kumocast

Kumo Research · Stock indices

Best day of the week to buy stocks: S&P 500, Nasdaq 100 and Dow Jones day by day (1953–2026)

Is Monday a bad day for stocks? Is Friday the best day to buy? We measured the return of every trading day of the S&P 500 since 1953, the Nasdaq 100 since 1985 and the Dow Jones since 1992, 37,719 index days in all, to see whether the day of the week really matters.

By · Published · 8 min read

Key findings

  1. Over the full history since 1953, Monday was the S&P 500's only losing weekday on average (-0.044% per day, higher on 49.6% of Mondays). Wednesday (+0.078%) and Friday (+0.061%) were the best.
  2. That famous Monday effect belongs to the past. Monday was negative on average in every decade from the 1950s to the 1980s, but since 1990 the differences between weekdays are too small to separate from chance. In the 2020s Monday has been the best day for the S&P 500 (+0.123%, up 60.9% of the time) and the Nasdaq 100 (+0.215%).
  3. For long-term buyers the day barely matters: investing $100 every week in the S&P 500 since 2000 gave a final value only 0.26% higher on the best weekday than on the worst. Time in the market mattered far more than the day.
  4. Monday is still the riskiest day: 11 of the S&P 500's 20 worst days since 1953 were Mondays, including the −20.5% crash of 19 October 1987 and −12.0% on 16 March 2020.
  5. Nasdaq 100: Wednesday has been the strongest day since 1985 (+0.162%, up 58.3% of the time) and Friday the weakest (+0.003%). For the Dow since 1992, no weekday stands out.

The question

"Never buy on a Monday." "Buy on Friday, sell on Monday." "Wednesday is the best day for stocks." Trading forums and investing guides are full of rules about the day of the week. Some come from real research: in the 1970s and 1980s, economists documented a Monday effect, or weekend effect, in which US stocks tended to fall on Mondays and rise on Fridays.

But markets change, and many patterns fade once people know about them. So we tested the question with the full daily history of the three main US stock indices:

  • S&P 500: every trading day from January 1953 to 09 November 2026 (18,588 days). We start in 1953 because the New York Stock Exchange also traded on Saturdays until 1952.
  • Nasdaq 100: every trading day from October 1985 (10,356 days).
  • Dow Jones Industrial Average: every trading day from January 1992 (8,775 days), the longest daily Dow history available to us from our data source.

For each day we measured the daily return: the closing price divided by the previous trading day's close. A Monday return therefore includes everything that happened over the weekend. These are price indices, so dividends are not included, and that does not change the comparison between weekdays.

The full history: Monday was the weak day

Average daily return by weekday

Full history: S&P 500 since 1953, Nasdaq 100 since 1985, Dow Jones since 1992, %

0.00%0.05%0.10%0.15%MondayTuesdayWednesdayThursdayFriday
  • S&P 500 (since 1953)
  • Nasdaq 100 (since 1985)
  • Dow Jones (since 1992)

Source: Yahoo Finance daily index closes · Chart: Kumo Research, kumocast.io

Daily returns by weekday: S&P 500, Nasdaq 100 and Dow Jones
Index and dayDaysAverageMedianUp daysStd. dev.Worst dayBest day
S&P 500 – Monday3,553-0.044%-0.001%49.6%1.14%-20.5% (19 Oct 1987)+11.6% (13 Oct 2008)
S&P 500 – Tuesday3,796+0.047%+0.020%51.2%0.99%-5.7% (07 Oct 2008)+10.8% (28 Oct 2008)
S&P 500 – Wednesday3,788+0.078%+0.082%55.6%0.97%-9.0% (15 Oct 2008)+9.5% (09 Apr 2025)
S&P 500 – Thursday3,737+0.030%+0.045%53.0%0.97%-9.5% (12 Mar 2020)+6.9% (13 Nov 2008)
S&P 500 – Friday3,714+0.061%+0.091%55.9%0.93%-6.8% (08 Jan 1988)+9.3% (13 Mar 2020)
Nasdaq 100 – Monday1,955+0.024%+0.117%54.5%1.71%-15.1% (19 Oct 1987)+12.6% (13 Oct 2008)
Nasdaq 100 – Tuesday2,125+0.070%+0.091%53.6%1.66%-9.1% (02 Jan 2001)+11.7% (05 Dec 2000)
Nasdaq 100 – Wednesday2,122+0.162%+0.180%58.3%1.63%-8.8% (15 Oct 2008)+18.8% (03 Jan 2001)
Nasdaq 100 – Thursday2,083+0.069%+0.128%54.7%1.63%-9.3% (12 Mar 2020)+10.8% (05 Apr 2001)
Nasdaq 100 – Friday2,071+0.003%+0.092%53.1%1.52%-9.7% (14 Apr 2000)+10.1% (13 Mar 2020)
Dow Jones – Monday1,652+0.072%+0.083%55.3%1.20%-12.9% (16 Mar 2020)+11.1% (13 Oct 2008)
Dow Jones – Tuesday1,802+0.062%+0.031%51.8%1.08%-5.1% (07 Oct 2008)+11.4% (24 Mar 2020)
Dow Jones – Wednesday1,800+0.033%+0.038%52.6%1.05%-7.9% (15 Oct 2008)+7.9% (09 Apr 2025)
Dow Jones – Thursday1,766+0.007%+0.046%52.4%1.10%-10.0% (12 Mar 2020)+6.7% (13 Nov 2008)
Dow Jones – Friday1,755+0.017%+0.080%54.9%1.03%-5.7% (14 Apr 2000)+9.4% (13 Mar 2020)

