The $10K Testwith JJ Free weekly email

Test #3 · Published October 20, 2026

Should you sell stocks before the midterm elections? We tested every midterm since 1934

By JJ, host of The $10K Test · · How the tests work

Watch on YouTube · Subscribe to the channel

My verdict: BUSTED.

The short answer: Not in this test. Over 23 midterm elections from 1934 to 2022, the run-up to Election Day really was bumpier, but the market still ended it higher in 15 of 23, and selling before every midterm and buying back by Election Day left 21.8% less money over the long run. In the 12 months after Election Day, stocks rose after 22 of the 23 midterms.

The key numbers

The claim

You've probably heard it from a friend, a neighbor or the radio: sell before the election and sit in cash until it's over. No published source we found says it outright; it's folk advice. But the fear has a real basis. Capital Group, one of the big fund companies, writes that "market volatility is higher in midterm years, especially in the weeks leading up to the election." It also writes the opposite claim people hear: "Since 1950, the average one-year return following a midterm election was 15.4%." The 2026 midterm is Tuesday, November 3, 2026. This test is about the calendar only: no parties, no candidates, no predictions.

Source: Capital Group, Matt Miller and Chris Buchbinder, January 26, 2026

How we tested it

Is the stock market more volatile before midterm elections?

Yes. In 11 of 23 midterm run-ups (47.8%), the S&P 500 fell 10% or more from a high at some point before Election Day. In all other years, over the same dates: 15 of 69 (21.7%), about 1 in 5. The typical daily move was 22% bigger in midterm run-ups.

Some were ugly. Inside those 3 months the market fell 24.65% at its worst in 1974, 19.31% in 2002 and 16.91% in 2022. Capital Group is right that election season is bumpier.

Should you sell stocks before the midterm elections?

Not in this test. Bumpy isn't the same as down: the market still finished the run-up higher in 15 of 23 midterms. On average, $10,000 in stocks grew to $10,239 by Election Day, against $10,093 in T-bills. Only 2 run-ups ended down 10% or more: 1946 and 1990.

Selling did win 10 of 23 times, including the last 2 (2018 and 2022). In 2022, $10,000 in stocks at the end of July was $9,268 by Election Day; in T-bills, $10,091. But in 1982 stocks went to $12,839 while T-bills reached $10,222. Measured a year after the election, when selling won it won by about $728 per $10,000; when holding won, by about $861.

Over the long run it added up. $10,000 in January 1934, price only, became $7,680,870 held, and $6,007,520 selling before every midterm and buying back by Election Day: 21.8% less. Buying back the day after Election Day: 29.3% less. Staying out until 12 months after each election: $560,239.

Do stocks go up after midterm elections?

Historically, yes. In the 12 months after Election Day, stocks rose after 22 of 23 midterms, by 15.17% on average ($11,517 per $10,000). The same 12 months in all other years averaged 6.55%. Since 1950: 19 of 19 up, averaging 15.11%, close to Capital Group's 15.4%.

The only loss was 1938 (-5.06%), and 1946 was flat (0.07%). After the 8 run-ups that lost money, the next 12 months rose 8 of 8 times, by 13.92% on average.

Is the midterm election stock pattern just luck?

Probably not pure luck, but not a promise either. Out of 200,000 random picks of 23 years from all 91, the average reached 15.17% or more only 1.4 times in 100. But 23 elections is a small sample, the pattern was found by looking back, and it is now widely known.

U.S. Bank reviewed 31 midterm elections from 1900 through 2025 and wrote: "Statistical testing did not find differences large or consistent enough to establish a dependable midterm-election effect." Its team added: "The economy remains the primary engine of market returns." Both can be true: in this test, the year after is history, not a forecast.

