Midterm years have been the weakest year of the four-year presidential cycle for US stocks, and the year after them the strongest. From 1950 to 2025, the S&P 500 returned 8.3% on average in midterm years, including dividends, against 20.8% in pre-election years; since 1928 the figures are 7.1% and 17.8% [3]. In the 12 months after each midterm election since 1950, the index rose in 18 of 19 cases, by a median of 14.4% [4]. These are small samples with large swings, and they say nothing reliable about any single year. As of October 12, 2026, the S&P 500 was up 14.0% for the year on price alone, ahead of the midterm election scheduled for November 3, 2026 [5].
What is the presidential election cycle in stocks?
The presidential election cycle splits every four-year term into four years: the post-election year (the first year of a term, such as 2025), the midterm year (the second, when all 435 House seats and about a third of the Senate are on the ballot, such as 2026), the pre-election year (2027) and the election year (2028). The question is whether stock returns differ by position in the cycle.
The table uses calendar-year total returns of the S&P 500 (price change plus reinvested dividends) compiled by NYU Stern professor Aswath Damodaran [3]. The median is the middle year, which a single extreme year cannot pull the way it pulls the average.
| Cycle year | Years (1928-2025) | Average, 1928-2025 | Median, 1928-2025 | Average, 1950-2025 | Share of years positive, 1950-2025 |
|---|---|---|---|---|---|
| Post-election (e.g. 2025) | 25 | 10.6% | 12.4% | 11.5% | 63% (12 of 19) |
| Midterm (e.g. 2026) | 24 | 7.1% | 5.0% | 8.3% | 63% (12 of 19) |
| Pre-election (e.g. 2027) | 24 | 17.8% | 22.5% | 20.8% | 100% (19 of 19) |
| Election (e.g. 2028) | 25 | 11.9% | 15.9% | 11.4% | 89% (17 of 19) |
Source: DataPorium calculation from NYU Stern annual S&P 500 total returns, 1928-2025 [3].
The range inside each group is wide. Midterm years since 1928 run from a loss of 25.9% (1974) to a gain of 52.6% (1954) [3]. Pre-election years have not had a calendar-year loss since 1950; the worst since 1928 was 1931, when the index lost 43.8% [3].
How did stocks do in the 12 months after each midterm?
To measure the year after each election, DataPorium compared the S&P 500's average level in November of each midterm year with its average level in November of the following year, using Robert Shiller's monthly index data [4]. Monthly averages smooth out single days, so the figures differ slightly from close-to-close changes. The table also shows each midterm year's calendar-year total return and which party held the White House, the House and the Senate before and after the election [6][7].
| Midterm | President (party) | House majority: before, after | Senate majority: before, after | Midterm-year total return | Next 12 months (Nov. to Nov.) |
|---|---|---|---|---|---|
| 1994 | Bill Clinton (D) | Democratic, Republican | Democratic, Republican | 1.3% | 29.2% |
| 1998 | Bill Clinton (D) | Republican, Republican | Republican, Republican | 28.3% | 21.5% |
| 2002 | George W. Bush (R) | Republican, Republican | Democratic, Republican | -22.0% | 15.4% |
| 2006 | George W. Bush (R) | Republican, Democratic | Republican, Democratic | 15.6% | 5.4% |
| 2010 | Barack Obama (D) | Democratic, Republican | Democratic, Democratic | 14.8% | 2.3% |
| 2014 | Barack Obama (D) | Republican, Republican | Democratic, Republican | 13.5% | 1.8% |
| 2018 | Donald Trump (R) | Republican, Democratic | Republican, Republican | -4.2% | 14.0% |
| 2022 | Joe Biden (D) | Democratic, Republican | Democratic, Democratic | -18.0% | 13.9% |
Sources: S&P 500 total returns [3]; 12-month changes calculated by DataPorium from monthly average index levels [4]; party control from the House and Senate party-division records [6][7].
All eight post-midterm windows from 1994 to 2022 were positive, with an average of 12.9% and a median of 14.0% [4]. The midterm years themselves averaged 3.7%, pulled down by 2002 and 2022 [3]. For the two most recent elections, exact daily closes tell the same story: the S&P 500 rose 11.7% from 2,755.45 on Election Day 2018 (November 6) to 3,076.78 a year later, and 14.5% from 3,828.11 on Election Day 2022 (November 8) to 4,382.78 a year later [5].
