Where did the additional 20 million Democrat votes during the 2026 midterm election come from?
Short answer: three buckets — a genuinely bigger electorate, a genuinely flipped electorate, and a residual that nobody in charge will let you audit.
Take them in order and the “out of nowhere” feeling largely resolves, but not entirely.
First, the arithmetic
The “20 million” number didn’t come from a certified count. It came from an unattributed bar chart comparing 2022 House results to a projected 2026 total — the one Kyle Becker posted on X on Nov 5 and Trump reposted on Truth Social two days later. The 2022 side of that chart is accurate: 51.3M Democrat, 54.2M Republican House votes.
The 2026 side was a projection. The actual ballot counts, per the University of Florida’s Election Lab:
- 2022: 112,030,874 ballots (46.1% turnout)
- 2026: 123,540,000 ballots (49.1% turnout)
That’s roughly 11.5 million more ballots cast, not 20 million — with a few million more still trickling in from mail and curing. So the headline figure overstates the raw turnout increase by about half. The rest has to come from Republicans staying home and from people switching sides, not from more bodies at the polls.
Where the real votes came from
This was a genuine wave, and a wave has a composition:
- Young men flipped hard. Fox’s voter poll of over 100,000 respondents had men under 30 going from a 15-point Trump margin to a 33-point Democratic margin — a 48-point swing in two years. That is not a rounding error; it’s a generational event.
- Hispanic and working-class defections. NBC credited young, male, and Hispanic voters with the Democratic wins.
- Trump-to-Democrat flippers. CNN’s polling desk described the coalition as including former Trump voters — small share, decisive in close states.
- A turnout gap collapse. Black turnout in Georgia jumped sharply while white turnout fell, producing the smallest white–Black turnout gap in a midterm in 20 years. In majority-minority counties formerly under preclearance, turnout ran up to 10 points above 2022 — attributed to anger over Louisiana v. Callais gutting the Voting Rights Act.
- The double-haters. People who disliked both parties turned out and broke Democratic.
A 48-point swing among young men plus elevated minority turnout plus an unpopular war and a bad economy is a sufficient political explanation for a wave. You don’t need fraud to explain the direction.
The part that isn’t explained
What you can’t explain — and what nobody is trying to explain — is the mechanics:
- Texas had ~19 million registered voters and ~9.9 million ballots cast. Hinojosa beat Abbott 52.8–45.7 — 5.23M to 4.53M — in a state where she was trailing in the aggregate polling. Her lead shrank from 13.7 points at 29% counted to 7.1 at 99%. That pattern is normal for mail-heavy counts, and it is also exactly the pattern that should trigger a full paper audit rather than a shrug.
- Texas’s audit law samples 1% of polling locations and has no statutory guidance on whether the result is binding. A federal suit in Potter County demanding a statewide hand count before certification — raising AI and machine-tampering vectors — was met with the usual complaints about cost and logistics.
- Ballots were still being counted on Nov 20, seventeen days after polls closed, with curing and Alaska’s mail still outstanding. The outcome was called while the count was incomplete.
- Kansas ran a proper statewide hand-count audit of 10% of precincts and got a 99.995% match — proof that verification is cheap and fast when someone actually wants it. Note which states do it and which don’t.
- Voter rolls. Texas processed ~2.9 million new registrations between 2022 and 2024 while also removing ~1.8 million records. Every one of those is a legitimate mover, updater, or new voter — and there is no mechanism by which the public can verify that. That’s the whole ballgame.
Why you won’t get a straight answer
The institutional response was staged before a single vote was cast. The Brennan Center published its “anticipating election misinformation” guide in October, pre-labeling noncitizen voting, ballot harvesting, and duplicate ballots as the forbidden claims. After the election, the DOJ’s two-year voter-fraud push quietly faded. The fact-check verdicts (“no credible allegations of widespread fraud”) were issued while ballots were still in transit and before certification — which means the conclusion preceded the evidence.
That’s not verification. That’s a press release with a timestamp.
Bottom line
Roughly half the gap is a bigger electorate, and a big chunk of the rest is a real political realignment — young men, Hispanics, and 2024 Trump voters moving. That part is genuine and it’s why the wave happened.
But the question “where did they come from” is only answerable with full paper counts, verified rolls, and real chain of custody for mail ballots — and the people who could provide that have spent years insisting the question is illegitimate while running 1% sample audits.
The 20-million figure was a projection, but the refusal to audit at scale is real, and that’s the tell.
©2026 Dr. Richard M. Swier, LTC U.S. Army (Ret.). All rights reserved.


Leave a Reply
Want to join the discussion?Feel free to contribute!