Every month of the S&P 500 since 1950, ranked by average return and by how often it rose. Plus recent Novembers, the “best six months” rule, midterm years, and how much an average like this is worth.
Market risk reading, cross‑asset context and an events calendar: see Exent →
Historically, yes on average, but not reliably. Exent’s data on S&P 500 index month‑end closes shows November gained a mean 1.9% from 1950 to 2025, the highest of any month, with a median of 2.0%. It was higher in 53 of 76 years, or 69.7% of the time.
That still leaves 23 Novembers that fell, roughly 3 in 10. Five of them lost 5% or more: 1973, 1974, 1987, 2000 and 2008. The average is a description of the past, not a promise for any single year.
| November, 1950–2025 | S&P 500 |
|---|---|
| Average (mean) change | +1.9% |
| Median change | +2.0% |
| Years higher | 53 of 76 (69.7%) |
| Years lower | 23 of 76 (30.3%) |
| Worst November | −11.4% (1973) |
| Best November | +10.8% (2020) |
S&P 500 price index, end of October to end of November, without dividends. Past results, not a forecast.
By average return, November, December and April led; by how often the index rose, December came first. September was the only month with a negative average and the only one that fell more often than it rose. The table lists all 12 months.
| Month | Mean | Median | Years higher (of 76) | Worst | Best |
|---|---|---|---|---|---|
| January | +1.1% | +1.6% | 46 | −8.6% | +13.2% |
| February | 0.0% | +0.3% | 41 | −11.0% | +7.1% |
| March | +1.0% | +1.4% | 49 | −12.5% | +9.7% |
| April | +1.4% | +1.2% | 53 | −9.0% | +12.7% |
| May | +0.4% | +1.1% | 47 | −8.6% | +9.2% |
| June | +0.2% | +0.1% | 42 | −8.6% | +8.2% |
| July | +1.3% | +1.3% | 46 | −7.9% | +9.1% |
| August | 0.0% | +0.9% | 42 | −14.6% | +11.6% |
| September | −0.6% | −0.3% | 34 | −11.9% | +8.8% |
| October | +0.9% | +1.2% | 45 | −21.8% | +16.3% |
| November | +1.9% | +2.0% | 53 | −11.4% | +10.8% |
| December | +1.4% | +1.4% | 55 | −9.2% | +11.2% |
S&P 500 price index, month‑end closes, 1950–2025, 76 years per month. Look at the spread as well as the average: every month had at least one loss of about 8% or more, and October holds both the worst month (the 1987 crash) and one of the best. The gap between the best and worst average month is about 2.5 percentage points, small next to the swings inside any single month.
Well: 9 of the last 10 Novembers were higher, and the one decline, in 2021, was small at −0.8%. Two of them, 2020 and 2023, were among the five best Novembers since 1950.
| Year | November change |
|---|---|
| 2016 | +3.4% |
| 2017 | +2.8% |
| 2018 | +1.8% |
| 2019 | +3.4% |
| 2020 | +10.8% |
| 2021 | −0.8% |
| 2022 | +5.4% |
| 2023 | +8.9% |
| 2024 | +5.7% |
| 2025 | +0.1% |
A strong recent run makes a pattern feel safer than it is. The 1970s and 1980s had runs of weak Novembers too, and the decade before 2016 included −4.4% in 2007 and −7.5% in 2008. Ten results are a story, not a statistic.
Not in a useful way. After the 31 Octobers that fell, November gained a mean 1.9% and was higher 23 times. After the 45 Octobers that rose, it gained a mean 1.8% and was higher 30 times. October’s direction told you very little about November.
| October was… | Cases | November mean | November median | November higher |
|---|---|---|---|---|
| Down | 31 | +1.9% | +2.7% | 23 of 31 (74%) |
| Up | 45 | +1.8% | +1.9% | 30 of 45 (67%) |
| October was… | Down | Up |
|---|---|---|
| Cases | 31 | 45 |
| November mean | +1.9% | +1.8% |
| November median | +2.7% | +1.9% |
| November higher | 23 of 31 (74%) | 30 of 45 (67%) |
The down‑October group rose a little more often, but with 31 cases a handful of years decides that gap. It also contains the worst November of all, 1973.
