korrents
Nate Silver

Nate Silver

@nate-silver · 33 positions · 1 change of mind · On ppll ↗

Forecaster, founder of FiveThirtyEight, author of The Signal and the Noise and On the Edge. Writes Silver Bulletin, where the predictions come with numbers and the numbers are later checked against what happened.

Nate Silver did not write this page.

We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.

The ideas Nate Silver's claims keep coming back to. The labels are ours; every quote is Nate Silver's own words.

Polls can miss in either direction, and their errors move together across races

Our label. 5 claims in Nate Silver's own words.

Across nine election years since 2008, polls were biased against Democrats once, against Republicans four times, and basically unbiased four times.

  1. Nate Silver Forecaster, founder of FiveThirtyEight In the 9 midterms or presidential years I’ve covered since FiveThirtyEight launched in 2008, polls have had a systematic bias against Democrats once, in 2012, a systematic bias against Republicans four times (the three Trump presidential elections + 2014), and were basically unbiased in the other four elections, including in the past two midterms. What if the polls are wrong again?natesilver.net · 3 Oct 2026
    In the 9 midterms or presidential years I’ve covered since FiveThirtyEight launched in 2008, polls have had a systematic bias against Democrats once, in 2012, a systematic bias against Republicans four times (the three Trump presidential elections + 2014), and were basically unbiased in the other four elections, including in the past two midterms.

Willingness to answer a poll is correlated with voting Democratic.

  1. Nate Silver Forecaster, founder of FiveThirtyEight That doesn’t mean they’ve solved it, because willingness to answer a poll is correlated with voting Democratic, and that’s inherently tricky. My pet peeve about the polling error debatenatesilver.net · 20 Oct 2026
    That doesn’t mean they’ve solved it, because willingness to answer a poll is correlated with voting Democratic, and that’s inherently tricky.

The models that treated races as a series of independent coin flips have died out.

Probabilistic thinking

  1. Nate Silver Forecaster, founder of FiveThirtyEight The models that wrongly treated races as a series of independent coin flips have died out, and deservedly so. My pet peeve about the polling error debatenatesilver.net · 20 Oct 2026
    The models that wrongly treated races as a series of independent coin flips have died out, and deservedly so.

Politics is nationalised and polarised: fewer competitive states, less ticket-splitting, weaker candidates

Our label. 3 claims in Nate Silver's own words.

Politics is increasingly nationalized.

  1. Nate Silver Forecaster, founder of FiveThirtyEight Politics is increasingly nationalized, so even though it’s easier for gubernatorial candidates to buck a state’s partisan trends than it is for their Senate counterparts, it’s getting more difficult to be a blue-state Republican or red-state Democrat. Do these Senate-governor splits make sense?natesilver.net · 23 Oct 2026
    Politics is increasingly nationalized, so even though it’s easier for gubernatorial candidates to buck a state’s partisan trends than it is for their Senate counterparts, it’s getting more difficult to be a blue-state Republican or red-state Democrat.

Perpetually increasing polarization means that fewer states are competitive, there is less ticket-splitting, and candidate effects are weaker.

  1. Nate Silver Forecaster, founder of FiveThirtyEight Perpetually increasing polarization means that: 1) fewer states are competitive; 2) there’s less and less ticket-splitting, so races for Congress are more predictable based on presidential voting and 3) “candidate effects,” including incumbency, are weaker, so there’s less opportunity for a candidate with strong personal qualities to overcome the partisan tide. Could Minnesota spoil the blue wave?natesilver.net · 9 Oct 2026
    Perpetually increasing polarization means that: 1) fewer states are competitive; 2) there’s less and less ticket-splitting, so races for Congress are more predictable based on presidential voting and 3) “candidate effects,” including incumbency, are weaker, so there’s less opportunity for a candidate with strong personal qualities to overcome the partisan tide.

Statewide ticket-splitting is expected to decrease even further in 2026.

