Polling from the last few days offers some evidence that Kamala Harris is enjoying a modest post-convention bounce, with Nate Silver’s aggregate having her lead on the national popular vote out from 2.3 points to 3.8 points. However, Harris has taken a turn for the worse on Silver’s forecast model, on which Donald Trump is now rated a 52.4% chance of winning with Harris on 47.3%, restoring him to a marginal favouritism he lost at the start of the month. This is entirely down to state-level polling from Pennsylvania, where “it’s been a while since we’ve seen a poll showing her ahead”. However, Harris remains a 56-44 favourite in The Economist’s model, which has hitherto tracked Silver’s very closely. Adrian Beaumont has more at The Conversation.
Adrian Beaumont update at 2:08pm William’s link above referred to an article I published on Sunday. I’ve done another US article for The Conversation today which incorporates Nate Silver’s latest forecast.
Huey Long doesn’t loom all that large in my own explanations as to why Trump talks about elections being stolen. Let’s see what Randy Newman thinks.
https://www.youtube.com/watch?v=riOxEDVfkt8
That pic of Trump at Arlington is so weird. Seriously WTF with a thumbs up ????
how else are you meant to pose at a memorial / cemetery / grave.
I remember the complaining in 2016 over why the 538 predictor was giving Trump any chance of winning. How dare they! And similar in 2020.
Then everyone accused 538 of predicting a red wave in 2022 when they did nothing of the sort.
By all means have a go at predictions, adjust polling as you see fit with your reasoning (which is what most of the aggregators do anyway). But this revisionist take on Silver and 538 is nothing more than shouting at shadows.
538 and now Silver and the other aggregators and predictors are a good cross check resource when your feed is dominated by hyper partisan websites. You can chose to use it as a reality check. Or you can stick to your bubble. Up to you.
Thanks Mavis, just watched that link. There really is no comparison with Harris/Walz measured & intelligent response to interview questions & Frumps demented ravings. Even if the line of questioning seemed a bit lifted from GOP talking points.
Well said, TK.
The interviewer asks those questions to give Harris the opportunity to address those ‘GOP talking points’. That’s what Harris wants.
I think it’s quite fair to knock Nate Silver’s model when it suddenly goes agin all the others, on the stated basis that there hasn’t been a recent favourable poll for Harris in Pennsylvania, when there have in fact been several such polls.
I’m sure it will correct itself as he says, but it reeks a bit, particularly in light of what Victoria has pointed out about his new employment.
None of that should stop anyone from pointing out daily/hourly that Nate was right in 2016 of course.
Victoria @ #24 Friday, August 30th, 2024 – 9:09 am
He consults to them, on a fee for service basis, because he is very good at what he does, and they are prepared to pay for his undoubted expertise, in providing input to their potential ‘customers’.
Polymarket’s prediction market has taken in an enormous sum from punters betting on Trump to win before Biden dropped out. Such punters were counting chickens before they were hatched, without noticing that the hen-house was balanced on a rotten foundation. Polymarket is similar to Betfair, in that it allows anyone to take a position as a bookmaker or a punter, either laying or accepting a bet. Currently its market on the presidential election is almost exactly 50:50. Polymarket itself takes no position at all, just taking a cut of payouts on settlement.
I’m posting this here, even though it’s only tangentially related, because this thread is about an election. The video explains why it is impossible to hold a “fair” election. The guy who proved this was awarded the Nobel Prize in economics. (Australian viewers will smile condescendingly at implied outcome of ranked choice voting being that the candidates will sing “Kumbaya”.)
https://youtu.be/qf7ws2DF-zk
(23 minutes)
Assuming by right one means he gave Clinton a 60-odd % chance of winning and she lost
I thought Silver explained that his predictor had Harris dropping below 50% chance because the model was expecting a post convention bounce that it wasnt getting (at least at the time).
IMO, this year is different and I didnt think there would be a significant post convention bounce for Harris. If I am right, Silvers prediction model would drop for Harris but her polling average may well still increase as part of the general momentum she has garnered since Biden dropping out. It is no biggie. Silver was honest about how his model adjusts the polling. Right or wrong.
@yabba a tremendous proportion of the volume on these sites is people trading in and out of positions on each side of the bet – I myself have deposited $20 and it is already matched in $92 of bets (yes I’m a tremendous whale). If the market suddenly moves in my favour (due to for example Nate Silver announcing that Harris chance of winning the election has plunged) I might do it again with the same 20 (now 32) dollars, which as you point out makes money for the exchange.
Now this is nevertheless a bit conspiratorially minded, but as I said it doesn’t smell great when the actual data clearly doesn’t match up and old mate is employed by a betting exchange.
Bone spurs/cowardice.
