ClaimsNate Silver

“Don’t bet on that changing dramatically in the final days leading up to the election.”

Nate Silver, author, Silver Bulletin 23 October 2026 Do these Senate-governor splits make sense? (Silver Bulletin) archived copy

Prediction AI policy and governance Political

Silver advised against betting that the pattern of Democratic Senate candidates doing surprisingly well in traditionally red states and less well in bluer states would change dramatically in the final days before the 2026 election.

Unresolvable The public record could not settle it.

Prediction: Statements about events outside the person's control, checked once they fell due: did they come true?

What would count as true or false

Compare Senate polling averages (Silver Bulletin, or other major aggregators) on about 23 October 2026 with those on election eve (2 or 3 November 2026), for the competitive 2026 Senate races in red states and in blue states. The pattern is that Democrats run relatively strong in red-state races and relatively weak in blue-state races. True: the pattern still holds on election eve, and the average Democratic margin in each group moves by under about 2 points. Mostly true: the pattern holds, but the averages move by 2–4 points or one race shifts notably. Mixed: the pattern holds in one group of states but clearly weakens in the other. Mostly false: the pattern largely erodes, with average movement over about 4 points in the direction that removes it. False: the pattern disappears or reverses before the election.

Set before the research began.

Evidence (9)

  1. E1 2026 Senate Forecast: Every Race, Polls and Map | Poliwave Poliwave

    “| State | D | I | R | Poll lead | Forecast | Polls | Newest | |----------------------------------------------------------|------|------|------|-----------|----------|-------|--------------| | Iowa | 46.7 | — | 46.4 | D +0.3 | D +0.7 | 41 | Nov 1, 2026 |”
  2. E2 2026 Senate Forecast: Every Race, Polls and Map | Poliwave Poliwave

    “Shares are the poll average (each pollster's latest poll from the last 90 days, newer polls counting more); where a race has one poll, the row shows that poll.”
  3. E3 2026 Senate Forecast: Every Race, Polls and Map | Poliwave Poliwave

    “| Iowa | 46.7 | — | 46.4 | D +0.3 | D +0.7 | 41 | Nov 1, 2026 |”
  4. E4 2026 Senate Forecast: Every Race, Polls and Map | Poliwave Poliwave

    “| Michigan | 49.6 | — | 44.9 | D +4.7 | D +6.7 | 68 | Nov 2, 2026 |”
  5. E5 2026 Senate Forecast: Every Race, Polls and Map | Poliwave Poliwave

    “Polls through Nov 2, 2026”
  6. E6 On the eve of the election, this is what the Senate battleground polling tells us MS NOW · 2 November 2026

    “In Georgia, New Hampshire, Minnesota and Michigan — seats the party needs to hold to control the chamber — Democrats are ahead in the polling averages by 6 to 10 points, all outside the margin of error, according to FiftyPlusOne.”
  7. E7 On the eve of the election, this is what the Senate battleground polling tells us MS NOW · 2 November 2026

    “In toss-ups Maine, Texas, Kansas and Ohio, FiftyPlusOne puts Democrats up by 5, 4, 4 and 3 points, respectively.”
  8. E8 Senate races to watch: 6 states shifting days before Election Day NewsNation (via Yahoo) · 28 October 2026

    “Decision Desk HQ's forecast over the past three weeks have shown Democrats and independents gaining ground in states that usually vote Republican, like Kansas, Nebraska and South Carolina, while Republicans are improving their odds in Iowa, Michigan and Minnesota.”
  9. E9 Democratic Victory or Polling Miss: 3 Possible Midterms Scenarios Newsweek · 23 October 2026

    “A Newsweek analysis finds that Democrats would win both chambers if today’s polling averages were exactly right.”

Reasoning

The criteria require comparing Senate polling averages from about 23 October 2026 with those on election eve. The evidence gives only election-eve figures. FiftyPlusOne shows Democrats ahead by 6 to 10 points in Georgia, New Hampshire, Minnesota and Michigan, seats the party needs to hold (E9), and by 3 to 5 points in the toss-ups of Maine, Texas, Kansas and Ohio (E10), and Poliwave gives one snapshot through Nov 2 (E6–E8). E11 says the Decision Desk HQ forecast moved toward Democrats in red states and toward Republicans in Michigan, Minnesota and Iowa, which suggests the pattern held, but that is a forecast as of Oct 28, not a polling-average comparison. No late-October baseline figures are given, so the size of any movement cannot be measured.

Cite this claim

Nate Silver, “Silver advised against betting that the pattern of Democratic Senate candidates doing surprisingly well in traditionally red states and less well in bluer states would change dramatically in the final days before the 2026 election.” (said 23 October 2026). Verdict: unresolvable. Reliability API, claim c_47dc06bd9ce0, method 0.1.0+28da144b, dataset release v1.1.2. https://epistacklabs.com/claims/c_47dc06bd9ce0

This claim is in dataset release v1.1.2, with the same verdict.

Reply to this claim

For Nate Silver, or someone speaking for them: a correction or counter-evidence. Replies are read before anything is shown; an approved reply shows your name, role, date, statement and link. Your email address is never shown and is deleted 30 days after the decision (privacy).

More prediction claims by Nate Silver

  1. Resolved true Prediction 30 June 2022
    “If the GOP takes over the House next year one thing you can be pretty sure of is COVID panels/investigations.”

    If Republicans take control of the House in 2023, they are pretty likely to hold COVID panels or investigations.

  2. Mixed Prediction 8 April 2025
    “Children's toys are gonna become much more expensive because they're made in countries like Vietnam in Southeast Asia that are heavily taraffed.”

    Children's toys will become much more expensive in the US because they are made in heavily tariffed countries such as Vietnam.

  3. Mostly true Prediction 13 August 2024
    “I think people might understand with Kamala Harris that it's inherently a difficult circumstance and a little bit of a long shot bid, but that her chances are probably better than Biden.”

    Kamala Harris's bid is inherently difficult and a little bit of a long shot, but her chances of winning the 2024 presidential election are probably better than Joe Biden's were.

  4. Mixed Prediction 8 September 2023
    “Most voters aren’t paying attention yet, and it will be nine months to a year before they do.”

    Most US voters will not start paying attention to the 2024 presidential election until nine months to a year after September 2023.

  5. Mostly false Prediction 8 September 2023
    “I’d argue that 1 (Biden winning) is more likely than 2 (Trump winning) which, in turn, is more likely than 3 (someone else winning).”

    Biden winning the 2024 presidential election is more likely than Trump winning, which is more likely than someone else winning.

All of them

Verdicts are made automatically; how verdicts are made.