Liberals’ Knives Come Out for Nate Silver After His Model Points to a Trump Victory
Silver: “Harris is in fact on the decline in polls over the past couple of weeks in most of the key swing states.”
And just like that, after suggesting that Vice President Kamala Harris’s honeymoon phase is over, polling guru Nate Silver went from a liberal darling to a charlatan who works in cahoots with Silicon Valley billionaire and past Republican megadonor Peter Thiel (regardless of the fact that Thiel has repeatedly said he is sitting out the 2024 election).
Silver, who publishes election data in the widely followed “Silver Bulletin,” reported a significant shift in the presidential race. His latest model showed former President Donald Trump’s odds of winning at 62.5% compared to 38.5% for Harris. He noted, “Harris is in fact on the decline in polls over the past couple of weeks in most of the key swing states.” Stacey Matthews posted on this story over the weekend.
One of the reasons I'm not super sympathetic to complaints about the convention bounce stuff is that Harris is in fact on the decline in polls over the past couple of weeks in most of the key swing states.
— Nate Silver (@NateSilver538) September 6, 2024
I'd also note that Harris's raw polling averages have DECLINED in most swing states since the start of the DNC. This data is NOT subject to the convention bounce adjustment. She's had a run of pretty mediocre state polling. pic.twitter.com/smiCFiznQW
— Nate Silver (@NateSilver538) September 6, 2024
It’s important to note that Silver is a Harris supporter, as he tells his followers in the post below.
My subjective view is a little sunnier on Harris than the model but also I’m planning on voting for her so you should probably discount that.
— Nate Silver (@NateSilver538) September 6, 2024
More bad news for Democrats followed on Sunday morning. A New York Times/Sienna poll showed Trump leading Harris nationally by a margin of 48-47. Factored into Silver’s model, this poll pushed Trump’s odds of winning slightly higher still to 63.8%, compared to Harris’s 36%.
According to The Washington Examiner:
The former president is also favored to win every swing state.
Silver’s current odds give Trump a 64% chance of winning Pennsylvania, 54% for Michigan, 53% for Wisconsin, 77% for Arizona, 75% for North Carolina, 68% for Georgia, and 61% for Nevada.
The new probability total is a nearly five-point boost for Trump since Thursday, when he was given a 58.2% probability, itself a boost from the 52.4% a week prior. The prediction shows a further eroding of Harris’s honeymoon support since President Joe Biden dropped out in July.
In a Substack post, Silver explained that the new poll was such bad news for Harris due to the large sample size and reliability of the poll, which he ranks as the second best. He said the new poll “confirms the model’s view that there’s been some sort of a shift in momentum in the race.”
Needless to say, Harris supporters, who had grown accustomed to her poll numbers moving in only one direction for six weeks, and tried to promote the narrative that her victory was inevitable, were outraged by the reversal in Silver’s model.
It took about a nanosecond for their mutiny to unfold. Their knives were out. Their campaign to discredit their former favorite pollster by tying him to Thiel, the co-founder of PayPal and Palantir Technologies, had begun.
Super important—do not trust Nate Silver’s polling. He is financed by Peter Thiel. https://t.co/M8cxmlb0M1
— SunningChameleon (@SunningCham88) September 8, 2024
Nate Silver now works for Polymarket, which is owned by Peter Thiel. He constantly includes garbage high school republican polls like “patriot polling” in his aggregate models
For all intents and purposes he is acting as a Republican op & should be regarded as such https://t.co/hreJPzptl8
— Leftward Bound 🌴 🥥 🇵🇸 🇺🇦 (@LeftwardBound) September 8, 2024
Soon I expect Nate to predict Trump will receive every last electoral vote
Cuz that’s the way Peter Thiel wants it
Nate Silver works for Polymarket, a Panama based betting mkt that offers odds on US elections. Peter Thiel is a maj investor. US citizens are barred from betting https://t.co/hW7QPSzhT8
— f(x) (@ef_of_eks) September 8, 2024
Nate Silver is funded by Peter Thiel who also funds JD Vance. You know the same way you are funded by Russia comrade Ben https://t.co/HlXZgXUH2h
— Jack Starr (@Jckstarr) September 8, 2024
I respect the naked partisan hack branch of Nate Silver haters than whatever this shit is pic.twitter.com/WDRJiZiWh6
— Pontifex Maximus-ComradeSnek☭ 🐍 (@ComradeSnake) September 8, 2024
.@Polymarket, Peter Thiel and Nate Silver. A very possible scenario brewing… pic.twitter.com/fEowRcSkQT
— f(x) (@ef_of_eks) September 7, 2024
Of course no one knows who will win the election. I gave up trying to predict election outcomes after the 2022 midterms when I was forced to delete the rough draft of a post I’d started in the hours before reality struck. My title had been, “How Sweet It Is.” Yes really.
