Best Time to Buy Stocks? 4 Days a Month That Carried the Market

The four-day window

In the research, the turn of the month is four trading days:

• the last trading day of the month (day −1);
• the first three trading days of the next month (days +1 to +3).

Everything else, about 16 trading days a month, is the "rest of the month".

The evidence

Richard Ariel was one of the first to document a monthly pattern, in A monthly effect in stock returns (Journal of Financial Economics, 1987): U.S. stock returns were bunched up around the start of each month. A year later, Josef Lakonishok and Seymour Smidt looked at a much longer record in Are seasonal anomalies real? A ninety-year perspective (Review of Financial Studies, 1988). Studying the Dow Jones Industrial Average from 1897 to 1986, they found unusually high returns around the turn of the month, along with other calendar effects, and used the four-day window later research picked up.

The most cited modern study is Equity Returns at the Turn of the Month by John J. McConnell and Wei Xu, published in the Financial Analysts Journal in March 2008. Using daily U.S. returns from 1926 to 2005, they reported:

• Turn of the month (4 days)0.15%
• Other 16 trading days−0.001%

Exact values from table 2 in this article. Bar lengths share a zero baseline; no new calculation or simulated price path. Read the table and its stated sample and limits in this article.

Their conclusion, as stated in the article's abstract, is pretty striking: over those eight decades, investors on average got no reward for taking market risk except at turns of the month. They also found that the effect:

• wasn't limited to small or low-priced stocks;
• wasn't limited to year-ends or quarter-ends;
• showed up in 31 of the 35 countries they looked at;
• wasn't explained by month-end buying pressure, measured by trading volume and flows into equity mutual funds.

They concluded that the pattern "remains a puzzle in search of an answer".

In index futures

If you trade ES or NQ, the most relevant follow-up looks at futures. In a November 2011 working paper, Calendar Anomalies in Stock Index Futures, Óscar Carchano and Ángel Pardo tested 188 possible calendar rules on S&P 500, DAX and Nikkei futures from December 1991 to April 2008. As summarized by CXO Advisory, a short version of the turn of the month, holding S&P 500 futures from the last close of the month to the close of the first trading day, was the only calendar effect still profitable after assumed trading costs over the whole sample. It was also one of only three that were statistically significant in both halves of it.

That's notable. Most calendar effects, like the weekend effect, weakened after they were published. This one has been more stubborn, though the futures sample is short and 188 rules were tested.

Possible explanations

None is proven, and McConnell and Xu tested and threw out some of them.

• Payment flows. Wages, pensions, dividends and interest get paid around month-end. If part of that cash goes into stocks at the start of the month, it could push prices up. James Ogden proposed this payment-date explanation in the Journal of Finance in 1990. Later tests haven't settled it.
• Pension and fund contributions. A lot of retirement plans invest contributions on a fixed monthly schedule. Practitioners often point to these flows, but McConnell and Xu found that measured fund flows didn't explain the effect.
• Window dressing and rebalancing. Fund managers may adjust portfolios before month-end reporting and rebalance at the start of the month.

So is the turn of the month the best time to buy stocks?

For an investor, only in a narrow sense. Four days at 0.15% add up to about 0.6% a month. That matters over decades but it's small next to the cost of being out of the market. The research measures when returns showed up, not a reason to wait for a date. For a trader, it's a documented tilt in the calendar, not a timing system.

What this means if you trade ES or NQ

It's a tilt, not a trade signal. An average of 0.15% a day over four days adds up over decades, but in any one month the noise is many times bigger, and a turn of the month can still be sharply negative.

Put it in points. With the S&P 500 near 7,800 in October 2026, 0.15% is about 12 index points. A normal ES session moves several times that, which is why the effect only shows up across many months.

Use it as context for the first days of a month. If your daily bias is bearish on the first trading day of a month, you're leaning against a documented historical tailwind. That's not a reason to change your read, but it is a reason to be clear about why you're bearish.

Read it alongside the calendar. The first trading days of a month often bring big data, like the ISM manufacturing survey and, usually on the first Friday, the jobs report. Those releases can easily outweigh any calendar tendency.

Check it on your own window. The research uses full trading days. Whether the effect shows up in the 9:30–12:00 ET morning window, or overnight, is a separate question that needs its own test.

How to test it yourself

1. Export at least ten years of daily closes for SPY, QQQ or your futures contract.
2. Label the last trading day and the first three trading days of each month.
3. Compare their average daily return and share of up days with all other days.
4. Split the sample by period, for example before and after 2010.
5. Subtract realistic costs for one round trip a month.

We plan to publish this test on the Nasdaq-100 once we've added long daily history, with the method above locked in ahead of time.

How it compares with other calendar effects

Limits

• We didn't run our own daily test. This article sums up published research.
• Average daily returns, not a strategy. The numbers say nothing about costs, taxes or the risk of only holding four days a month.
• The main sample ends in 2005. Later evidence comes from shorter samples and secondhand summaries.
• Data-mining risk. Calendar effects were found by searching through a lot of possible rules, which tends to overstate how strong they are.

Sources

• McConnell, J. J., & Xu, W. (2008). Equity returns at the turn of the month. Financial Analysts Journal, 64(2), 49–64. IDEAS/RePEc; abstract at CFA Institute
• Ariel, R. A. (1987). A monthly effect in stock returns. Journal of Financial Economics, 18(1), 161–174.
• Lakonishok, J., & Smidt, S. (1988). Are seasonal anomalies real? A ninety-year perspective. Review of Financial Studies, 1(4), 403–425.
• Ogden, J. P. (1990). Turn-of-month evaluations of liquid profits and stock returns: A common explanation for the monthly and January effects. Journal of Finance, 45(4), 1259–1272.
• Carchano, Ó., & Pardo, Á. (2011, November). Calendar Anomalies in Stock Index Futures [Working paper]. Summarized in: LeCompte, S. (2011, November 21). Stock Index Futures Calendar Effects. CXO Advisory. cxoadvisory.com
• Quantpedia. Turn of the Month in Equity Indexes (strategy summary). quantpedia.com. Accessed October 12, 2026.

FAQ

What is the turn-of-the-month effect? The tendency for stocks to earn higher returns on the last trading day of a month and the first three trading days of the next.

Does the turn-of-the-month effect still work? It stuck around long after it was documented and has been found in S&P 500 futures, but it's a small average tilt, not a reliable signal for any one month.

Why do stocks rise at the start of the month? Suggested explanations include salary, pension and fund flows, and month-end rebalancing. None has been proven.

What's the best time to buy stocks? No date guarantees a gain. Historically, the last trading day of the month and the first three of the next carried most of the U.S. market's average return, but the effect is small in any one month.

---

Research and education only. Not financial advice. Past results do not predict future returns.

How this article was made: drafted with AI assistance from the research papers listed above; facts and figures checked by Gabriel Décarie and Karel Panneton.