Does it matter which party wins the midterms?
In this small set, no clear pattern separates the outcomes. Grouping the eight post-midterm windows from 1994 to 2022 [4][6][7]:
- Democratic president: five midterms (1994, 1998, 2010, 2014, 2022), an average 12-month gain of 13.7%.
- Republican president: three midterms (2002, 2006, 2018), an average 12-month gain of 11.6%.
- House majority changed hands: five midterms (to Republicans in 1994, 2010 and 2022; to Democrats in 2006 and 2018), an average of 13.0%.
- House majority unchanged: three midterms (Republicans kept it in 1998, 2002 and 2014), an average of 12.9%.
With three to five cases per group, a single year moves each average by several points, so the gaps between groups carry little information. The same windows also include very different economic backdrops: the end of the 2000-2002 bear market, the period before the 2007-2009 financial crisis, and 2022, when the Federal Reserve raised its target range seven times, from 0%-0.25% to 4.25%-4.50% [2].
Certified results for every House and Senate race since 1976, with margins and swings by state and district, are on DataPorium's election pages at dataporium.tech/elections [1].
How does the post-midterm year compare with other years?
A fair test compares the post-midterm windows with every other 12-month window measured the same way. Across all 75 November-to-November windows from 1950 to 2024, the S&P 500 rose 75% of the time, with an average gain of 9.3% and a median of 11.3% [4]. The 19 windows that began in a midterm November averaged 14.9%, and the other 56 averaged 7.4%, with 68% positive [4].
The one post-midterm window since 1950 that did not rise ran from November 1986 to November 1987 and included the October 19, 1987 crash; the index ended the period essentially flat (-0.04% on monthly averages) [4].
How much weight can 19 midterms carry?
Not much, for four reasons:
- Small samples. There have been 24 midterm years since 1928, 19 since 1950 and 8 since 1994 [3]. When many calendar patterns are tested on the same history, some will look strong by chance alone.
- Other forces dominate. Interest rates, earnings, recessions and valuations move stocks far more than the election calendar. The midterm years 2002 and 2022 lost 22.0% and 18.0% for reasons that had little to do with the elections [2][3].
- The pattern is well known. When a calendar pattern is widely published, investors who try to trade on it can change prices ahead of time, which may weaken the effect in later years.
- Method matters. Calendar-year total returns, monthly averages and daily closes give different numbers for the same period; the direction here is consistent, but the exact percentages depend on the method.
Since 1950, the S&P 500 has been weakest on average in midterm years and strongest in the year after, but with 19 cases and wide swings, the pattern describes the past and does not forecast any one year.
Frequently asked questions
Do stocks go down in midterm election years?
Not usually, but they have done worse than in other cycle years. Since 1950, midterm years were positive 12 of 19 times, with an average total return of 8.3%, the lowest of the four cycle years [3].
What happens to the stock market after midterm elections?
Historically, it has tended to rise. Measured from the November of each midterm year to the next November, the S&P 500 gained in 18 of 19 cases since 1950, with a median of 14.4%; the exception, 1986-1987, was flat [4].
Is the stock market better under a Democratic or Republican Congress?
The midterm data do not show a meaningful difference. From 1994 to 2022, the 12 months after a midterm averaged 13.0% when the House changed hands and 12.9% when it did not, under both parties [4][6].
What is the best year of the presidential cycle for stocks?
The pre-election year, the third year of a term. Since 1950 the S&P 500 averaged a 20.8% total return in those years, and none of the 19 had a loss [3].
Sources & References
- [1] DataPorium Election Results (US House and Senate results since 1976)
- [2] DataPorium Interest Rates (Federal Reserve rate decisions since 1994)
- [3] NYU Stern School of Business (Aswath Damodaran), Historical Returns on Stocks, Bonds and Bills, 1928-2025 (updated January 5, 2026)
- [4] Robert J. Shiller, Online Data, U.S. Stock Markets 1871-Present (ie_data.xls, accessed October 13, 2026)
- [5] Federal Reserve Bank of St. Louis, FRED, S&P 500 (SP500), daily close (accessed October 13, 2026)
- [6] U.S. House of Representatives, History, Art & Archives, Party Divisions of the House of Representatives (accessed October 13, 2026)
- [7] U.S. Senate, Party Division (accessed October 13, 2026)