That said, Harris’s gravity-defying momentum, much of it the result of Democrats’ tremendous relief that the spavined, senile President Joe Biden had left the race, has peaked.
For both candidates, much is riding on the outcome of Tuesday night’s presidential debate in Philadelphia.
During a weekend interview, Speaker Emerita Nancy Pelosi absurdly asked journalist Kara Swisher if she thought Trump would even attend the debate. Swisher said she did and asked Pelosi if she knew something Swisher did not.
Pelosi’s response? “I know cowardice when I see it.”
At #TribFest24 @karaswisher asks @SpeakerPelosi if she has advice for Trump ahead of the debate.
Pelosi: "Do you think he's going to show up?"
Swisher: "I do. Do you know something I don't know?"
Pelosi: "I know cowardice when I see it." pic.twitter.com/4UHjygR2xc— Texas Tribune (@TexasTribune) September 7, 2024
Pelosi’s random remark aside, all eyes will be focused on the Philadelphia debate stage tomorrow night. And as Bette Davis famously uttered in the 1950 film “All About Eve,” “Fasten your seat belts, it’s going to be a bumpy night.”
Elizabeth writes commentary for The Washington Examiner. She is an academy fellow at The Heritage Foundation and a member of the Editorial Board at The Sixteenth Council, a London think tank. Please follow Elizabeth on X or LinkedIn.
DONATE
Donations tax deductible
to the full extent allowed by law.
Comments
How do you know a Hail Mary is coming on third and long?
When Nate Silver tells the truth.
Frankly, if Silver predicts a Trump win, if I were a Trumper that would scare me. LOL
Silver’s claim to fame is that he called Hilary as *only* a 2:1 favorite instead of a lock.
I’ll take it but not at ease over 60% odds.
How do we know he isn’t shilling Trump as an overwhelming favorite so his base decides to stay home on election day?
Operation Demoralize can work both ways.
Energised voters will come out to vote regardless of what this guy is saying. I think some of you have lived too long in Democrat laaaaalaaaaa land 🤣🤣
Leftists have always been some of the worst “Shoot the Messenger!” types. It’s an innate part of the Western Leftist psychopathology.
I’m sure that more than a few democrats are talking about having Nate Silver arrested for “misinformation” and “colluding with Russia to alter the 2024 election”.
But Putin endorsed Kumella…
Has any candidate EVER been endorsed by BOTH Cheney’s AND Putin?
This woman is historic and knows no limits.
She is far more predictable than Trump. Of course Putin endorsed her. Putin knows she will obey the instructions of the foreign policy Blob. She will not freelance. Well, the Blob runs the Cheneys too.
He didn’t account for voting fraud, which points to a Harris victory regardless.
Voting fraud is an intentional feature of our voting system.
Hence why Trump has to win outside the margin of fraud!
The X-commenters hanging on the ‘accuracy’ of the 2022 polling is rich. Midterm polling generally seems to be more accurate because the favorites – incumbents – frequently prevail. But, presidential polling for the last 25-yaers has been not very reliable. In all but one election cycle – 2012 – Democrats have under-performed ‘predictions.’ The chances that’s a coincidence are slim.
Democrats started systematically outperforming their polls since Dobbs. This is the very first post-Dobbs Presidential election. We have no idea what will really happen.
Lol. The final RCP average for Ron DeSantis in November of 2022 – five months after Dobbs – had him winning by 10-points. He won by almost 19-points. Rubio was only supposed to by 8-points. He won by 16-points. Vance in Ohio was only supposed to win by 3-points. He won by 8.5-points. DeWine in Ohio was supposed to win 14-points. He won by 20-points.
Polling was all over the map in 2022….and it was better in 2022 compared to 2020.
“Pelosi’s response? “I know cowardice when I see it.”
Come on, Nancy, you don’t even know a gay tryst when you see it.
LOL. Good one.
