structural trend of many years remain two profoundly different phenomena.

Editor’s note

This article was originally published in Italian on money.it by Gerardo Marciano on August 12, 2026 as «Wall Street, agosto accende un segnale d’allarme: è solo una correzione o l’inizio di un lungo ribasso?». It has been translated and adapted for an international audience by the Money.it International desk.