S&P 500 from 2 Jan 1953, Nasdaq 100 from 2 Oct 1985, Dow Jones from 3 Jan 1992, all to 09 Nov 2026. Price indices, close to close; Monday includes the weekend.

Over the whole history, the S&P 500's weekday pattern is clear. Monday was the only weekday with a negative average return (-0.044%) and the only one that ended higher less than half the time (49.6%). Wednesday (+0.078%, up 55.6% of days) and Friday (+0.061%, up 55.9%) were the best.

These daily numbers look tiny, but they add up. Had you held the S&P 500 only on Mondays since 1953 and stayed in cash on the other days, each $1 would have shrunk to about $0.17 by 2026, before dividends. Held only on Wednesdays, the same $1 would have grown to about $16.23. In the long run, the other four weekdays did all of the work.

The Nasdaq 100 has a different shape. Its best day by far was Wednesday (+0.162% on average, up 58.3% of the time), and its weakest was Friday (+0.003%). Monday was slightly positive. For the Dow since 1992, Monday was actually the best day (+0.072%) and Thursday the weakest (+0.007%), but none of the Dow's differences are large enough to rule out chance.

The Monday effect has faded

The S&P 500 Monday effect, decade by decade

Average daily return by weekday in each decade, % (2020s = Jan 2020 – Nov 2026)

-0.10%0.00%0.10%1950s1960s1970s1980s1990s2000s2010s2020s
  • Monday
  • Tuesday
  • Wednesday
  • Thursday
  • Friday

Source: Yahoo Finance ^GSPC daily closes, 1953–2026 · Chart: Kumo Research, kumocast.io

Splitting the S&P 500 by decade shows why the old rules no longer work. Monday's average return was negative in the 1950s, 1960s, 1970s and 1980s, and in the 1950s and 1960s only 41% and 40% of Mondays were up. Then it changed:

  • In the 1990s, Monday became the S&P 500's best day (+0.119%).
  • In the 2000s and 2010s, Monday was close to zero, like most days in those flat-to-modest decades.
  • In the 2020s (January 2020 to November 2026), Monday has again been the best day: +0.123% for the S&P 500 (up 60.9% of the time), +0.215% for the Nasdaq 100 (up 63.0%) and +0.100% for the Dow. Thursday has been the weakest day for all three.

Before and after: the S&P 500 weekday pattern

Average daily return by weekday, %

-0.10%0.00%0.10%MondayTuesdayWednesdayThursdayFriday
  • 1953–1989
  • 1990–2026
  • 2020–2026

Source: Yahoo Finance ^GSPC daily closes · Chart: Kumo Research, kumocast.io

A standard statistical test (analysis of variance) makes the same point. For 1953–1989 the S&P 500's weekday differences are far too large to be chance (p-value below 0.001). For 1990–2026 the p-value is 0.63: differences of this size would be common even if the weekday made no difference at all. The same test gives no significant weekday pattern for the Nasdaq 100 or the Dow since 2000.

