How the Post-Midterm Stock Selloff Spread

In short: US stocks had their best post-midterm day in decades on 4 November, then lost more than that gain on 5 November. The reversal began with a change of view about AI infrastructure policy. It became a large move because retail sellers, then hedge funds, then rule-based funds sold in sequence, and few buyers stepped in.

The result and the rally

Democrats won control of both the House and the Senate in the 3 November elections. Fidelity described the outcome as divided government from 2027, with Democrats controlling Congress and Republicans the White House.

The result matched the scenario JPMorgan had rated most likely. Before the vote, the bank gave a "Blue Wave" a 66% likelihood, with only a small S&P 500 reaction in either direction. Its note also warned that a blue wave "could mean a selloff in artificial intelligence stocks" if it brought tighter regulation of data centers, though it expected bills limiting AI expansion to struggle against a presidential veto.

The first reaction was the opposite of a selloff. CNBC tied it to President Trump accepting the result and saying he would work with Democrats. On Wednesday 4 November, according to CNBC's market coverage:

AFP, citing CFRA Research, reported it was the Dow's biggest post-midterm gain since 1982, and the S&P 500's second-biggest post-midterm one-day gain over that period. Bonds rallied and oil dropped: the 10-year Treasury yield fell more than 6 basis points to 4.884%, and US crude fell 8.29% to $88.12 a barrel.

The reversal

By Thursday the reading had changed. CNBC reported that investors feared a Democratic sweep of Congress, together with wins in key governor races, "could decelerate the data center buildout" on which AI spending depends. The closes on 5 November:

The Nasdaq and the small-cap Russell 2000 both entered a correction. The iShares Semiconductor ETF fell more than 12%, its worst day since March 2020.

How the selling spread

Bloomberg News reconstructed the day in a piece republished by Financial Advisor. Its account is a sequence, with each stage feeding the next.

1. Retail sold first. "The first half hour of Thursday trading belonged to retail." Vanda Research said net selling of single stocks by retail investors ranked behind only two days, both in March 2020.
2. Professional investors followed. Bloomberg reported that hedge funds and long-only managers accelerated the selling. Goldman Sachs's prime brokerage desk estimated fundamental long/short funds lost 1.3% on the day.
3. Rule-based funds compounded it. Strategies that trade on volatility and trend rather than views began to sell as prices and volatility moved.
4. Few buyers appeared. "The problem was no major buyers emerged on the other side."

The third stage is where the size came from. The estimates Bloomberg reported were made during the session by named strategists, and they describe selling pressure, not confirmed trades:

Each of these is mechanical. Options dealers who are short gamma must sell as prices fall to stay hedged. Volatility-control funds cut stock exposure when measured volatility rises, and the VIX went from 16.6 at Thursday's open to near 28 by the close, per Bloomberg. Leveraged ETFs rebalance at the close in the direction of the day's move. Trend followers sell once prices break through levels their models track. None of them needs a view on AI policy; they respond to the price move itself.

The same reflex shows up elsewhere in market plumbing. Clearing houses also demand more collateral when volatility rises, which is why their margin models are called procyclical.

How broad it was

The damage was not limited to technology. CNBC's tally for 5 November:

• Sectors: all 11 S&P 500 sectors fell, led by information technology, down 7.30%
• Breadth: declining stocks beat advancing ones on the New York Stock Exchange by 2,344 to 416, almost 6 to 1
• Largest companies: the "Magnificent Seven" lost more than $1 trillion in combined market value, including more than $440 billion for Nvidia
• Volatility: the VIX gained 11.57 points, its biggest point gain since 4 April 2025, and hit its highest level since 31 March

Bloomberg put the total at $3.7 trillion of US market value wiped out in a day, and called it the Nasdaq 100's worst drop since March 2020. The figures measure different things, so they do not add up to each other. The market-wide circuit breakers did not trigger; CNBC noted during the session that conditions would have to get a lot worse before they paused trading.

The partial rebound

The selling did not continue at that pace. On Friday 6 November the Nasdaq rose 1.09% to 24,579.42 and the S&P 500 rose 0.31% to 7,376.86, while the Dow slipped 0.46% to 51,285.73. Bloomberg noted chip stocks were leading the recovery. For the week, the Nasdaq lost 3.4%, the S&P 500 2.7% and the Dow 0.9%.

On Monday 9 November the Dow rose 2.15% to 52,389.75, the S&P 500 rose 2.35% to 7,549.94, its best day since 8 April, and the Nasdaq rose 2.74% to 25,253.14.

What the episode shows, and what it does not

It shows how quickly a change in narrative turns into forced selling. The news on Thursday was not new: the result was known on Wednesday, and JPMorgan had named the AI risk before the vote. What changed was how investors read it. Once prices fell, strategies that size positions by volatility and trend added selling without reference to the news at all.

It does not show what Congress will do about data centers. The new Congress is not seated until January, and the selling reflected fears about policy, not enacted policy. The flow figures above are estimates made in real time by strategists, not reported trades. More on the systems between an order and a settled position is in Markets & Infrastructure.