Every midterm since 1934: $10,000 from the end of July to Election Day, and the year after
Midterm (last close before results)Run-up: stocksRun-up: T-billsWorst drop inside run-upYear after: stocks
1934 (Nov 5)$10,392 (+3.92%)$10,006-11.24%+42.57%
1938 (Nov 7)$10,892 (+8.92%)$10,002-13.67%-5.06%
1942 (Nov 2)$11,075 (+10.75%)$10,010-2.18%+24.26%
1946 (Nov 4)$8,536 (-14.64%)$10,010-22.55%+0.07%
1950 (Nov 6)$10,852 (+8.52%)$10,035-3.78%+16.17%
1954 (Nov 1)$10,295 (+2.95%)$10,025-4.42%+33.22%
1958 (Nov 3)$10,926 (+9.26%)$10,060-2.32%+11.06%
1962 (Nov 5)$10,021 (+0.21%)$10,076-10.52%+24.78%
1966 (Nov 7)$9,657 (-3.43%)$10,149-12.86%+12.89%
1970 (Nov 3)$10,791 (+7.91%)$10,166-4.34%+12.69%
1974 (Nov 5)$9,470 (-5.30%)$10,226-24.65%+18.69%
1978 (Nov 7)$9,322 (-6.78%)$10,218-12.94%+6.41%
1982 (Nov 2)$12,839 (+28.39%)$10,222-6.02%+19.89%
1986 (Nov 4)$10,427 (+4.27%)$10,144-9.42%+1.12%
1990 (Nov 6)$8,750 (-12.50%)$10,205-17.04%+25.14%
1994 (Nov 8)$10,161 (+1.61%)$10,136-4.98%+27.07%
1998 (Nov 3)$9,912 (-0.88%)$10,120-14.58%+21.97%
2002 (Nov 5)$10,041 (+0.41%)$10,043-19.31%+14.90%
2006 (Nov 7)$10,832 (+8.32%)$10,137-1.78%+6.71%
2010 (Nov 2)$10,835 (+8.35%)$10,004-7.14%+3.71%
2014 (Nov 4)$10,422 (+4.22%)$10,001-7.4%+4.48%
2018 (Nov 6)$9,784 (-2.16%)$10,059-9.88%+11.66%
2022 (Nov 8)$9,268 (-7.32%)$10,091-16.91%+14.49%

S&P 500 price only (no dividends); 3-month T-bills. Run-up = last close of July to the last close before results (2022: from 2022-07-29). Year after = the next 12 months.

The verdict

BUSTED. The run-up to a midterm really is bumpier, with a 10% drop inside it in 11 of 23 midterms. But the market still finished the run-up higher in 15 of 23, and selling before every midterm and buying back by Election Day left 21.8% less money from 1934 to 2026. The second claim, that stocks do well the year after a midterm, is CONFIRMED as history: up after 22 of 23 midterms, 15.17% on average. Just not a promise.

What this test leaves out

Data: S&P 500 index, price only, daily, 1934 to October 2026 (the 90-stock S&P index before March 1957); 3-month T-bills from FRED; no taxes. Every number on this page comes from the test results.

Questions people ask

Should you sell stocks before the midterm elections?
Not in this test. Selling before each of 23 midterms since 1934 and buying back by Election Day turned $10,000 into $6,007,520 instead of $7,680,870 (price only), 21.8% less.
Does the stock market go down before midterm elections?
It gets bumpier, but it usually doesn't end down. The S&P 500 fell 10% or more from a high inside 11 of 23 midterm run-ups, yet still finished the run-up higher in 15 of 23.
How do stocks do the year after a midterm election?
In the 12 months after Election Day, stocks rose after 22 of 23 midterms since 1934, by 15.17% on average. The only loss was 1938, down 5.06%.
What if you sell before the election and buy back after the results?
It was worse. Buying back the day after Election Day turned $10,000 from 1934 into $5,433,752, 29.3% less than holding.
When is the 2026 midterm election?
Tuesday, November 3, 2026. In this test the run-up ran from the last trading day of July to the last close before results, about 3 months, and T-bills never lost money in any of the 23 run-ups.
Is the midterm election stock market pattern reliable?
Not as a promise. 23 random years averaged 15.17% or more only 1.4 times in 100, but 23 elections is a small sample, and U.S. Bank's statistical test of 31 midterms found no dependable effect.

