Nobody Knows Anything, 2026 Midterms Edition

250 House seats, ~55 Senate seats.

The past few months have been fascinating, watching pundits wrestle with what would become a historic election.

All presidents lose Congressional seats in the midterms (with only 4 exceptions). Given how close the margins had been, a shift in at least the House was likely. But this particular president has been uniquely polarizing. He won in 2024 by broadening traditional Republican demographics, adding a coalition that included young males, the uneducated, and Hispanics. That left the GOP coalition vulnerable in a normal midterm. How did that it turn out?

~~~

Everywhere I looked – stock futures, prediction markets, polls – turned out to be a panoply of errors.

Stock futures were in a deep selloff by 6 p.m.; prediction markets underpriced Democratic candidates; and the Polls, which got the general direction of the race right, had significant errors in the magnitude of the changes and wildly misread Democratic wins in traditionally GOP strongholds.

The S&P sold off about 100 points; now it’s up 150 points from those lows. At ~noon today we are +3% on the Dow and +2% on the SPX and NDX; that’s quite a swing.

maybe double-check your math…?

About those Prediction Markets:

The FT reported that “Polymarket’s gamblers had a bad night of forecasting, undermining the reputation of prediction markets.” (I have been criticizing their inaccuracy forever).

And the Polls!

The margins were far wider than even the final polls suggested; polling missed the “Texas Shock” where Talarico won a Senate seat by ~10 points and over a million votes; Gina Hinojosa won the Texas Governor’s race by 7% and 750k votes; traditional red states like Kansas and South Carolina elected Democratic senators for the first time in nearly a century. The so-called “Shy Trump” voter vanished; exit polls said 43% of voters cast ballots to oppose Trump vs 16% who supported him. Turnout Models missed intense motivation among Dem-leaning voters.

Some right-wing politicians imagined the MAGA coalition would last, but it showed signs of fraying almost immediately after the 2024 election. Each of the president’s three signature policies was a factor in the coalition pieces I mentioned to start: 1) ICE‘s aggressive immigration enforcement; 2) War, and a muscular foreign policy supporting multiple wars; 3) Inflation, driven partly by a silly tariff policy. 

~~~

Whenever I channel William Goldman to emphasize how little we know about the future, I feel I am cheating, taking advantage of a flaw in the human condition that most either don’t know or don’t care about. We evolved in a very specific environment to deal with threats very different from the ones we face today.

Tribalism, a bad information diet, and an inability to accurately assess our own skill sets are a poor tool set when analyzing social events and politics. It is an utterly fatal combination when putting capital to work in markets…

Previously:
Why Polls Fail: Attitude, Intention, Behavior (November 3, 2026)

Bad Polling is a Behavioral Problem (October 6, 2024)

Finding Investing Lessons in 2020 Election Fails (December 10, 2020)

Polls Were Wrong (again). They Often Are (November 4, 2020)

What Models Don’t Know (May 6, 2020)

Nobody Knows Anything archive

The post Nobody Knows Anything, 2026 Midterms Edition appeared first on The Big Picture.