The kool aid drinkers on the left need another dose b/c it seems to be wearing off. They had a flipping disaster on their hands with Biden candidacy. They tried damned hard to convince everyone, including themselves, it wasn’t so. Then the 1st Presidential Debate happened. They panicked b/c the folks who KNEW had been falsely reassuring those who suspected how bad Biden actually was. A soft coup occurred in party. Biden jammed up any hope for an open convention by throwing his weight to Harris out of spite/anger IMO.
Now they are stuck with a nominee who received no support and offered no policy positions during the d/prog ‘non traditional’ ’24 primary. The convention was an anointing in part and sort of a ‘put on a happy face’ feigned enthusiasm to mask their reluctance/trepidation. Harris is not a good politician; not retail in person, not wholesale speaking through a camera.
The d/prog are stuck like Chuck and tomorrow night could be the final nail in the coffin. Assuming that Trump can for 90 minutes just do what he did with Biden and let Harris destroy herself by staying out of the way. Just stay in his own lane and let her fail tomorrow evening and he almost certainly wins over the electorate IMO.
Exactly! I hope Trump just answers the questions posed by the moderators and ignores Que Mala, except to fact check the BS lies (“very fine people,” “inject bleach,” etc. Some people still do believe that crap, so he should respond in a lighthearted manner–ala Reagan’s “there you go again”–to that stuff but for people at home, not her). Let her babble nonsensically and that’s it. Game over.
With a dash of ‘WRONG’ for flavor if Harris gets out of hand.
Fuzzy, your post is just gibberish. Trump needs to focus on the economy, the border, and our military strength against China and Russia. He has to stop going on and on about personal attacks that drive away Independents and Undecideds. This election is for Trump to lose, and he is doing everything he can to lose this election. Why does he keep attacking Harris for not being Black, for giving blow jobs to politicians (when Trump had sex with porn star prostitutes like Stormy Daniels, and is convicted of sexual abuse, and is still bragging about grabbing women by their pussies). Trump is undisciplined. Someone in his campaign needs to grab him by the balls and say STOP THIS!
So, lonejustice…. it’s NOT ok for Trump to say “grab em by the pussy” (which isn’t even what happened), yet it’s ok for YOU to say someone needs to “grab him by the balls?”
I mean… the jokes just write themselves, usually with you as the punch line.
The rest of your post is just your usual compendium of hot garbage, lies, verbal diarrhea, and flatulence.
I suggest you start posting over at Democratic Underground. They’re more your type.
Being gay, I found the “grabbing by the pussy” remark irrelevant. I think I will stop there. LOL
The only people attacking Harris about not being black are Democrats blaming everyone else for attacking Harris for not being black 🤣🤣
@JR, she called it on the Charlottesville lie
You are right – tomorrow night is incredibly important. But I still expect the Democrats (or their foreign allies or cheerleaders) to do extraordinary things if she blows it tomorrow. Particularly Iran. They do not want a Trump II, and may be willing to take great risks to prevent it. Perhaps risks as high as launching a new pandemic.
I will make a stunning prediction: The NYT, WaPo, CNN, AP, et. al. will declare Harris the winner. Everyone here should expect it. The silver lining is that, three years ago, Reuters surveyed people in 46 countries about their media, and the U.S. came in last. No one believes them, so it won’t matter who they crown.
This isn’t the worst thing. I think Americans have slowly but relentlessly concluded that they cannot trust the media, and as a result more people are deciding for themselves. Long overdue, but it’s happening. The media’s stock in trade is trust, and once that’s lost they will have one hell of a time regaining it.
https://www.poynter.org/ethics-trust/2021/us-ranks-last-among-46-countries-in-trust-in-media-reuters-institute-report-finds/
Three preludes to my comment, which will be long and well-researched, with links.
First, I am a “double hater” who will cast his third straight write-in vote this year. I live in WA State, so the result is foreordained, which leaves me free to indulge in puerile cynicism. Second, I am a former journalist, when that was an honorable undertaking, as in “start with the facts, progress to reasonable inferences arising from them, and if you are an editorialist, base your opinion on the former. Otherwise, keep your opinions out of the story.”
Third, I am a retired financial analyst (two careers) who lives and breathes the numbers. Fourth, I have been following both the economy and the polls closely all year.
Now here we go.