Attacking Walz’s 24 year service in a uniform.
Dissing Congressional Medal of Honor winners.
Describing soldiers as losers.
Assault on an Arlington caretaker prompting public Army statements of support – for Team Trump’s victim.
Breaking the Arlington non-partisan rules. Arlington is about as close to a sacred site as you can get in the US.
Using a dead soldier’s grave as a media prop without the family’s consent.
Using a weird thumb’s up gesture at Arlington.
There are a lot of vets. Is the cumulative weight of Trump’s malignant contempt starting to make a difference? Or are the respective bubbles air tight?
Re Nate Silver. He backs his ability to analyse probabilities by occasionally entering poker tournaments. In 2024 so-far he has spent 7 days doing that and has earned $US6,500 per day spent. In the past he has had several $150,000 plus paydays. He made his reputation advising baseball franchises on player trades. He made many millions of dollars doing that, because they knew he was worth it, proven by results.
Not to mention he correctly predicted the 2016 and 2020 elections.
”
citizensays:
Friday, August 30, 2024 at 9:32 am
Trump’s happy snap at ANC. Why the thumbs up from Trump? It’s a weird photo – everyone looking so happy like they are at an amusement park instead of a solemn memorial to fallen soldiers.
”
Exactly. It is so weird. He is so weird. He was with weird bunch.
@Team Katich
Definitely not a biggie and whoever said “But this revisionist take on Silver and 538 is nothing more than shouting at shadows.” was being a total drama queen.
In my experience, Pollbludger has a much higher bar for that term. You may be confusing it with the term ‘turn of phrase’.
Unless you meant ‘total dancing queen’. Something I am more familiar with having spent some time in Sydneys inner East.
”
ScromoIIsays:
Friday, August 30, 2024 at 11:03 am
There was a recent poll in Ohio. Trump was up by 13 points.
Has the Harris/Walz evening interview started yet? Let’s pray that they don’t make fools of themselves.
”
oh my scromo!
Test after test after test for Harris-Walz. But Trump can say any racist, sexist about Harris and change position on any so called ‘policy’, it is ok.
People like you, meherbaba and others like Alabama, FUBAR, PP think that the current enthusiasm for Harris-Walz ticket is ‘too good to be true’ and a passing phase. They will mess it up any moment with one mistake and it will all come down like a house of cards.
There are some others, who are waiting with bated breadth, hoping everything goes well.
We all know American politics is broken and about half the electorate are either racist or cynical.
Here is a thought.
Maybe, just maybe a lot of US people are tired of Trump BS and Vance fakery.
Maybe, jus maybe Walz brought of breadth of fresh air to US politics.
Maybe, just maybe For the first time in a longtime they are witnessing honest and inspiring speeches from leading contenders.
Or
Maybe I am just reading too much into the situation. 🙂
I expect that all poll analysts will be having serious qualms about the validity of their weighting factors for their various demographic groups, in the light of such observations that “voter registrations of young black and latino females are up by 175%”. Do we expect that those that have felt moved to register may well be keen to vote? Any feeling about who they may be likely to vote for :)? Are any of the pollsters actually reaching these new registrants?
How should Nate and the other pundits adjust their algorithms? As a model builder myself, I really think that they have a big problem. History, in such cases, is quite probably bunk.
If they are honest about it, they could report the growing uncertainty of their results, and maybe start work on a new model with which to guide the pollsters towards getting better data. (It’s just a thought. Perhaps some are doing that already.)
Re: weighting factors/models: I’m sure Nate Silver doesn’t want to spend another four years explaining how getting it wrong makes him clever.
”
Omar Comin’says:
Friday, August 30, 2024 at 3:50 pm
I think it’s quite fair to knock Nate Silver’s model when it suddenly goes agin all the others, on the stated basis that there hasn’t been a recent favourable poll for Harris in Pennsylvania, when there have in fact been several such polls.
”
Is Silver prediction, analysis and judgement clouded regarding Pennsylvania because he is disappointed with Harris for not picking Shapiro?
IMO, Shapiro gives the impression of being another politician and Walz as ‘next door neighbour’. Harris already gives the impression of being a politician
I think a lot of US people are yearning for a non-politican and normal bloke.
Vance wants US people to think that he is bloke like Walz, Walz is.
Silver going slightly contrarian?
I say ‘Moo’ to the herding effect.
Just watched Harris’s interview. Thought she did very well.
The closest race? We are constantly told it’s reliably and solidly Republican.
Last week when I posted that 400,000 people registered to volunteer for Harris-Walz team, BW asked whether 400,000 is a big number in US context. Does anyone have an answer for that?
