Tuesday, September 9, 2025

September Seasonality of US Stock Indices | Jeff Hirsch

With Fed scheduled to make an announcement on September 17, the 12th trading day, it appears the market is pulling typical mid-month gains forward.

 Average September Market Performance 204-2024.
 
Once the market gets the interest rate cut it expects, it still may not be enough to avoid historical end-of-Q3 weakness.
 
 
 
 
2025 S&P 500 Equal Weight Cycles Composite (One-Year Seasonal Cycle, 
Four-Year Presidential Cycle, 10-Year Decennial Cycle, 1928-2024), 
Ned Davis Research, August 8, 2025.
 
September in post-election years exhibits a bearish trend with an overall average decline of -1.0% since 1950 and -0.93% from 1928-2024, driven by a -0.36% average in the first ten days and a steeper -1.13% in the last ten, reflecting policy uncertainty and late-month weakness. 
 
 
The market often starts with Day 1 showing a bearish tilt, down ~58% of the time since 2008 with an average decline of -0.3% to -0.5%, influenced by low post-Labor Day volume.

Days 2–5 display mixed performance with a slight downward bias due to portfolio rebalancing and "window dressing" by fund managers, while mid-September (Days 6–15) sees amplified losses, with Day 6 at -0.17%, Day 7 at -0.22%, Day 10 at -0.26%, and Day 15 (quadruple witching) at -0.25%, marked by heightened volatility averaging -0.48% in post-election years. 
 
S&P 500 average performance per day and daily percentage hit rate (1928-2024).

This mid-month weakness is tied to market adjustments and quadruple witching dynamics, contributing to a cumulative bearish shift.
 
S&P 500 seasonality first ten sessions and lst ten sessions of the month since 1928.

Late September (Days 16–20) offers a modest 0.2% bounce, though inconsistent and often fading into choppy trading by Days 25–30, which remain neutral to slightly bearish due to end-of-quarter portfolio adjustments. Hit rates drop below 50% mid-to-late month, and a 4.2% standard deviation in early September peaks mid-month, underscoring volatility.