1. Nate Silver is useful for only one thing: His pollster ratings that show partisan lean. Whenever I see a story that cites a particular poll, I check Silver’s pollster ratings. I don’t think his predictive model is worthwhile, and regard his political commentary and analysis as pedestrian. But the pollster ratings are golden. Scroll down at the link below.
https://www.natesilver.net/p/pollster-ratings-silver-bulletin
2. To gauge the national polls in the aggregate, and their trends, I start with the Real Clear Politics average. This year, I adjust it for Trump’s overperformance in 2016 (2.5%), and 2020 (2.9%). RCP shows that Harris’s unadjusted average peaked at 1.9% on August 31, and now stands at 1.2%. The narrowing is because Trump is gaining, not because Harris is declining.
https://www.realclearpolling.com/polls/president/general/2024/trump-vs-harris
As I just noted, I adjust that average for Trump’s outperformance in the last two presidential elections. At Harris’s recent peak, I considered Trump ahead by 1.9 minus 2.7 (the midpoint of the two other years), or 0.8%. Today, I think he’s ahead 1.2 minus 2.7, or 1.5%.
https://www.realclearpolling.com/historical-polling-archive
3. State polling is notoriously unreliable, and I refuse to make it up. This is frustrating for obvious reasons, but I think if Trump wins the popular vote he will be elected. That much, I think I can say with as much confidence as I can muster.
4. Ever since World War II, with only one exception, if the national unemployment rate rose in the second quarter, i.e., June was higher than March, the incumbent party’s candidate lost. The exception was in 1956, when unemployment rose slightly (by 0.1%), and was not confirmed by other economic indicators.
Conversely, if the unemployment rate fell in the second quarter, the incumbent party’s candidate won in every such case. The reason this works is that unemployment is a reliable proxy for the state of the economy, and Americans vote their pocketbooks before anything else. Forget about the stock market. Voters don’t spend the S&P 500.
This year, unemployment went from 3.8% to 4.1% between March and June, and then to 4.3% in July and back to 4.2% in August. This increase matches a long list of other weak indicators all year. Thus, my belief since June has been that this is Trump’s election to lose.
https://data.bls.gov/timeseries/LNS14000000?years_option=all_years
Can he lose? Of course! To quote the disclaimer in mutual fund advertising, “past performance does not guarantee future results.” Trump’s ham-fisted personal attacks on Harris have been, even for him, just plain stupid. His running mate selection was subpar; had he chosen N.D. Gov. Doug Burgum, and been disciplined in his commentary in the past five or so weeks, I think he’d be running away with it.
BOTTOM LINE
I expect Trump to be elected, but not by much of a margin. He very well could lose with a bad debate on Tuesday. He should talk forcefully about the economy (unemployment and inflation, especially the latter), illegal immigration, and Harris’s lack of performance in office. Will he do that, or will he play Archie Bunker Trump, the Rodeo Clown From Queens? We shall see.
Good analysis Random Crank. I think Trump is well aware of what he needs to do to win. He performed well in the debate with Biden – even when Biden called him a convicted felon and a rapist. … Regarding state polling, I’ve read that there has been less of it this year. I don’t know whether that’s true or not. … My career has been the mirror image of yours. My first real job after college was in financial services. And I began blogging in 2018.
My error about three preludes rather than four notwithstanding, the finance career taught me both the strengths and limitations of numbers, but especially not to get them wrong. (In the case of my miscount above, there’s no edit function here, or I wouldn’t have goofed.)
Right now, it’s a tie ballgame, within the margin of error. The potential tragedy for Trump is that it truly didn’t have to be this way. He knew right away what he needed to do, but he just couldn’t control his Outer Borough combativeness.
He wasn’t wrong about Harris, but a presidential campaign is about a lot more than just being right. It’s even more about how you make your case, and when. His blunders are indicative of why I have never voted for him. Trump is deeply insecure, which by itself is not a disqualifier, but his failure to control himself is a character issue that I cannot get past.
Harris’s character is no better, so between that and a range of D policies, she’ll never get my vote either. The Dems need to reanimate Harry Truman, and the Rs need to dig up Dwight Eisenhower. American history shows a bunch of cases where one dominant party has melted down, but not until now a time when both of them did.
No matter who is elected, this country will be worse off.
I was with you until Eisenhower. Not a fan.
For several reasons, but just to give one, he was the worst president for Israel of all, including 0bama and Biden.
