Nate Silver is now giving Trump the edge of winning.
As a result, the forecast is still in toss-up range, but Trump’s chances of winning are his highest since July 30. And the chance of an Electoral College-popular vote split working against Harris has risen to almost 18 percent.
Yeah, it seems like something really has to go wrong at the debate for Trump for there to actually be any separation or better than toss-up result.
I don’t see the polls doing anything but shifting from leaning one way to the other until Election Day.
Well, fingers crossed that he melts down at the debate and (somehow!) undecided voters actually make up their damn mind. Eight or nine weeks of this knife edge is not healthy for me.
I take what Silver says with a shaker of salt. I think the memory of 2016 has spooked the hell out of him and he has determined that the benefit of every doubt in his projections is going to go the way of DJT.
I think we have to take all the polls with a grain of salt and accept that they just might not be working.
Silver fully admits that his model is suppressing Kamala’s numbers post convention. The model’s post-convention dampening effect fades out after about three weeks – just in time for the debate.
8/31/2024: One reason the model has moved so much on new Pennsylvania polling, for instance, is because since earlier polling occurred before what the model considers to be two “landmark events” — the convention and a candidate exiting the race — so the model looks at older polling as being really out of date. Might it be overdoing things a bit? Sure. But I don’t think there’s anything inherently wrong with how it’s handling these circumstances, either, and it’s also naturally self-correcting (e.g. if/when Kamala Harris gets some better Pennsylvania polls, her numbers will improve).
9/2/2024: I’ll work through what will happen if Kamala Harris holds her current position in polls. Hint: she’ll eventually become the favorite again.
(my emphases above)
I think there’s something to this. Could help explain the difference between Silver’s current model at Silver Bulletin and G. Elliott Morris’s current model at FiveThirtyEight.
He addresses that in the link:
The model’s convention bounce adjustment is one thing — and something you’re welcome to debate — but the fact is that most polls we’ve seen over the past week are coming in below our current polling averages for Kamala Harris. (Our polling averages don’t make any convention adjustments — those come at a later stage of the model.)
For instance, the latest YouGov weekly poll (Harris +2 nationally) or this one from a high-quality pollster in Michigan (Trump +1). This batch of CNN polls was better for Harris — but not in Pennsylvania, where CNN showed a tie even though the poll went into the field the day after the DNC ended in what should have been the post-convention afterglow period for Harris.
Or that they work within their reasonable limits to inform us. We may be looking to polls to do something they simply have never been capable of reliably doing.
Fair enough, especially now that we are at a point when it isn’t possible to have a runaway win. As I said before, we’re two nations fighting over control of one government.
I think all of this hand wringing over polls is misplaced, for several reasons.
One, polls themselves are mere snapshots of the electorate, and have notable biases. Clearly, they aren’t reaching an entire cross section of the voters, and may be missing out on new voter enthusiasm, or shy supporters (perhaps Republican women will break for Harris). The only true poll that matters is the one that occurs on Election Day.
Secondly, these forecasts are based on a massaging of these flawed poll numbers, by prognosticators with a bias, unconscious or not, towards a more conservative outlook; better to err on the side of a close race. Although it’s covered by numbers and statistics, this polling stuff is really just guesswork.
At best, they reflect a trend, that the wave of enthusiasm for Harris has softened, as all of those new supporters have already been added into the equation. But her lack of gains in recent weeks isn’t necessarily weakness.
Thirdly, there are certain inflection points to come which I think will have significant influence on the outcome of the election. The debate is the first one: if Trump appears tired or angry and bitter, it’s going to turn people off. There’s also that movie I mentioned upthread, which might get some buzz as it depicts Trump as a loathsome asshole. His DC indictment is moving, with Judge Chutkin holding hearings. Even if people don’t agree with, or understand, the underlying charges, Trump is being tainted with scandal. NY’s senfencing might happen this month, too.
Obviously, Harris or Walz could hit a snag; Walz has MAGA family members, so that could get ugly.
But Harris is focused on policy. She’s talking to people about their problems and their future. Trump is just talking about Harris, or about himself.
None of this is to say that the election is over, or in the bag. But there is ample room for optimism at this point, even as we are in the thick of the fight.
All right – RealClear Polling came through. They’ve now got the Rasmussen Reports Trump +1. And here’s the direct link (partially paywalled, but you can see enough at the top). Rasmussen actually sampled 1,838 likely voters, so Newsweek garbled the sample size somehow.
Meanwhile, Emerson now has Harris over Trump nationally by 3.7 points (Google Docs – scroll down to rows 35-40). And Outward Intelligence now has Harris up 6 points against Trump with third-party candidates included.
Very good post, thank you!
The polls are indeed a useful but flawed lens through which to discern what’s going on, full of grease, dust, and gnats. Probably a lot greasier than in days past…
Here are two possible perspectives on how Harris is doing:
My perhaps wishful thinking version:
Whoa, good news:
Trump brought in a hefty $130 million in August. Pretty good, right? (Though that was actually down from $138.7 in July.) But get this: Harris brought in $300 million.
(Michael Bluth voice) “You didn’t say ‘away.’”
But I may need to tune out of this thread and all poll-related stuff. I saw a tweet recently that said something to the effect of “All the pundit class says it’s really close with Trump a bit ahead. Talking to real people, Kamala is burying him.” Somewhere in the middle is the truth, I’m sure. I’m clinging to the hope that the younger cohort just aren’t answering pollsters and are revved up enough to surprise everyone on November 5. But I’m living in a permanent state of cold sweat until the results come in.
One factoid that gave me comfort was hearing news that a million new voters registered… as soon as Harris entered the race.
Something tells me they weren’t moved to do so by Trump.
Are you referring to the below? It’s reporting that vote .org says they have registered 1 million new voters this cycle (not just since Harris entered the race). They do report that they’ve registered 376k new voters since Biden withdrew.
It might have been that and I either forgot, misunderstood or got it garbled in my head since it was reported. Still, that’s quite a lot of new folks coming in to participate in the election.
Thanks for the cite!
Whoops, haha.
Most of the people I watch on YouTube seem genuinely positive about where things are. Also, Trump himself does not seem overly confident. (Compare his demeanor now to when he was beating up on Biden.)
That’s my guess. E.g., registrations of young Black women are up 175% versus 2020. The fundraising numbers suggest huge enthusiasm. And so on.