  1. Nate Silver Forecaster, founder of FiveThirtyEight Based on FLIPR’s current projections, we expect statewide ticket-splitting to decrease even further, with a median gap of just six points in 2026. Do these Senate-governor splits make sense?natesilver.net · 23 Oct 2026
    Based on FLIPR’s current projections, we expect statewide ticket-splitting to decrease even further, with a median gap of just six points in 2026.

The 2026 midterm environment favours Democrats more strongly than cautious expert ratings admit

Our label. 3 claims in Nate Silver's own words.

Cautious expert race ratings, like Cook's, are misaligned with the long history of strong opposition performance in midterms under unpopular presidents.

  1. Nate Silver Forecaster, founder of FiveThirtyEight Conversely, expert ratings imply a much different political environment. The Cook Political Report, for example, counts seven races as “toss-ups,” including states like New Hampshire where models and prediction markets imply that Democrats are heavy favorites. This cautious view is misaligned with the long history of strong opposition performance in midterm elections, especially when the incumbent president is as unpopular as Trump. Expert ratings are ignoring signs of a blue wavenatesilver.net · 20 Sept 2026
    Conversely, expert ratings imply a much different political environment. The Cook Political Report, for example, counts seven races as “toss-ups,” including states like New Hampshire where models and prediction markets imply that Democrats are heavy favorites. This cautious view is misaligned with the long history of strong opposition performance in midterm elections, especially when the incumbent president is as unpopular as Trump.

It would take a massive October surprise or a bigger-than-2016 polling error to save the Republican majority in the House, despite Republicans’ aggressive redistricting.

  1. Nate Silver Forecaster, founder of FiveThirtyEight It would take a massive October surprise or a bigger-than-2016 polling error to save the Republican majority there, despite Republicans’ aggressive redistricting. Will Republican gerrymanders become dummymanders?natesilver.net · 13 Oct 2026
    It would take a massive October surprise or a bigger-than-2016 polling error to save the Republican majority there, despite Republicans’ aggressive redistricting.

Democrats' voter-targeting efforts are plausibly more efficient and effective than Republicans' efforts under Trump.

  1. Nate Silver Forecaster, founder of FiveThirtyEight Democrats’ voter-targeting efforts are plausibly more efficient and effective than the Republican ones under Trump. Why did Ohio stop being a swing state?natesilver.net · 1 Oct 2026
    Democrats’ voter-targeting efforts are plausibly more efficient and effective than the Republican ones under Trump.

Democrats' polling strength in 2026 sits in rural, red and energy-producing states rather than blue ones

Our label. 2 claims in Nate Silver's own words.

Democratic polling overperformance has generally been strongest in the central U.S., in rural, agricultural, and energy-producing states.

  1. Nate Silver Forecaster, founder of FiveThirtyEight But generally, Democratic polling overperformance has been strongest in the middle part of the country, in rural, agricultural states and in states that are big energy producers. Diesel prices could crush Republicans in the heartlandnatesilver.net · 22 Sept 2026
    But generally, Democratic polling overperformance has been strongest in the middle part of the country, in rural, agricultural states and in states that are big energy producers.

Fuel price inflation is much higher in diesel-heavy farm states and may explain some surprisingly competitive Senate races.

  1. Nate Silver Forecaster, founder of FiveThirtyEight Inflation in fuel prices is much higher in diesel-heavy farm states — and may explain some surprisingly competitive Senate races. Diesel prices could crush Republicans in the heartlandnatesilver.net · 22 Sept 2026
    Inflation in fuel prices is much higher in diesel-heavy farm states — and may explain some surprisingly competitive Senate races.

In football stats, separate skill from noise and opportunity: efficiency predicts, touchdowns mislead

Our label. 5 claims in Nate Silver's own words.

Dropped passes should, with high confidence, be attributed to the receiver rather than the quarterback.