Have compiled the table below and in the words of rampaging Roy Slaven make of it what you will. My take is that 538 were reasonably accurate in forecasting results for Biden but not so with Trump by consistently under estimating their forecast for him.
2020 538 Polling 2020 Actual Voting Margin (Votes) 2024 538 Polling
23/8/2024
United States Biden 52.0 Biden 51.31 7,059,526 Harris 47.2
Trump 43.4 Trump 46.85 Trump 43.7
Georgia Biden 48.6 Biden 49.47 11,779 Harris 46.0
Trump 47.4 Trump 49.24 Trump 46.6
Pennsylvania Biden 50.3 Biden 50.01 80,555 Harris 46.6
Trump 45.6 Trump 48.84 Trump 44.9
Arizona Biden 48.7 Biden 49.22 10,457 Harris 45.7
Trump 46.1 Trump 48.91 Trump 44.4
Nevada Biden 49.8 Biden 50.06 33,596 Harris 45. 1
Trump 44.4 Trump 47.67 Trump 44.1
Wisconsin Biden 52.1 Biden 49.45 20,682 Harris 47.9
Trump 43.7 Trump 48.82 Trump 44.2
Michigan Biden 51.8 Biden 50.62 154.188 Harris 46.7
Trump 43.2 Trump 47.84 Trump 43.8
Minnesota Biden 51.9 Biden 53.40 233,012 Harris 49.3
Trump 42.7 Trump 45.28 Trump 40.3
North Carolina Biden 48.9 Biden 48.59 Harris 45.8
Trump 47.1 Trump 49.93 74,483 Trump 45.6
Florida Biden 49.1 Biden 47.86 Harris 43.9
Trump 46.6 Trump 51.22 371,686 Trump 47.8
Harris interview shows she’s ready—it’d be nice if the media could rise to her level
https://www.dailykos.com/stories/2024/8/29/2266697/-Harris-interview-shows-she-s-ready-it-d-be-nice-if-the-media-could-rise-to-her-level?pm_campaign=front_page&pm_source=top_news_slot_1&pm_medium=web
Kamala Harris spoke at a packed campaign rally on Aug. 29 in Savannah, Ga., after her first TV interview earlier in the day.
Ven: Compare and contrast Trump thumbs up photo at arlington with this photo.
“It’s been a Beltway media obsession for weeks, pretending that they’re the only ones who can explain to voters what the ticket is all about (despite doing what they do so poorly).
When CNN interviewer Dana Bash began asking for Harris’ “day one” agenda, the vice president spat out a bunch of policy proposals, and the response? “You were vice president for three years already, why haven’t you done it already.” Well, for one, she was vice president. It was right there in her question. But the question also feigns ignorance about the political situation in D.C. Bash knows the Republicans have run the House the last year and a half. She knows the Supreme Court has thrown out various efforts as well, such as student debt relief.
An honest interviewer needs to contextualize the political situation, otherwise it rewards obstructionism. Republicans don’t pay a price for their obstructionism, and the Supreme Court doesn’t get the criticism it so plainly deserves.
So how to approach it?
“What would you do if your party had complete control of Congress?”
“If Republicans retain control of any part of Congress, what could you accomplish in a gridlocked government?”
“The Supreme Court has made it harder for administrative agencies to issue rules. How does that change your view of your potential role as president?”
Those questions would be honest, give viewers proper context, and educate them about the political reality. Instead, people walk around wondering why government doesn’t work, and the Beltway media has zero interest in educating them.
Oops, that didn’t come out as I thought it would. There is supposed to be four columns viz. 2020 Polls, 2020 actuals, difference in votes and 2024 polls.
The betting markets had been at loggerheads since the conclusion of the Democratic Convention. A power struggle had taken place for favoritism between Harris and Trump.
Across the board between the betting agencies that matter, they were showing a dead heat:
Harris 50%
Trump 50%
I can report that there has been a slight shift in the market. The betting now reads:
Trump 50.5%
Harris 49.5%
I suspect that the next shift will take place after the debate.
At the moment, very close. If you were smart enough to ride on the support for Harris earlier and got the overs, it may be worth considering hedging your bets. But not yet, wait for the debate in my opinion, where I expect Kamala to win that contest.
@enjaybee can you do it as a picture?
Ven @ #78 Friday, August 30th, 2024 – 5:46 pm
8,000 volunteers per state. Quite a few more than would ever be involved here. Quite a few, no matter which way you look at it. Would be concentrated in key states, of course. Working on getting new registrations, helping people to cast early and postal votes, and getting people to voting booths on election day. Critically important, in the context of non-compulsory voting.
@enjaybee can you do it as a picture?
I’m a bit of Luddite when it comes to computer use. That table was from an Excel spreadsheet but I am quite happy to load it as a picture if you talk me through how to do it and more importantly would it work as picture any way?