Average daily return by weekday in different periods (share of up days in brackets)
Index and periodMondayTuesdayWednesdayThursdayFriday
S&P 500, since 1953-0.044% (50%)+0.047% (51%)+0.078% (56%)+0.030% (53%)+0.061% (56%)
S&P 500, 2000–2026+0.025% (54%)+0.060% (52%)+0.037% (54%)+0.029% (54%)+0.006% (54%)
S&P 500, 2020–2026+0.123% (61%)+0.070% (49%)+0.082% (55%)-0.032% (52%)+0.047% (54%)
Nasdaq 100, since 1985+0.024% (55%)+0.070% (54%)+0.162% (58%)+0.069% (55%)+0.003% (53%)
Nasdaq 100, 2000–2026+0.038% (56%)+0.064% (54%)+0.077% (56%)+0.077% (54%)-0.036% (52%)
Nasdaq 100, 2020–2026+0.215% (63%)+0.084% (52%)+0.127% (58%)-0.029% (53%)+0.011% (55%)
Dow Jones, since 1992+0.072% (55%)+0.062% (52%)+0.033% (53%)+0.007% (52%)+0.017% (55%)
Dow Jones, 2000–2026+0.042% (54%)+0.057% (52%)+0.028% (53%)+0.020% (54%)-0.001% (54%)
Dow Jones, 2020–2026+0.100% (59%)+0.048% (49%)+0.058% (53%)-0.038% (53%)+0.050% (56%)

Yahoo Finance daily closes. 2020–2026 = 2 January 2020 to 09 November 2026.

Does the day you buy matter for investors?

Buying $100 every week: the weekday hardly matters

Value on 09 Nov 2026 of each $100 invested weekly in the S&P 500 since January 2000, by buying day (price only)

  • Monday$444.32
  • Tuesday$443.16
  • Wednesday$443.88
  • Thursday$443.42
  • Friday$443.66

Source: Yahoo Finance ^GSPC daily closes; Kumo Research calculation · Chart: Kumo Research, kumocast.io

Suppose you invest a fixed $100 in the index every week at the close, always on the same weekday. Does the choice of day make a difference by 2026?

Hardly. From January 2000, buying the S&P 500 every Monday gave the best result: each $100 invested was worth $444.32 on 09 November 2026. The worst day, Tuesday, gave $443.16, just 0.26% less. For the Nasdaq 100 the gap between best and worst day was 0.50% and for the Dow 0.26%. Since 2016, Thursday was the best buying day for the S&P 500 and Monday the worst, and the gap was 0.66%.

Even over the full 74 years since 1953, when Monday really was weak, weekly Monday buying beat the worst day by only 2.35%. Over the same period, each $100 invested this way grew to between $5,172 and $5,294, before dividends. The decision to keep buying mattered thousands of times more than the day.

Weekly $100 purchases: value per $100 invested, by buying day
Index and startMondayTuesdayWednesdayThursdayFridayBest vs worst
S&P 500, since 1953$5,293.56$5,171.87$5,180.07$5,199.82$5,189.022.35%
S&P 500, since 2000$444.32$443.16$443.88$443.42$443.660.26%
S&P 500, since 2016$214.26$214.54$214.98$215.68$215.250.66%
Nasdaq 100, since 2000$1,002.01$998.20$998.99$997.00$999.600.50%
Dow Jones, since 2000$348.32$347.40$348.05$347.83$347.900.26%

$100 invested at the close every week on the chosen day, valued at the 09 November 2026 close. Price only, before dividends, costs and taxes.

The best day to trade stocks: Monday is the most volatile

For traders, the more useful question is not which day goes up, but which day moves. Here Monday stands out. Since 2000, the S&P 500's daily moves have been biggest on Monday (standard deviation 1.31%) and smallest on Friday (1.12%). The Dow shows the same pattern (1.26% vs 1.08%).

The biggest crashes cluster at the start of the week. Of the S&P 500's 20 worst days since 1953, 11 were Mondays: Black Monday (19 October 1987, −20.5%), 16 March 2020 (−12.0%) and 9 March 2020 (−7.6%) among them. News piles up over the weekend while the market is closed, and Monday is when it gets priced in. Our study on stock perps at the Monday open shows how 24/7 crypto markets now try to price that weekend news before Wall Street opens.

How often each day ends higher has also shifted. In the 2020s, the S&P 500 closed higher on 60.9% of Mondays but only 48.9% of Tuesdays, the only weekday below half. For the Nasdaq 100 the figures were 63.0% for Monday and 51.7% for Tuesday. Six or seven years is a short sample, so treat these as a description of the recent past, not a rule: the same test that found no reliable weekday pattern since 1990 finds none in the 2020s either.

