S&P 500 Midterm-Year Pattern Since 1946: Does the 20-for-20 Record Hold for October 2026?
One line about midterm years circulates every two years: since 1946, the S&P 500 has been positive in the twelve months after every single midterm election — twenty for twenty. It is the kind of statistic that gets repeated without anyone opening the spreadsheet. What follows takes the claim apart in three checks: what the counting window really is, what the raw annual returns say when the record is rebuilt from scratch, and whether October 2026 looks anything like the setup the pattern needs. Every figure below was checked on October 12, 2026 against the primary sources listed at the end.
The claim, as it is usually stated
Truist co-chief investment officer Keith Lerner's research, cited by Yahoo Finance on October 5, 2026, tracks the S&P 500's return in the 12 months following every midterm election since 1946. Every single one of those 12-month windows has been positive — a perfect record across 20 midterm cycles. The average gain across that window is 14.4%, and the best outcome on record followed the 1954 midterms, a 34% gain. That is a genuinely unusual track record for any seasonal pattern in markets; most "always happens" claims about stocks fall apart within a decade or two of being tested. This one has held for eight decades, through wars, oil shocks, a dot-com bust and a global financial crisis.
Check 1 — Which windows are actually being counted
A second, overlapping statistic from the same research: since 1950, the S&P 500 has averaged a 6.6% gain in the fourth quarter of midterm years specifically, rising in that quarter 84% of the time. J.P. Morgan Asset Management's research arrives at the same 6.6% fourth-quarter average independently, which is a useful cross-check — two firms using different methodologies landing on the same number.
Why do the Truist/UBS figures differ from the calendar-year table?
They measure different 12-month windows. Truist and UBS track the period starting on Election Day in November; the calendar-year table in this article tracks January 1 through December 31 of the midterm year itself, which includes months the other statistics exclude.
The pattern is not just a Q4 story. Carson Group chief market strategist Ryan Detrick's research, since 1950, breaks midterm years down by month: September is the weakest month of the year on average, down 0.8%. October is the strongest, averaging a 3% gain and finishing positive 73.7% of the time. November is a close second, averaging 2.7% with gains 78.9% of the time.
UBS Global Research's Maxwell Grinacoff, looking at all 19 midterm elections since 1950, describes the mechanism: markets tend to go choppy from late August through early October, with a median decline of about 1.4% in that stretch, before rallying through year-end at a pace that has historically outpaced non-midterm years. UBS puts the September-through-year-end average at about 6% in midterm years, versus roughly 4% in other years. The three exceptions where the second half of a midterm year actually finished negative were 1978, 2002 and 2018.
| Source | What it measures | Historical average | Win rate |
|---|---|---|---|
| Truist (Lerner) | S&P 500, 12 months after Election Day, 1946–present | +14.4% | 100% (20 of 20) |
| Truist (Lerner) / J.P. Morgan | S&P 500, Q4 of midterm years, since 1950 | +6.6% | 84% |
| Carson Group (Detrick) | S&P 500, October of midterm years, since 1950 | +3.0% | 73.7% |
| Carson Group (Detrick) | S&P 500, November of midterm years, since 1950 | +2.7% | 78.9% |
| UBS (Grinacoff) | S&P 500, late Aug–early Oct dip, 19 midterm years since 1950 | −1.4% (median) | — |
Check 2 — Rebuilding the record from raw annual returns
The statistics above all measure a rolling window that starts on Election Day in November, not the calendar year itself. That is a meaningfully different question from "how did the S&P 500 do during the midterm year as a whole," so the table below answers that second question directly, using annual total-return data (price plus reinvested dividends) published by NYU Stern's Aswath Damodaran for every U.S. midterm year from 1950 through 2022 — 19 years in total. This is a separate calculation from the Truist and UBS figures above, not a restatement of them, and it is worth running because a calendar-year number is the one most readers can sanity-check against their own account statements.
| Midterm year | S&P 500 total return | Midterm year | S&P 500 total return |
|---|---|---|---|
| 1950 | +30.81% | 1990 | −3.06% |
| 1954 | +52.56% | 1994 | +1.33% |
| 1958 | +43.72% | 1998 | +28.34% |
| 1962 | −8.81% | 2002 | −21.97% |
| 1966 | −9.97% | 2006 | +15.61% |
| 1970 | +3.56% | 2010 | +14.82% |
| 1974 | −25.90% | 2014 | +13.52% |
| 1978 | +6.51% | 2018 | −4.23% |
| 1982 | +20.42% | 2022 | −18.04% |
| 1986 | +18.49% |
Summing those 19 figures and dividing by 19 gives an average calendar-year return of +8.30% across every U.S. midterm year since 1950 — lower than the headline "12 months after the election" figure, which makes sense, since the calendar year includes the typically weak January-through-September stretch that the post-election window skips. Twelve of the 19 years finished positive and seven finished negative, a 63% win rate, well short of the 100% record for the rolling 12-month-after-election window. The seven losing years — 1962, 1966, 1974, 1990, 2002, 2018 and 2022 — line up with recognizable shocks: the Cuban Missile Crisis year, the 1966 credit crunch, the 1973-74 oil embargo and bear market, the 1990 Gulf War recession, the 2002 dot-com unwind, the 2018 rate-hike selloff, and the 2022 inflation/rate-hike bear market. The pattern is real in the sense that it shows up reliably in the second half of these years specifically; it is not a guarantee that the whole calendar year will be kind.
Check 3 — Does October 2026 match the setup the pattern needs?
