Short answer: Across long U.S. history, the strongest average months for the S&P 500 have usually been November, December, and April, and the weakest has usually been September. The well-known “Sell in May and go away” / Halloween pattern — stronger November–April than May–October — shows up in decades of data and in academic work. That does not mean you should trade the calendar. Seasonality is a historical average, not a forecast, and after taxes, trading costs, and missed best days, most long-term investors are better off staying invested than rotating in and out by month.
This refresh updates BearSavings’ guide to bullish and bearish months with sourced averages, clear definitions (Santa Claus rally vs a generic December rally), and a practical “what to do instead” section. It is education for a U.S. audience, not investment advice.
What stock market seasonality is (and is not)
Seasonality means some calendar windows have delivered different average returns than others over long samples. Researchers and market almanacs track it for the same reason they track presidential cycles or midterm patterns: the averages are interesting. They are not a trading system by themselves.
Useful framing:
- Average ≠ this year. A month can be “historically weak” and still finish green — September 2024 and September 2025 both did on price-return series many investors watch.
- Window matters. “Since 1950,” “since 1928,” and “last 10 years” can rank months differently.
- Price vs total return. Some published monthly averages exclude dividends; total-return series look a little better in every month. Label the series you cite.
- Not a substitute for a plan. Paycheck investing, asset allocation, and costs usually matter more than whether the calendar says “September.”
For the “should I wait for a better month?” version of this question, see our dollar-cost averaging vs lump sum guide and why staying invested is less risky than not investing.
Direct answer: which months are best and worst?
Using average S&P 500 monthly price returns since 1950 (YCharts data as charted by Visual Capitalist, March 2025):
Historically stronger months (roughly):
- November: about +1.82% average — often the strongest calendar month in that sample
- December: about +1.49%
- April: about +1.46%
- July: about +1.28%
- March: about +1.13%
- January: about +1.07%
- October: about +0.91% (volatile reputation, middling-to-positive average)
Historically weaker months (roughly):
- September: about −0.72% — the clear laggard in that sample
- February and August: about flat to slightly negative on average (−0.01% each in the same chart)
- May and June: small positive averages (+0.30% and +0.11%) that still sit in the weaker half of the year
LPL Research (commentary updated September 2, 2025) similarly notes that over roughly the last 75 years, September’s average S&P 500 return is about −0.7%, with positive Septembers only about 44% of the time — the lowest positivity rate of any month in their cut of the data. They also note that October–December has averaged about +4.2% since 1950, finishing higher about 80% of the time.
Those averages answer the search query. They do not tell you what next September will do.

Month-by-month snapshot (S&P 500 averages since 1950)
Treat this as a map of history, not a buy/sell calendar. Figures are approximate average monthly price returns from the Visual Capitalist / YCharts (via Bilello) series since 1950:
- January (~+1.07%): Often positive on average; overlapping ideas include the January effect (especially small caps) and early-year flows. For the January deep dive on this site, see Is January really the best month for stocks? and The January Effect explained.
- February (~−0.01%): Near flat on long-run average — a reminder that “winter” is not uniformly strong month by month.
- March (~+1.13%): Typically solid inside the historically stronger half.
- April (~+1.46%): One of the best single months in the sample; often discussed alongside tax-deadline and contribution-driven flows (correlation, not proof of cause).
- May (~+0.30%): Starts the weaker half in the Almanac framing; still positive on average.
- June (~+0.11%): Often muted.
- July (~+1.28%): A notable exception inside May–October — summer is not uniformly weak.
- August (~−0.01%): Thin-liquidity reputation; long-run average near flat.
- September (~−0.72%): The consistent long-run laggard in many U.S. samples.
- October (~+0.91%): Famous crash anniversaries (1929, 1987) create fear; the average month is still positive in this series.
- November (~+1.82%): Often the strongest calendar month since 1950 in this dataset.
- December (~+1.49%): Strong on average; not the same thing as the short Santa Claus rally window (defined below).
2024 reality check (same Visual Capitalist chart): several “strong” months disappointed and several “weak” months worked — for example April 2024 about −4.16%, September 2024 about +2.02%, November 2024 about +5.73%, December 2024 about −2.50%. History is a distribution, not a script.
“Sell in May,” the Halloween indicator, and the Best Six Months
“Sell in May and go away” is the folk saying. The cleaner research label is the Halloween indicator: stocks have often done better from roughly November through April than from May through October.
What reputable sources actually say:
- Stock Trader’s Almanac (Yale Hirsch): The “Best Six Months” strategy focuses on roughly November 1 through April 30, then reduces equity exposure in the other half. An Almanac note summarizing the pattern since 1950 cited an average DJIA gain of about 7.5% in November–April versus about 0.4% in May–October. More recent Almanac materials describe the S&P 500 November–April span since 1950 as advancing roughly 77% of the time with an average gain around 7.1%.