After a bad Friday

S&P 500 Monday after a down or up Friday
PeriodAfter a down FridayAfter an up FridayAfter a Friday of −1% or worse
1953–1989-0.40%, up 33% (740 Mondays)+0.07%, up 53% (970 Mondays)-0.82%, up 31% (108 Mondays)
1990–2026+0.06%, up 57% (761 Mondays)+0.06%, up 54% (922 Mondays)-0.02%, up 56% (226 Mondays)
2000–2026+0.07%, up 57% (562 Mondays)-0.00%, up 52% (658 Mondays)+0.01%, up 57% (182 Mondays)

Average Monday return, share of Mondays that closed higher and number of cases. Only Mondays that directly follow a Friday session.

Another old rule says a down Friday leads to a down Monday. From 1953 to 1989 it did: after a losing Friday, the S&P 500 rose on the next Monday only 32.6% of the time, with an average of -0.40%. Since 2000 the pattern has reversed: after a losing Friday, Monday was up 57.1% of the time (+0.07% on average), more often than after a winning Friday (51.8%).

Overnight vs the trading day

S&P 500 weekend/overnight gap vs trading session, by weekday (2007–2026)
DayAverage gap (prev. close → open)Gap upAverage session (open → close)Session up
Monday+0.011%50.0%+0.012%53.6%
Tuesday+0.034%51.5%+0.048%51.9%
Wednesday+0.027%52.5%+0.022%54.0%
Thursday+0.000%50.6%+0.011%54.7%
Friday+0.014%52.6%+0.020%55.8%

Yahoo Finance ^GSPC daily open and close, 3 January 2007 – 09 November 2026. Monday's gap covers the weekend.

Since 2007 we can split each day into the gap (from the previous close to the open, which on Monday includes the weekend) and the session (from open to close). On Mondays the average weekend gap was +0.011% and the Monday session +0.012%, both close to zero. Tuesday had the strongest session (+0.048%). None of these differences is big enough to trade on after costs.

What this means

  1. There is no reliably best day of the week to buy stocks today. The Monday effect was real from the 1950s to the 1980s, but it has faded since 1990 and even reversed in the 2020s.
  2. For long-term investors the day does not matter. The gap between the best and the worst buying day was a fraction of a percent over 26 years.
  3. For traders, Monday matters most because it is the most volatile day, with weekend news priced in at the open and most of history's worst crashes.
  4. Past weekday patterns are not a strategy. Any edge of a few hundredths of a percent a day disappears after spreads, fees and taxes.

You can follow the current trend, momentum and key levels on our S&P 500 forecast, Nasdaq 100 forecast and Dow Jones forecast pages. For how the three indices compare over the long run, see our study of Nasdaq 100 vs S&P 500 vs Dow. Kumocast provides free market data and analysis, not trading signals or financial advice.

Method, data and limits

Data: daily closing prices of the S&P 500 (^GSPC), Nasdaq 100 (^NDX) and Dow Jones Industrial Average (^DJI) from the Yahoo Finance chart history. S&P 500: 2 January 1953 – 09 November 2026; Nasdaq 100: 2 October 1985 – 09 November 2026; Dow: 3 January 1992 – 09 November 2026 (the Dow's daily history at this source starts in 1992). The Nasdaq 100 series was cross-checked against the official series on FRED (NASDAQ100).

Measures: daily return = close ÷ previous trading day's close − 1, grouped by the weekday of the close. Monday returns include the weekend. "Up" = a positive return. Statistical tests: one-way analysis of variance across the five weekdays and Welch t-tests for each weekday against the rest. Weekly buying test = $100 invested at the close on the chosen weekday of every week (weeks where that day was a holiday are skipped), valued at the last close; the result is shown per $100 invested so that different numbers of holidays do not matter. Gap = open ÷ previous close − 1; session = close ÷ open − 1, S&P 500 only, from 2007 because earlier opening prices in this source are often just the previous close.

Limits: price indices only (no dividends); no trading costs or taxes. Small average differences can come and go by chance, so a pattern in one decade says little about the next. Not investment advice.

How to cite

Kumo Research (2026). "Best day of the week to buy stocks: S&P 500, Nasdaq 100 and Dow Jones day by day (1953–2026)". Kumocast. https://www.kumocast.io/research/best-day-of-the-week-to-buy-stocks

This study describes past market data. It is not investment advice and past behaviour does not guarantee future results.

Live data for these markets

More from Kumo Research

Alex Wu, Independent market analyst

About the researcher · Crypto, DeFi, perps and FX · driven by data · methodology