Nate Silver Senate Odds Jumped to 60 Percent Range for Democrats

Let’s discuss Silver’s three models plus the betting markets. How FLIPR Forecasts the Midterm Elections FLIPR is the new acronym for Nate Silver’s Election Models. Aug 11, 2026 FLIPR stands for Forecast with Leading Indicators, Polls and (Expert) Ratings. - Lite derives as much information as possible from polls alone; - Classic is polls + fundamentals; - And Deluxe is polls + fundamentals + expert ratings. Why forecasting midterms is hard Compared with our presidential forecast, which mostly defaults toward being “polls only” by election day, FLIPR is more of an even blend between polls and fundamentals (and in Deluxe, expert ratings, though these have less influence than the other major components). This is both because non-polling data is richer for House, Senate, and gubernatorial races than for the presidency — there are around 500 races combined in each midterm, rather than just one presidential race — and because polling data is sparser. Many important House races receive little to no nonpartisan polling, for example, so we need other methods to forecast these races. Basically, FLIPR seeks to estimate what a poll of a given race would say today if it were conducted on a likely-voter basis. In this way, it hopes to avoid relying on “stale” polling. FLIPR also applies the likely voter adjustment to its averages of the generic ballot and presidential approval, which our regular averages do not. This makes a material difference. As of early August, for example, Democrats are up by about 6.5 points in our default generic ballot average, but that advantage expands to around 8 points after likely voter adjustments, which is the version that FLIPR uses. (The likely-voter adjusted version of the generic ballot can be found on the FLIPR landing page.) Lite The “Lite” version of our forecast is supposed to be “polls only”. But that creates a problem for races that have little or no polling. Our solution is to impute polling-like estimates for these races based on FLIPR’s calculation of the national political environment, and polling in similar races using an internal system we call CANTOR (Congressional Algorithm using Neighboring Typologies to Optimize Regression). Deluxe Every year, we have a debate about whether to retire the Deluxe version and stick to a model that uses purely objective inputs. Even though the expert ratings reflect other people’s judgments and not ours, there are some undesirable properties from including them. For instance, if experts are looking at Silver Bulletin forecasts to help calibrate their ratings, the entire process becomes somewhat recursive. Furthermore, in recent elections, the Deluxe version of our forecasts typically only “calls” one or two additional races correctly across the ~500 races we forecast every year compared to Classic. Although the expert forecasts are good at discriminating between relatively stronger and weaker races for each party, they have a weird and persistent bias. Namely, expert raters tend to implicitly assume a neutral political environment (i.e., a generic ballot of about zero) until quite late in the cycle. As of early August, for instance, the unadjusted expert ratings imply a political environment of around D +2 versus the roughly D +7 environment calculated by FLIPR Classic. In a D +2 environment, a candidate like Talarico probably wouldn’t win despite Paxton’s weaknesses, but a D +7 would require much less overperformance. FLIPR’s predecessors have always been aware of this tendency and “de-biased” the expert ratings before applying them. But in auditing the code this year, we discovered some shortcomings in the way this de-biasing process had been applied by the model. Basically, because of some artifacts in how races are rated, there was a bit of leakage from the experts’ implicit assumptions about the national environment (i.e., wrongly assuming the generic ballot = 0) into the Deluxe forecast. And we specifically do not think the experts’ views of the national environment provide much utility at all; their value is solely in the rank-ordering of different races relative to the national environment. In working on the model this year, we developed better techniques for removing this bias completely. Having read all that, the sane thing to do seems to go with classic. I do that, but I also look ahead. What I have found is the models are catching up to what I have thought all along. I factor in where I think inflation, gasoline prices, and the mood is headed. That’s not always easy, but all year my presumption has been Trump will make things worse for himself and Republicans. And that is exactly what’s happened. 