- Bouman & Jacobsen (2002), American Economic Review: In “The Halloween Indicator, ‘Sell in May and Go Away’: Another Puzzle,” they found higher November–April returns than May–October returns in 36 of 37 countries studied — a famous academic stamp on the folk wisdom (AER abstract).
- Investopedia’s overview of the adage stresses the same half-year split, then immediately warns that automatic selling can miss strong summers and that costs and taxes matter.
Practical takeaway for buy-and-hold readers: the half-year gap is one of the more durable calendar patterns in the literature — and still a poor reason to liquidate a long-term S&P 500 or broad index plan every spring. Many May–October stretches have been excellent (including parts of 2020 and 2025 on price data).
The September effect
September owns the worst long-run average in many U.S. large-cap samples:
- Visual Capitalist / YCharts since 1950: about −0.72%
- LPL Research (~75-year window, 2025 commentary): about −0.7%, positive only ~44% of the time
- CFRA figures widely cited in financial press: roughly −0.7% average since 1945, with November and December among the best months (~+1.5% each in those summaries)
Investopedia’s September Effect page notes the long negative average (including a “since 1928” framing) while emphasizing that recent median Septembers have sometimes looked better, that economists largely discount it as a tradable edge, and that September is not down every year. Stocks have been down in September only a bit more than half the time in some long samples — closer to a coin flip than a guaranteed crash.
Common stories offered for September weakness (mutual-fund window dressing, back-to-school cash needs, post-summer positioning) are hypotheses, not proven causes. Treat them as folklore with a statistical shadow.

October: scary stories, middling averages
October’s reputation comes from extreme history — the 1929 crash week, Black Monday in 1987, and other crisis chapters — not from a reliably negative average month. In the since-1950 Visual Capitalist / YCharts series, October’s average is about +0.91%.
That mix is why October gets labeled “volatile” more accurately than “bearish.” Bad Octobers are memorable; average Octobers are often fine. For crisis-context reading on this site (different intent), see how war affects stock market prices and protecting your downside in investing.
Santa Claus rally — and why it is not “all of December”
Media often calls any fourth-quarter bounce a “Santa Claus rally.” Yale Hirsch coined a narrower definition in the Stock Trader’s Almanac (1972): the last five trading days of December plus the first two trading days of January.
Verified points commonly cited (Investopedia summarizing Almanac / Hirsch family figures):
- Average S&P 500 gain around +1.3% over that seven-session window since 1950 (similar averages are quoted since 1969)
- Positive results a large majority of the time (often cited around ~78–79%)
- Almanac lore: “If Santa Claus should fail to call, bears may come to Broad and Wall” — a failed rally has sometimes preceded weaker years (pattern, not prophecy)
December the calendar month can still be strong on average (~+1.49% since 1950 in the Visual Capitalist series) even when the short Santa window is what Almanac readers track. Do not conflate the two.
January follow-ons (January Barometer, first-five-days, small-cap January effect) are sibling seasonal ideas — covered on BearSavings at Is January really the best month for stocks? rather than duplicated here.
A quick note on election-year seasonality
Presidential election years sometimes soften the “Sell in May” story. Investopedia cites Ned Davis Research figures that, since 1950, the S&P 500 has risen about 78% of the time from April 30 to October 31 in presidential election years, versus roughly 64% in non-election years. Almanac materials also discuss election-year overlays on monthly patterns.
Use that as context, not a trade. Election years still contain recessions, bubbles, and random macro shocks. Seasonality layered on politics is still an average.
Recent reality: 2024 and 2025 did not follow the script
Seasonality fans got a useful stress test in the last two full years. Using YCharts’ S&P 500 monthly price return series (ex-dividends; discontinued series still widely referenced):
2024 highlights:
- Strong “good” months that behaved: November ~+5.73%
- Strong months that didn’t: April ~−4.16%, December ~−2.50%
- Weak-reputation months that worked: May ~+4.80%, September ~+2.02%
2025 highlights (same YCharts price series):
- Soft start: February ~−1.42%, March ~−5.75%, April ~−0.76%
- Powerful “weak half” stretch: May ~+6.15%, June ~+4.96%, July ~+2.17%, August ~+1.91%, September ~+3.53%
- Late year: October ~+2.27%, November ~+0.13%, December ~−0.05%
- January 2025 ~+2.70%; January 2026 ~+1.37% on the same series
Point for readers: if you had sold every May on principle, you would have stepped aside for some of the best months of 2025. Averages describe the past; your calendar still has to live through surprises. (Always prefer total-return figures for long-horizon planning; the price series above is used because it matches the published monthly tables many seasonality charts cite.)
Why seasonality is a weak trading edge after costs and taxes
Even if the average gap is real, implementing it is hard for taxable U.S. investors:
- Transaction costs and spreads eat a thin seasonal premium, especially in taxable accounts.
- Capital gains taxes on repeated May sales / November buys can dominate the historical half-year gap. See our 2026 capital gains tax brackets overview for the rate map (not advice).
