Should We Trust Seasonality? — 2026-09-01
Core thesis: September's historical reputation as the worst month (SPX avg. -1.31%, NDX avg. -1.84%) masks a more nuanced reality: actual month-to-month outcomes are nearly random, and recent performance contradicts the pattern, making seasonality an unreliable trading guide.
Key points:
- Historical data is skewed by outliers: Only 3–4 calendar months show negative average returns since 2000; September's losses are >1%, while other down months barely register negative.
- Actual outcomes defy the pattern: SPX closed lower in only 5 of the past 10 Septembers; NDX in 6 of 10—"about as close to random as we can get."
- Recent trend contradicts seasonality: Last two Septembers were solidly positive (SPX +2.02%, +3.53%; NDX +2.48%, +5.40%), breaking a brutal 2020–2023 streak.
- Current conditions favor fundamentals over seasonality: Rising yields (10-year Treasuries approaching 5%), Persian Gulf tensions (tanker attacks, US strikes on IRGC), suppressed pre-Labor Day volumes, and Friday's jobs report loom larger than calendar patterns.
- This morning's dip is modest: Early September weakness is not predictive; a few hours of trading don't set monthly tone.
Takeaway: Traders should weigh which odds they're playing—historical September weakness, near-random recent outcomes, or the two-year winning streak—but ultimately focus on fundamentals and geopolitical risk rather than relying on seasonality alone.