62 percent now seems about right. Things can change, but I don’t expect them to, other than for Trump to make matters worse for himself and Republicans with Trumpian hissy fits and policy errors. What Happened Starting September 10 (Yellow Highlight) The streak of excellent polls for Democrats continues. The latest New York Times/Siena poll shows the generic ballot at D +8, and this week’s YouGov/Economist wave has it at D +12 among likely voters. As a result, Democrats’ chances of retaking the Senate are up to 59 percent in our Deluxe forecast. That’s the highest they’ve been this cycle. What happened is more polls, and they broke in the direction I expected. All year long I have expected Hispanics, young voters, swing voters, and independents to break hard for Democrats. The recent polls support that idea in spades. That’s how you get D+8 and D+12 in national polls. Spotlight Texas Talarico has had a lead in 8 of the last 10 polls with one tied. Only four of the polls look current to me. Spotlight Ohio Once again I consider only four polls are current. This time, all of them favor the Democrat. Spotlight Alaska There is only one current poll in Alaska, and it’s not even that current, August 20-23 by Alaska Survey Research. Silver has Democrat Mary Peltola up 50.8 to 49.2, in the Deluxe model, Peltola up 57-43 in the lite, but Sullivan ahead 56-44 in the classic. I will discuss Alaska in greater detail in a separate post, but I think it’s well over 60 percent likely Democrats win Alaska, and primarily not because of Alaska’s screwy ranked voting setup. Senate Slate Classic If Silver’s model proves accurate and I am correct about Alaska, two big ifs, Democrats may easily take seven Senate seats. And that would not surprise me in the least. Kalshi Senate Map For the first time, Kalshi Betters have the Democrat odds of taking the Senate 55 percent or better. I take the betting markets seriously so I downgrade the Silver forecast a bit. Kaslshi has Ohio, Michigan, and Alaska as tilt Democrat. Polymarket Senate Polymarket, Nate Silver, and the polls, seem more aligned than Kakshi. For a long time Kalshi was ahead of Polymarket on many races, now it is is behind, overall. Polymarket Texas Senate Polymarket Ohio Polymarket Alaska Again, I will discuss Alaska separately, but both Polymarket and Kaslsi are high compared to all the polling experts and Nate Silver. Synopsis To win the Senate, Democrats need two of Alaska, Ohio, Texas, and Iowa (assuming they take Maine and North Carolina, and hold Michigan). Those are assumptions I believe very likely. Democrats have so many ways to pickup at least four while Republicans are spending money trying to make sure they don’t lose Kansas and Nebraska. Note that Vance is campaigning in Kansas. Excluding Nebraska and Kansas as extreme long shots, the least likely Democrat pickup at the moment is Iowa. But Trump has so seriously damaged his farm support, that even Nebraska is remotely in play. Iowa is a tossup or slight tilt Republican in absence of polling. Skew The skew is extremely one-sided. A very good night for Republicans would be a loss of only 3 seats. My base case is Democrats pick up 4-5 seats with a very decent shot at 6-7. Mentally parsing this together, I think the Silver Classic model at ~62 percent is about right, with a bit of a cushion due to Alaska to be discussed later. A sustained stock market decline or rise in the price of gasoline and diesel will make matters worse for Republicans. Related Posts September 14, 2026: Drone Damaged Saudi Oil Pipeline Down for 3-5 Weeks. Wait! How Long? Four million barrels per day are now offline, likely for months, not weeks. September 15, 2026: 3 Trump-Appointed Justices in 7-2 Majority Block His Mail-In Ballot Order So much for the “rogue judge” theory unless you now blame Trump for appointing them. September 15, 2026: 10-Year Treasury Yield Highest Since July 2007, 30-Year Since July 2004 That’s the highest in 19 years and 22 years respectively. September 15, 2026: Massie Reading Articles of Impeachment Against Hegseth on the House Floor We’ve only just begun with impeachment hearings. Cheers to Massie. Most valuable act by an opposition party would be to shut down funding to the criminal wars and terrorism, even if that means shutting down the government (including the military, of course). Impeachment? If they impeach without