On Nate Silver, I’m currently reading his latest book – On The Edge.
The first half of the book is on gambling – Texas Hold’em, Casinos, Poker Machines, Sports Betting. Nate opens the kimono somewhat on how the odds work. Spoiler, slot machines are the worst – but they make it fun whilst your guaranteed loss plays out.
The second half is on The Village – the establishment. And The River – high risk crypto and venture capitalism.
”
yabbasays:
Friday, August 30, 2024 at 5:59 pm
Ven @ #78 Friday, August 30th, 2024 – 5:46 pm
Last week when I posted that 400,000 people registered to volunteer for Harris-Walz team, BW asked whether 400,000 is a big number in US context. Does anyone have an answer for that?
8,000 volunteers per state. Quite a few more than would ever be involved here. Quite a few, no matter which way you look at it. Would be concentrated in key states, of course. Working on getting new registrations, helping people to cast early and postal votes, and getting people to voting booths on election day. Critically important, in the context of non-compulsory voting.
”
yabba
Michigan Comes Into Focus: Trump is losing his foothold and Harris Support is Exploding
https://www.dailykos.com/story/2024/8/29/2266491/-Swing-State-Michigan-Canvass
“Something is happening here. And what it is is exactly clear. Hope Springs from Field PAC [website] volunteers basically flooded the zone last Saturday, especially in the 7 Toss Up Electoral College Battleground states. Four states saw turnout of over 1,700 volunteers, including Michigan, which was second on Saturday in our highest volunteer turnout state.
“On Saturday, 1,768 volunteers came out to canvass in the swingy grey (and pink) districts (MI-3, MI-4, MI-7, MI-8 & MI-10).
Hope Springs volunteers knocked on 132,069 doors last Saturday and talked to 11,569 voters. 7,404 of those voters answered questions to at least part of the Issues Survey.”
“Volunteers registered 27 new voters and re-registered 77 voters in compliance with Federal (HAVA) law. ”
“Volunteers registered 27 new voters and re-registered 77 voters in compliance with Federal (HAVA) law. ”
@sprocket_ Hope you stole it
No doubt he’s a cryptobro and I’ll enjoy sledging him that little bit more now that I see he has stans. Apparently he has millions of dollars from baseball draft consulting(?) so I’m sure he is well insulated from criticism by internet proles such as myself.
Silver mentions that currently his model relies a lot on the calculation that PA is the tipping point state and that state relies heavily on one recent Emerson poll.
I’ve mentioned that while Emerson was on the money nationally in 2020, they did have very big issues in PA in 2022 senate race where they underestimated the Democrat by 6pts.
Sprocket
That’s interesting. I might grab a copy, Nate Silver – On The Edge.
The worst form of betting here in Australia is Horse Racing. The commission rates make it impossible to win and fund all the multi-million dollar races that the Industry programs.
Compared to a game of two-up for example, if you place a successful $50 bet, you keep $30 yet $20 goes to the Industry and betting operator.
Anybody reading this – you can do much better with your money than betting on horse racing!
My point about @6:21 pm post is that there separate volunteer organisations other than Harris campaign team working for Harris & Democrats election.
@enjaybee I’m not actually sure myself, you might need permission from the blog owner to post pictures.
Centre
Nate does put horse racing towards the bottom of the pile. The book is interesting for it’s Game Theory discussion and algorithm examples – as well as the ways both the house and the smart punters approach things. He does cover bookmaking on elections to some degree.
@enjaybee
Posting images to PB is done by inserting a link to the image. And that image needs to reside “on the internet” on a publicly available image server. Since PB doesn’t offer that you’ll have to choose another. I use “postimage”. People on PB can walk you through that process, but the first time it may be frustrating, and tie up the thread. (The open thread might be the place for that.) It might also depend on which combination of Microsoft, Apple, or Google products you use.
The good news is that once you’ve got the tools it doesn’t take long.
This one took less than a minute. The bad news is the first time is a bit frustrating.
Cool Sprocket, I will give it a read…
Ven @ #87 Friday, August 30th, 2024 – 6:21 pm
Interesting. Thanks Ven.
I expect this is significant. I wonder how it was in 2020.
It is a bit hard to think.that Harris will not win Pennsylvania. But win the presidency.
See electoral. vote.com.
This election is incredibly close. I don’t think it’s going to be possible to even have an inkling of who’s the likely winner even on election day!
Frankly I find it astounding that Trump has the support he has. The media double standards that apply to him are also baffling. America is a very strange country.
@ late riser
I’ll give this a miss for the time being. My son will probably be able to help me the next time I see him.
No worries, enjaybee.