- Missed best days often cluster around stressful months — the same windows seasonal traders try to dodge. Sitting in cash through a sharp rebound is how “risk reduction” becomes permanent underperformance.
- Crowding: once a pattern is famous, some of the edge may migrate earlier (for example, more August selling ahead of September).
- Sample quirks: changing market structure, index composition, and monetary regimes mean 1950–1980 is not the same machine as 2009–2025.
Academics still debate why Halloween / September patterns exist. For portfolio construction, “interesting anomaly” is the right bucket — next to narrative bias, not next to your contribution schedule.
What long-term investors should actually do
If you came here asking which months are best and worst, keep the answer and drop the temptation:
- Automate contributions through strong and weak months alike — payroll 401(k) deferrals and IRA transfers beat calendar timing for most households. 2026 contribution ceilings live in our 401(k) contribution limits guide.
- Prefer broad, low-cost funds over seasonal sector hopping. If you are choosing wrappers, see index fund vs ETF.
- Use a written allocation, not a weather forecast. See our asset allocation by age guide for decade-by-decade stock/bond framing — then pick a target mix you can hold.
- If you receive a windfall, compare lump sum vs DCA with evidence, not with September folklore: dollar-cost averaging vs lump sum.
- Rebalancing on a schedule (or when bands trip) is a cleaner “calendar” habit than selling every May.
- Expect volatility in historically rough months without needing a prediction. Emotional preparedness is underrated; see my net worth dropped $800k… for a behavioral case study (personal story, not a model portfolio).
Compounding math still dominates seasonal noise — sanity-check horizons with a compound interest calculator and a long S&P 500 lens.

Frequently asked questions
What is the best month for the stock market historically?
In the widely cited S&P 500 average monthly price returns since 1950 (YCharts via Visual Capitalist), November leads at about +1.82%, with December and April close behind. “Best” still means average, not guaranteed.
What is the worst month for stocks?
September is the usual answer in long U.S. samples — about −0.72% average since 1950 in that same chart, and about −0.7% in LPL’s ~75-year cut — with the lowest share of positive months. Individual Septembers can still rally hard.
Does “Sell in May and go away” work?
The Halloween / Best Six Months pattern (stronger November–April than May–October) is real in a lot of historical data and in Bouman & Jacobsen’s 2002 multi-country study. Mechanically selling every May is another question: taxes, costs, and missed summer rallies (including parts of 2025) often erase the paper edge for taxable investors.
What is the Santa Claus rally?
It is not “stocks go up in December.” Per Yale Hirsch / Stock Trader’s Almanac, it is the last five trading days of the year plus the first two of the new year, with a long-run average S&P gain near +1.3% and a high historical win rate. Treat it as seasonal trivia unless your plan already keeps you invested.
Is the October effect real?
October’s crash lore is real; a reliably negative average October is not. Since 1950, October’s average in the Visual Capitalist / YCharts series is about +0.91%. Fear and average returns are different statistics.
Should I wait until November to invest new money?
Usually no. Waiting for a “good month” is a timing decision. For new paycheck money, invest on your schedule. For a large one-time sum, compare lump sum vs DCA on evidence — not on folklore — in our DCA vs lump sum article.
How did 2025 fit the seasonal script?
Poorly, in the ways that matter for traders: several May–September months were strong on YCharts’ S&P price-return series (May ~+6.15%, June ~+4.96%, September ~+3.53%), while December finished roughly flat to slightly negative. Seasonality described the past better than it predicted that year.
Is this investment advice?
No. BearSavings publishes education. Seasonal averages are historical descriptions. Your risk tolerance, time horizon, tax situation, and written plan belong to you (and, if you use one, your advisor).
Final words
Bullish months in U.S. history have clustered around late fall through early spring, with November, December, and April often near the top of average-return tables. Bearish months are led by September, with a softer May–October half-year in Almanac and academic work. Know the map so a red September feels less like a unique crisis — then keep funding the plan that compounds through all twelve months.
Sources stamped in the READY_FOR_PUBLISH package (Visual Capitalist / YCharts, LPL Research, Stock Trader’s Almanac, Bouman & Jacobsen 2002 AER, Investopedia definitions, CFRA summaries as cited in secondary coverage). Figures are historical and not guarantees.
Keep learning: courses on this topic
Self-paced courses to put this article into practice.
Udemy
Personal Finance from Zero: Budget, Save, Borrow & Invest
Money touches every day of your life, yet almost nobody is taught how it works. Where does your pay go between payday and the end of the…
View course
Udemy
Stock Trading: Invest and Trade Safely with Stock Sector ETF
Stock Lovers, Traders, Investors, Stock Market Aficionados! Master the basics of trading ETFs. NO NEED to pick and choose individual…
View course
Udemy
How To Confidently Invest & Analyze Stocks Like an Expert
Are you tired of feeling overwhelmed by the stock market or unsure if a stock is actually a good investment? This course, "Investing in…
View course

Leave a Comment