The S&P 500 closed at 7,811.54 on October 9, 2026, up 0.59% on the day and within half a percent of its 52-week high of 7,844.52, according to Yahoo Finance market data. That close puts the index up roughly 14% in price terms since the December 31, 2025 close of 6,845.50 — a strong year by any measure. It is also, less obviously, a midterm congressional election year, and midterm years have a well-documented seasonal pattern that has nothing to do with any particular president or policy. This analysis was prepared October 12, 2026, using the most recently completed U.S. trading session (October 9) as the reference point; no intraday or forecasted index level is used, and nothing below is a prediction of what the S&P 500 will do between now and the November 3, 2026 election.
Two things about 2026 don't fit the textbook version of the pattern. First, the "late August to early October median decline of 1.4%" that UBS describes as typical has not shown up this year in the same way: the S&P 500 is sitting within half a percent of its 52-week high as of the October 9 close, not several points below it heading into the stretch UBS flags as choppy. Second, the index was already up about 14% on the year by early October — well above the roughly 4-6% full-year averages the research above describes for the broader September-to-year-end window, because most of 2026's gain had already happened before Labor Day. Both facts cut against treating the seasonal pattern as something currently "due" to play out from here; a market already near its highs starting from a higher base has, mechanically, less room to repeat an average built partly from years that started the fourth quarter in a hole.
None of this breaks the base rate. It just means 2026's starting point for the pattern is different from a typical midterm year's, and the historical average was built by blending years that looked very different from this one going in — including three that lost more than 15% outright.
How much of the record holds up
A perfect 20-for-20 record sounds like a rule; it is really a sample of 20 overlapping, not-fully-independent observations, drawn from a market that has had a long-run upward drift for structural reasons (economic growth, earnings growth, inflation) that have nothing to do with the election calendar. A seasonal pattern built on 20 cases can be real and still be a weak basis for a specific near-term call, because the dispersion inside that average is large: the calendar-year table above ranges from +52.56% (1954) to −25.90% (1974), a nearly 80-point spread in the same "midterm year" bucket. The 12-months-after-election figures that make headlines (100% win rate, average +14.4%) describe the central tendency without describing how different the individual paths to get there have been, and they say nothing about valuation, interest-rate policy, or earnings — the factors J.P. Morgan's own research argues matter more than the calendar.
Does this mean the S&P 500 will rise between now and year-end 2026?
No. It means that in 20 of 20 prior midterm years measured from Election Day, the following 12 months were positive, and that Q4 specifically has averaged a gain 84% of the time since 1950. A historical frequency is not a forecast, and this article gives no buy, sell or hold instruction for any security.
Is a midterm year always volatile?
Historically, yes, more than other years on average — UBS specifically notes September and October have been the most volatile months since 1928, with that effect more pronounced in midterm years. Volatility and direction are separate questions; a volatile period can still end higher.
If I were weighing this pattern against a near-term decision, I would treat it as one data point that very mildly favors the rest of 2026 playing out in line with most prior midterm years' second halves, and nothing more than that — it does not tell a reader whether this specific October is more like 1998 (+28% for the year) or 2018 (−4%), both of which carried the same base rate going in. A seasonal tailwind is not a substitute for a reason tied to this cycle's specific earnings, rates and valuation picture.
This article is general market education, not individualized investment, tax or legal advice, and it does not recommend buying, selling or holding any security. Historical seasonal patterns are not forecasts and do not guarantee future results. Index levels are as of the October 9, 2026 close as reported by Yahoo Finance; figures from third-party research are reproduced as cited and were not independently re-derived except where this article says otherwise. Consult a qualified, licensed financial professional about your own circumstances before acting.
Tip: Seasonal patterns are easier to evaluate on paper than to sit through in real time. An investment journal helps you write down a plan before a volatile month, not during one, and a stock market history book goes deeper into how past cycles actually played out. (These are Amazon Associate links — we may earn a small commission on qualifying purchases.)
Correction October 25, 2026: one answer below originally said the following 12 months were positive in "19 of 19" prior midterm years measured from Election Day. The correct count is 20 of 20, which matches the Truist row in the comparison table above; the 19-year figure belongs to the 1950–2022 calendar-year table, which starts four years later (source: Truist via Yahoo Finance, October 5, 2026).
Sources
- Yahoo Finance — S&P 500 (^GSPC) quote and historical close data, accessed October 12, 2026 (October 9, 2026 close: 7,811.54)
- Yahoo Finance, reporting Truist co-CIO Keith Lerner's research — "Every midterm election since 1946 has meant one thing for stocks", October 5, 2026
- Fortune, reporting Carson Group's Ryan Detrick — "In midterm years, October is usually the best month for stocks", September 23, 2026
- UBS Global Research (Maxwell Grinacoff) and J.P. Morgan Asset Management midterm-election market research, as reported in the above and in contemporaneous coverage, September 2026
- NYU Stern School of Business (Aswath Damodaran) — historical annual returns on the S&P 500, 1928–2025
- dqydj.com — 2025 S&P 500 year-end close and total return calculation (December 31, 2025 close: 6,845.50)
Related Reading
- AAII Bearish Sentiment Hit a 95th-Percentile Extreme — With the S&P 500 Near Record Highs
- The Yield Curve Un-Inverted 25 Months Ago. Here's What Happened After Every Prior Un-Inversion Since 1978
- The S&P 500 Enters Q4 Up 14.6% for the Year. Here's What Happened the Last 21 Times It Was Up 10-20% Through September
Start Here: every analysis by topic · About · Privacy Policy
Comments
Post a Comment