InsightDeck / Seasonality / Any month / Pre-holiday effect

Pre-holiday effect

The saying

The trading day before a market holiday is said to be good for stocks.

About this saying

What it says

A saying about the trading day before a market holiday, which comes round several times a year.

Why people say it (explanations offered)

Less selling before a break and a lighter mood are among the reasons given. None of them has been established.

What to keep in mindThe comparison is with all trading days. Days before closures that were not holidays (funerals, storms, paperwork) are left out.

Checking the data

Kenneth R. French Data Library (based on CRSP data). Whole US stock market (value-weighted), dividends included, US dollars. Window: Trading day before a weekday market holiday. 915 observations, 1926–2026 (data through 2026-09-30). Values are total returns over the window, dividends included.

626 / 915Times up (1926–2026): 68%, 95% range 65〜71%
55%Comparison: share up in all trading days
153 / 242Times up since 2000 (63%); comparison over the same years 54%
+0.27%Average; median +0.26%; comparison average +0.04%

Return on the day before a holiday

One bar per year (that year's average). Navy: average up; gray: down. Shaded from 2000

InsightDeck chart
-1%0%1%2%2000–1930194019501960197019801990200020102020

Source: Kenneth R. French Data Library (based on CRSP data). Whole US stock market (value-weighted), dividends included, US dollars

By period: share of observations up

Navy: the saying's window. Thin gray: the comparison. Vertical line: 50%

InsightDeck chart
50%1926–194969%(172/248)1950–199971%(301/425)2000 on63%(153/242)All years68%(626/915)
Pre-holiday effectall trading days

Comparison: all trading days (26338 windows)

Before and after the cited study or almanac

Before and after Robert A. Ariel, High Stock Returns before Holidays: Existence and Evidence on Possible Causes, Journal of Finance 45(5), 1611–1626 (1990). Thin gray: the comparison over the same years

InsightDeck chart
50%Before 199073%(431/593)1990 on61%(195/322)
Pre-holiday effectall trading days

Average: before 1990 +0.35% (comparison +0.04%); 1990 on +0.13% (comparison +0.05%). Many anomalies are said to weaken once they become known.

Year-by-year reviews

Reviews will be written one at a time from 2026 on. Each February, once December data is out, the previous year is reviewed together. The data appears one to two months after a window ends.

Past patterns may not continue. Test enough months and windows and some will look reliable by chance alone. This is not meant for making trading decisions.