  1. Nate Silver Forecaster, founder of FiveThirtyEight We are quite confident that drops should be attributed to the WR, not the quarterback. QBERT 2026 NFL quarterback ratingsnatesilver.net · 5 Sept 2026
    We are quite confident that drops should be attributed to the WR, not the quarterback.

Touchdown rates are noisy and opportunity-dependent, and don't transfer well when a quarterback changes teams.

  1. Nate Silver Forecaster, founder of FiveThirtyEight TDs are both noisy and opportunity-dependent, and TD rates don’t “travel” all that well when a QB changes teams. How our ELWAY forecasts worknatesilver.net · 7 Sept 2026
    TDs are both noisy and opportunity-dependent, and TD rates don’t “travel” all that well when a QB changes teams.

Efficiency measures such as yards per play and first downs gained are generally more predictive than scoring alone.

  1. Nate Silver Forecaster, founder of FiveThirtyEight Efficiency measures, such as yards per play and first downs gained, are generally more predictive than scoring alone. 2026 ELWAY NFL projectionsnatesilver.net · 7 Sept 2026
    Efficiency measures, such as yards per play and first downs gained, are generally more predictive than scoring alone.

A 17-game NFL season with imbalanced schedules inherently introduces randomness no projection model can eliminate.

Probabilistic thinking

  1. Nate Silver Forecaster, founder of FiveThirtyEight A projection model will never solve the problem that a 17-game NFL season with imbalanced schedules is inherently going to introduce a degree of randomness. Inside ELWAY’s 2026 NFL forecastnatesilver.net · 9 Sept 2026
    A projection model will never solve the problem that a 17-game NFL season with imbalanced schedules is inherently going to introduce a degree of randomness.

AI is developing about as the worriers expected, and regulation is both needed and more likely than assumed

Our label. 5 claims in Nate Silver's own words.

AI has arguably been hitting its marks so far, roughly as anticipated by people who are really worried about existential threats.

  1. Nate Silver Forecaster, founder of FiveThirtyEight I think AI has basically been hitting its marks so far as anticipated by people who are really worried about existential threats. We’re not ready for superpersistent AInatesilver.net · 15 Sept 2026
    I think AI has basically been hitting its marks so far as anticipated by people who are really worried about existential threats.

It seems like common-sense risk management that the AI industry needs far more regulation and probably some enforceable slowdown.

  1. Nate Silver Forecaster, founder of FiveThirtyEight It just seems like common-sense risk management that the industry needs far more regulation, and probably some sort of enforceable slowdown. We’re not ready for superpersistent AInatesilver.net · 15 Sept 2026
    It just seems like common-sense risk management that the industry needs far more regulation, and probably some sort of enforceable slowdown.

Substantive AI regulation is probably more likely to pass than conventional wisdom suggests.

  1. Nate Silver Forecaster, founder of FiveThirtyEight I don’t know what odds I’d give, but I’m probably bullish on the odds of substantive regulation passing relative to the conventional wisdom. A supermajority wants to slow AI downnatesilver.net · 17 Sept 2026
    I don’t know what odds I’d give, but I’m probably bullish on the odds of substantive regulation passing relative to the conventional wisdom.

The current pro-AI coalition opposing regulation is not expected to be very durable, since its factions hate one another.

  1. Nate Silver Forecaster, founder of FiveThirtyEight So I don’t expect this pro-AI coalition to be very durable. The AI denialists and the Trump fans hate one another. A supermajority wants to slow AI downnatesilver.net · 17 Sept 2026
    So I don’t expect this pro-AI coalition to be very durable. The AI denialists and the Trump fans hate one another.

AI is likely to become a more relevant issue before 2028.

  1. Nate Silver Forecaster, founder of FiveThirtyEight But not everyone is so bullish on AI, and with the arguably alarming rate at which the models are improving, this issue is likely to become more relevant before 2028. The Democrats’ AI primary is heating upnatesilver.net · 6 Oct 2026
    But not everyone is so bullish on AI, and with the arguably alarming rate at which the models are improving, this issue is likely to become more relevant before 2028.