As much as I favor Israel over the Arabs, and wish them well, I think they are quite far down the list when I evaluate presidential administrations.
Three preludes, I wrote, and gave four. Some numbers genius I am. Oops. LOL
@Random Crank, I do not agree with all of that but appreciate the details.
Archie Bunker Trump showed up at the debate. And he may have won or at least a draw.
I wish Silver and others would state their results in terms of odds. The odds ratio is the probability of success divided by the probability of failure. So if Trump’s win probability is 62.5% then his win odds are 1.667 (rounding). To understand why odds are a better think of studying for a test. It takes a lot more studying to raise your score from 94% to 95% then to raise it from 60% to 65% even though it’s five points in both cases. This is a general effect. A lot of software projects fail before completion because those last few points are the most difficult. That’s why many contractors will walk off the job (if they can) at 95% complete because they know the last 5% if the most difficult in many cases. So don’t pay your contractor until the job is 100%. Odds ratio is also the cost of a fair bet. If that 62.5% Trump win probability is correct then you should pay no more than $1.67 to win a dollar betting on Trump. If we want to track polling look at changes in the odds ratio.
All that being said, I don’t take polls seriously until mid October. Even then I don’t take them very seriously because of response bias and cheating. For this election, the main purpose of the polls is to explain a Harris win when we know the Democrats need to cheat to win with this idiot.
A distinction without a material difference, IMO. People who need it put in terms of odds can easily do it if they want to. Yes, it harder to grab the last 5% or 10%, but so what? It’s an election, not a software project.
Probabilities bunch up at the end points making for a very non-linear scale except near the center. The odds ratio re-expresses probabilities on a scale from zero to infinity. Much better behaved for visual interpretation. Sure one can calculate the odds, but that’s tedious if a number of polling results are presented as a table or a graph. Trends and differences are easier to interpret when presented in terms of odds instead of probabilities. Often a transformation is helpful as with some financial data where a log scale is more revealing than a linear scale.
President Trump will win. All he needs to do is keep his cool and be respectful in the debate, but, point out, factually and fairly, that the U.S. middle class cannot withstand another four or eight years of crushing Biden-Harrisflation. Harris has been in office for almost four years; she’s done nothing but sow impoverishment, an illegal alien invasion, higher crime, and, emboldened/enabled malignant and despotic regimes and terrorists.
Agreed, but this was obvious after Senile Joe dropped out and Kamalatoe replaced him. Trump went right back into Insult Monkey mode, which worked against the Hildebeast and forced Joe out, but has already backfired against Harris. We shall see.
Better Nate than lever… 🙂
Could be that Silver has started to adjust his model for Trump’s overperformance of polling in prior elections.
Wasn’t that long ago Nates 538 was the left wing bible.
He’s still gonna vote for Harris and publicly admitted it. He does PO the wokiestas b/c he puts integrity as a priority in his analysis of the polling data trying to be as accurate as possible and he tries to point out the shortcomings and potential for error in his polling and analysis… Like the probability of Trump over performance due to folks not being willing to talk to polls and the polling outfits being unable find the ‘hidden’ Trump supporters.
He’s a lefty but at least is upfront and honest about it.
There’s a lot of hidden Harris voters. They just don’t know it, and probably never will.
“ During a weekend interview, Speaker Emerita Nancy Pelosi…”
There is NO SUCH TITLE AS “SPEAKER EMERITA!”
Silver used to be a baseball statistician.
His problem when he first started 538 was that he assumed all the polls were equally valid, no matter how badly they oversampled one Party, and just did the math on the probable outcomes of the election as if the polls were entirely accurate.
So, he was wrong. A lot. Typically in favor of D’s, because most of the polls are run by lefty orgs that oversample D’s deliberately to make it seem like they are doing better – helps with the D’s fundraising and voter turnout.
He has gotten much, much worse. Now, he weights the polling firms and entirely excludes some (all of those excluded are those that are not run by lefties). He hides his weighting system, so no one has the slightest clue what he’s doing.
This makes his results favor D’s even more than they used to.
So, if he’s saying Trump has a 60% chance of winning, then you could bet your life savings that Trump is actually up far more than that.
Funny that libs are attacking him though, despite knowing that he’s a reliable life-long liberal Democrat. For them, everything is a litmus test and you can’t stray off the plantation even for a moment, on any issue, without being shunned.