What's in the video

  1. 0:00 Intro
  2. 0:39 The claim
  3. 1:50 How we tested it
  4. 3:06 The bumps are real
  5. 3:57 But selling didn't pay
  6. 5:44 The year after
  7. 7:12 Why it happened
  8. 7:51 What to do instead
  9. 8:59 The verdict
  10. 9:29 Next test
  11. 9:48 Free weekly email + disclosure
Read the script of the video

The words JJ read on camera for this test. The final video may differ slightly.

If you had sold your stocks before every midterm election since 1934, and bought back by Election Day, you'd have about a fifth less money today. And in the 12 months after the election, the market went up 22 times out of 23.

The next midterm is Tuesday, November 3. So, should you get out first? Today we test it on 92 years of the stock market.

But the people who say sell get one thing right. It's the reason so many people panic every October, and I'll show you what it is in this video.

I am JJ, and this is the 10K Test. Every week I take a piece of money advice everybody repeats and test it against real market history. Not opinions. What the money actually did.

Today: the midterm election. You've probably heard it from a friend, a neighbor, or somebody on the radio. Sell before the election and sit in cash until it's over, because elections make the market crash.

And you've probably heard the opposite, too: stocks do great in the year after a midterm.

No politics and no hunches here, just 92 years of real prices, and we show you the misses too. If that's your kind of channel, subscribe.

It's easy to see why people believe the first one. Every fall in an election year, the ads get louder and the headlines get scarier. Even Capital Group, one of the big fund companies, writes that market volatility is higher in midterm years, especially in the weeks leading up to the election.

One promise before we start: this is about the calendar. No parties, no candidates, and no guesses about who wins.

Here's exactly what we did, so you can check it.

We used the S&P 500 index going back to 1934. Before 1957, that's the smaller, 90-stock S&P index that came before it. For cash, we used 3-month Treasury bills. The T-bill record we used starts in 1934, so that's where we start.

That gives us 23 midterm elections, from 1934 to 2022.

For each one, we put $10,000 in stocks at the end of July and checked it at the last market close before any results came in, on or just before Election Day. That's about 3 months. We did the same with $10,000 sitting in T-bills. Then we followed the stocks for 12 months after Election Day.

And to see whether midterms are really different, we ran the exact same dates in every other year, too.

Two things work in favor of selling here. We used the index price only, without dividends, so the stocks don't get paid while the T-bills earn their full interest. And no taxes, which you'd often owe when you sell in a regular account. So this is the best case for selling.

Let's start with the part the claim gets right.

In 11 of the 23 midterm run-ups, the market fell 10% or more from a high at some point before Election Day. That's almost half. In all the other years, over the same dates, it happened 15 times out of 69. About 1 in 5.

The day-to-day swings were bigger, too. The typical daily move in a midterm run-up was about 22% bigger than in other years.

And some of them were ugly. In 1974, the market fell almost 25% inside those 3 months. In 2002, about 19%. In 2022, about 17%.

So if election season has ever felt bumpy to you, you're not imagining it. Capital Group is right about that, and our numbers agree.

But bumpy isn't the same as down.

Even with all those drops, the market finished the run-up higher in 15 of the 23 midterms. On average, $10,000 in stocks grew to about $10,239 by Election Day. In T-bills, about $10,093. Stocks beat cash 13 times out of 23.

A crash that lasted all the way to Election Day was rare. Only 2 run-ups ended down 10% or more: 1946 and 1990.

Now, I promised to show you the misses. Selling did win 10 times. And it won the last 2. In 2022, $10,000 in stocks at the end of July was down to about $9,268 by Election Day. In T-bills, it would have been about $10,091. If you sold in 2022, you looked smart.

But look at 1982. Stocks went from $10,000 to about $12,839 in those 3 months. T-bills: about $10,222.