On long‑run averages it held, but year to year it often failed. From 1950, the six months from end‑October to end‑April gained a mean 7.0%, against 2.1% for end‑April to end‑October. The winter half was higher in 58 of 76 periods, the summer half in 50.
| Period | Mean | Median | Higher | Worst |
|---|---|---|---|---|
| End‑Oct → end‑Apr (“best six months”) | +7.0% | +6.2% | 58 of 76 | −16.6% |
| End‑Apr → end‑Oct (“sell in May”) | +2.1% | +2.7% | 50 of 76 | −30.1% |
| Period | End‑Oct → end‑Apr (“best six months”) | End‑Apr → end‑Oct (“sell in May”) |
|---|---|---|
| Mean | +7.0% | +2.1% |
| Median | +6.2% | +2.7% |
| Higher | 58 of 76 | 50 of 76 |
| Worst | −16.6% | −30.1% |
The summer half still rose in two years out of three. Staying out of the market from May to October meant giving up those gains, and recent years show how costly that was. May–October gained 13.3% in 2024 and 22.8% in 2025, while the November–April periods of 2021–22 and 2024–25 lost 10.3% and 2.4%.
| Starting year | Nov → Apr | May → Oct |
|---|---|---|
| 2019 | −4.1% | +3.1% |
| 2020 | +27.9% | +12.3% |
| 2021 | −10.3% | +10.1% |
| 2022 | +7.7% | −6.3% |
| 2023 | +20.1% | +0.6% |
| 2024 | −2.4% | +13.3% |
| 2025 | +5.4% | +22.8% |
“Nov → Apr” starts at the end of October of that year and ends in April of the next. A strategy that switched to cash every May would also have paid trading costs and, in a taxable account, realized gains twice a year. The broader evidence on stepping in and out is in does market timing work.
Somewhat, on a small sample. In the 19 midterm years from 1950 to 2022, November gained a mean 2.7% (median 2.2%) and was higher 15 times, against a mean 1.6% and 38 of 57 in other years. Four midterm Novembers fell, including −5.3% in 1974 and −4.0% in 1994.
The stronger result is the six months that follow. From end‑October of a midterm year to end‑April, the index gained a mean 14.8% and rose all 19 times, against 4.3% and 38 of 56 in other years. Nineteen cases is still a thin record. The full study, with every midterm since 1950 and what came 3, 6 and 12 months later, is in the stock market after midterm elections.
There are several popular explanations, none proven. Each is plausible, and each could also be a story told after the fact about a pattern that is partly coincidence, not cause:
If these forces were strong and dependable, November would rarely fall. It fell 23 times in 76 years. Calendar effects that are widely known also tend to shrink, because traders try to get in early.
Not much, on its own. Each month has only 76 observations here, and a monthly average moves a lot with a few big years. Take out November 2020 and 1962 and the average drops from 1.9% to about 1.6%. The November lead over the next best months, December and April, is about half a percentage point.
This is why the definition page on monthly returns treats “best month to invest” patterns as mostly noise. The November edge is modest and was not reliable year to year. Trading on it also ignores taxes, spreads and the cost of being out of the market in the months that rose. The same caution applies to other calendar studies, such as how stocks behave on Fed days. For someone weighing whether to put money in now, the record on investing at an all‑time high says more than the month on the calendar.
The study uses month‑end closes of the S&P 500 index from December 1949 to October 2026, price only, without dividends. A month’s return is its month‑end close divided by the previous month‑end close, minus one. Each calendar month has 76 values, from 1950 to 2025. Six‑month periods run from the April or October close to the next October or April close.
Midterm years are 1950 to 2022, every four years. Dividends would add a little to every month alike. These are past results of one index; they describe how the market behaved, not how it will behave.
A 76‑year November average says nothing about this year’s conditions. Those conditions change, and they matter more for a decision than a calendar average.
Yes, on average. From 1950 to 2025 the S&P 500 gained a mean 1.9% in November, the highest of any month, and rose in 53 of 76 years. It still fell in 23 of them.
By average return, November (+1.9% from 1950 to 2025), followed by December and April at +1.4%. By how often the index rose, December came first, higher in 55 of 76 years.
September. From 1950 to 2025 the S&P 500 averaged −0.6% in September and rose in only 34 of 76 years. The single worst month was October 1987, at −21.8%.
On averages since 1950, November to April gained 7.0% vs 2.1% for May to October. But the summer half rose in 50 of 76 years, and it gained 13.3% in 2024 and 22.8% in 2025, so staying out would have missed real gains.
Nine of the ten Novembers from 2016 to 2025 were higher. The exception was 2021, at −0.8%. The best was 2020, at +10.8%.
The S&P 500 closed at 7,676.77 on Oct 26, 2026, up 12.1% for the year to date. Past Novembers do not tell you how this one will go.
Market intel provides a Risk ON / Neutral / Risk OFF reading from four pillars, cross-asset analysis, an events calendar and regular written notes. Context for reviewing a plan, not a crash forecast.
Educational content, not investment advice. Examples are illustrative; past performance does not guarantee future results.