refunding the wars, then you know it is performative. Vance is campaigning in Kansas? Is he trying to tank the election for the Republicans? Six things that corporate media indulges in so they can avoid talking about anything meaningful: Everyone knows none of these things have any meaning at all or are newsworthy in any way. Truly, we are in the presence of a gifted mind. C’mon, Creamer. We all say a lot of obvious stuff to each other, especially if it is said partially to lighten the mood. I think that was the point of whirlaway’s “topic avoidance topic bingo card” entries. Comment system blocked my original phrasing. I was able to publish after removing “fake-news”. System is glitchy or else they’re starting to block not just URLs but also ideas. Speaking of my old buddy and pal, the ever lower than whale poo Ken Paxton, here’s the latest: https://x.com/Acyn/status/2079678270879260859/video/1 I’m a bit surprised that New Hampshire doesn’t come up in the assessments as a potential D loss. Sununu seems like a credible threat given his family name and NH being historically a bit of a wild card. Yet it seems to be tracked as a gimme for the Dems. Not sure that makes total sense to me. Iowa is red. Thankfully Mish is now saying it’s the least likely pickup. It isn’t happening. Yet it still appears on the 50/50 type poll. Farmers are not liking diesel prices but they’ll sure love corn at $5.xx to make them a lot of money, assuming it dries out enough to harvest their bumper crop. And they just aren’t pulling the d lever. The Iowa D levers get pulled in the college towns and urban DSM. Wait till they are paying $10 per gallon for diesel in a few weeks and having to get in line to get it Just got back from my visit to my hometown in Canada. Diesel prices were 2.55 a liter which makes it essentially $10 a gallon Canadian. They’ll blame Biden and the drag queens. Costco has started rationing motor oil due to a severe global shortage of premium Group III base oils, aggravated by geopolitical conflicts in the Middle East. The warehouse retailer implemented purchase limits alongside substantial price hikes on synthetic oils. A few more hissy-fits like the one today with the President threatening to stop the Kennedy Center reconstruction unless the judge allows his name to also be on the center, and yep, GOP turnout is likely diminished and Dems encouraged. My best hope is gridlock, too. 2026 should have been a better, or at least as not a serious seat loss. It’s been a “Death By a Thousand Truths” campaign season. (posts) Plenty O’Drama. But then there was 2016, and Nate Silver and Hillary Clinton. It all looked so good on the screen and in blog posts. That miss permanently altered my confidence in such projections, and in him. It reminds me of when the OJ Simpson “not guilty” verdict was announced. I was in a courtroom lobby, working, with many, shall we say, less than affluent members of the public present. A huge spontaneous cheer went up among those people, suddenly alerting me, with a chill, that what made sense to me and my cohort for what people thought, was way off base. Silver had trump at 25% chance of winning in 2016. I dont see how that assessment would reduce confidence in his work. He gave Trump a realistic chance at winning, even if he had him as underdog. We can only hope as gridlock is our only source of potential salvation over the next two years and potential impeachments. Regardless of the odds I’m still concerned that Mike Johnson will do what he says he’ll do. Simply refuse to acknowledge them into the House. He’s already done that with Adelita Grijalva (Ariz.). I don’t know that there’s any legal solution and the administration is counting on people being more concerned about gas and food than democracy Things that won’t happen for $1,000 Alex. Sheesh I invite him to try that genius idea, seeing as the people concerned about gas and food will probably respond by wheeling out a guillotine if they don’t get gas and food soon. Mike’s head is so far up Trump’s ass… he will continue to confuse his own thoughts… with those of the “stable genius”. Trump does not care if a guillotine comes for Mike Johnson… as you’ll recall… he didn’t care about the rope they brought for Mike Pence: ”In an interview with journalist Jonathan Karl, when asked about the chants to hang Pence, Trump did not condemn them. Instead, he defended the mindset of the crowd, stating it was “common sense” to be angry because Pence had not blocked what Trump falsely claimed were fraudulent election votes.”