Measured a year after the election, when selling won, it won by about $728 per $10,000. When holding won, it won by about $861. So each time it's close to a coin flip, but the coin is tilted against selling.

And over 92 years, that tilt adds up.

$10,000 in the S&P 500 in January 1934, price only, is about $7.7 million today. Sell before every midterm and buy back by Election Day, and it's about $6.0 million. About a fifth less.

And if you waited for the results and bought back the day after, it was worse: about 29% less.

Now the other claim: the year after a midterm is good for stocks.

This one held up. In the 12 months after Election Day, stocks went up after 22 of the 23 midterms. The average gain was about 15.2%. Per $10,000, that's about $11,517 a year later.

The same 12 months in all the other years averaged about 6.6%.

Capital Group says that since 1950, the average one-year return after a midterm was 15.4%. We checked it ourselves and got 15.1%, and all 19 of those years were up. Close enough.

The only loss was 1938, down about 5%. And 1946 was basically flat: up 0.07%.

Here's the number that surprised me most. 8 midterm run-ups lost money. In the 12 months after each of those 8, the market went up, by about 13.9% on average. If those drops had scared you out for good, you'd have missed the part that paid.

Could the year after just be luck?

We checked that, too. If you pick 23 years at random out of all 91, the average comes out that high fewer than 2 times in 100. So it's probably not pure luck. But it's only 23 elections. And a pattern you find by looking back can stop working, especially once everybody knows about it.

So why would the run-up be bumpy and the year after be strong? Honestly, nobody can prove the reason. The usual explanation is uncertainty. Before the vote, nobody knows what the next 2 years will look like. Once the votes are counted, that question is answered, whatever the answer is.

U.S. Bank's research team looked at 31 midterm elections going back to 1900, and their statistical test didn't find a dependable midterm effect. As they put it, "The economy remains the primary engine of market returns." So treat the year after as history, not a promise.

So what do you do with this? Three things from the same data.

One: don't let the calendar move your money. Selling before the midterm won only 10 times out of 23, and over 92 years it cost about a fifth of your money.

Two, and this is the thing the sell crowd gets right: expect the bumps, so they don't push you out. About half of midterm run-ups had a 10% drop at some point. If you know that going in, a drop in October is just election season, not a signal. And remember, after the 8 run-ups that lost money, the next 12 months went up every time.

Three: if you need money in the next year, for a roof, a car, or your living expenses, that money belongs somewhere safe in any year, not just election years. T-bills didn't lose money in a single one of these 23 run-ups. Stocks fell 10% from a high in almost half.

And one warning about the year-after rule: don't load up because of it. 1938 lost money, 1946 went nowhere, and 23 elections is a small sample.

So, sell before a midterm election. Verdict: BUSTED. The run-up really is bumpier, with a 10% drop about half the time. But the market still finished higher most of the time, and selling cost money more often than it saved.

And stocks do well in the year after a midterm? CONFIRMED, as history: up 22 times out of 23, about 15% on average. Just not a promise.

Next on the 10K Test: sell in May and go away. The saying is right about the calendar. Summers really were the weaker half. And following it still cost $99,000 for every $10,000. How can both be true? If you want another test right now, there are more on the channel.

Every week I send a free email with the full numbers behind each test, the tables, the charts, and what we're testing next. Free weekly email at 10Ktest.com.

And tell me in the comments what money advice you want tested next.

This video is for education and entertainment, not personal financial advice. Your situation is different; talk to a professional before changing your investments because of an election.

Disclosure: I don't own an S&P 500 index fund. I do own some individual stocks, including Apple, which is part of the S&P 500. Nobody paid for this video.

More tests

Every test →  ·  All videos →

Education and entertainment only. Not personal financial advice. Your situation is different; talk to a professional before changing your investments because of an election. Disclosure: I don't own an S&P 500 index fund. I do own some individual stocks, including Apple, which is part of the S&P 500. Nobody paid for this video.