Midterm election year stock market performance: dated observations and limits
Sources checked 2026-09-29.
What the annual claim measures
The claim that the S&P 500 was higher a year after every midterm since 1950 describes an election-to-one-year-later window, not the next session. LPL Research reported that pattern in its 2022 midyear discussion, with an average gain of 14.5%. At that publication date the completed 1950 through 2018 sample contained 18 midterms (n = 18, counted every four years). It did not yet include the year following the November 2022 election. A positive frequency in this selected sample is not a probability for 2026.
Keep the baseline and sample attached
U.S. Bank's October 3, 2022 note reports a 16.3% average over the 12 months following 15 midterms since 1962 (1962 through 2018, n = 15). It compares a -1.1% average in the preceding 12 months (n = 16) with an 8.0% average for all 12-month periods starting October 31 since 1961. That is a different sample from LPL's 1950 start. Neither annual window measures the first session after voting. These are source-reported historical returns, not SquawkFlow price calculations.
Two more windows, not two more confirmations
Carson's 2022 bear-market discussion reports a 14.1% average for the year after midterms since World War II; its text does not state the observation count, so no count is inferred here. Separately, AAII's May 10, 2018 report attributes a 1.1% average decline in May through October of midterm years to CFRA's Sam Stovall, using data starting in 1946. The same report cites LPL's 1.5% average gain for all May-through-October periods from 1950 through 2017. Those start dates and pre-election windows differ from the post-election claim. They must not be pooled into one average or treated as independent replications.
2010: policy on the first post-election day
The midterm was November 2, 2010. On November 3 the Federal Reserve announced the additional Treasury-purchase program commonly called QE2. The first post-election session therefore also contained a monetary-policy announcement. A close-to-close change on that date cannot establish an election-only effect.
2018: the FOMC met in the same week
The November 6, 2018 midterm was followed by the November 7 to 8 FOMC meeting. The first post-election session and the policy-decision session are separate dates within that week. A weekly return includes both windows and other market information.
2022: CPI and FTX are separate dated events
The November 8, 2022 midterm was followed by the November 9 session. The October CPI report was released November 10, and FTX's bankruptcy petition date was November 11. The FTX crisis and the inflation release belong in the surrounding-week context; neither justifies naming the whole interval an election effect. A next-session return and a return ending after CPI answer different questions.
What this cannot tell you
A historical pattern is not a forecast. Small, selected samples, changes in market structure, different return definitions and overlapping economic events limit comparisons. None of these observations identifies what caused a return or says what any market will do after November 3, 2026. This page reproduces no price history, chart or Yahoo-derived table. The Election Center carries the separate dated options measurement and its own limits.