The Economist hasn’t yet returned to the party with a full prediction model yet, but it does have a poll tracker showing Kamala Harris leading Donald Trump by 47.8-45.0 on the national popular vote. The accompanying trend chart is of particular interest in showing the race with a discontinuity between Biden-Trump and Harris-Trump, such that the data for the latter is all from the past few weeks and not a long-term series working of hypothetical match-up polling like other such efforts I’m aware of. What is shows is that Harris was immediately at parity with Trump after the withdrawal of Biden, who trailed by about three points, and that this reflected a bigger drop in the neither-of-the-above vote than Trump’s. The three-point lead Harris is credited with now was opened up over a fortnight of favourable momentum that has tapered off over the past week or two.
Nate Silver’s model now gives Harris a 56.0% chance of an electoral college majority with Trump at 43.5%, a gap that has been slowly widening over the past week or two. The only states it now records as more likely than not to flip to Trump are Arizona and Georgia, with Harris gaining favouritism since the model was launched a fortnight ago in Wisconsin, Michigan, Pennsylvania and Nevada — the difference between her losing the Electoral College 312-226 and winning it 276-262. Any one out of Wisconsin, Michigan or Pennsylvania would tip the balance back to Trump.
Adrian Beaumont’s freshly minted thoughts on the matter can be found at The Conversation.
Ok, thanks citizen.
Can Kennedy trust Trump?
If Kennedy sucks up to Trump really hard, I think he can trust Trump. Trump loves it…
It seems like the Republicans are desperate so they’re calling the stooge RFK in early. Apparently his advances toward the Democrats were met with ridicule so he went on a trichinosic Twitter rant against the evil bastards.
If Trump wasn’t such an insecure loser they could’ve let this play out for longer.
Mavis
“I think BTsays/BT is rooting for Trump.”
You can think what you like, and whether you are right or wrong makes zero difference to this blog.
But in case you or anyone else needs a reminder, BTSays and BT are two completely different people.
https://www.pollbludger.net/2024/08/15/us-presidential-election-minus-12-weeks-2/comment-page-11/#comment-4354000
YaramahZ – as stated earlier, Tennessee was the state with the biggest margin of votes for the GOP in 2020, not any of the ones you cited. FWIW.
Based on the available poll breakdowns RFK draws votes from Trump nationally but he draws more votes from Harris in the crucial battleground states. But that is 2% in polls with 600 respondents so well within the margins.
————————————-
Good points. Even when state polls have higher sample numbers they still seem to be less reliable than national polls with 1000 sample. It is why predictors like 538 dont rely on state polls to come up with the chances of overall EC victory. So trying to drill down into what may happen with RFKs 5% or so in state polls of, yes, usually a 600ish sample is fraught.
And who knows how many of the people who polled for RFK would actually have voted for him anyway, or are actually capable of getting to a voting booth on the correct day and can remember their own name.
Having said all that….. it would suck if Kennedy withdrew and endorsed Trump. But not all that surprising. He can be 2024’s version of 2016’s Chris Christie.
https://www.pollbludger.net/2024/08/15/us-presidential-election-minus-12-weeks-2/comment-page-11/#comment-4354077
Ven – I find that very interesting.
But what would make it more interesting would be how this compares to previous conventions for both parties. Don’t suppose you could find that out, could you?
I mean, if the DNC always has more viewers even when they lose then it means nothing. But if they normally have fewer viewers than the RNC then it’s really significant. For example. Plus overall numbers watching compared to previously gives some idea of voter engagement with the political process currently.
https://www.pollbludger.net/2024/08/15/us-presidential-election-minus-12-weeks-2/comment-page-11/#comment-4354091
Confessions – yes it will help Trump if Kennedy drops out, especially if he endorses Trump.
What I’ve no idea is where it will help Trump much, and if there’s any state where it will help Harris more – I assume not though some may be a neutral effect.
“I think it doesn’t matter now whether RFK stays in or drops out – the 5% left are cooker voters who will stay home or vote for another loon.”
Well, except for the fact that Trump would be that “another loon” so it does matter.
BT, my apologies for comingling you with BTsays.
BTSayssays:
Wednesday, August 21, 2024 at 8:57 pm
Mavis
“I think BTsays/BT is rooting for Trump.”
You can think what you like, and whether you are right or wrong makes zero difference to this blog.
But in case you or anyone else needs a reminder, BTSays and BT are two completely different people.
================================================
This sort of confusion never occurred when “Gympie” was still here.
RFK has already unleashed the full worm upon on the Democrats so unlikely he’ll be endorsing them.
Mavis
Not a problem! – was just a friendly reminder tbh.
“Based on the available poll breakdowns RFK draws votes from Trump nationally but he draws more votes from Harris in the crucial battleground states. But that is 2% in polls with 600 respondents so well within the margins.”
I’ve just looked at 538 for all 7 states and just can’t see the evidence of that. I mean, one could cherry-pick individual polls but it’s basically swings and roundabouts all over the place. Perhaps Trump would benefit slightly more in Pennsylvania than elsewhere, but maybe Harris somewhere else. But it’s too little anywhere, and fraught with too many caveats and not enough polls, to suggest a clear advantage for either purely through Kennedy dropping out.
Dropping out. . . AND endorsing Trump. . .might make a little more difference. But frankly he will be largely forgotten a month after he drops out. Still it is likely to be another election of fine margins so any little difference it does make is not to be sniffed at I suppose.
Meidas Touch’s take on today’s DNC:
https://www.youtube.com/watch?v=3NcwdL25XRo
BTSays says:
Wednesday, August 21, 2024 at 9:22 pm
“Based on the available poll breakdowns RFK draws votes from Trump nationally but he draws more votes from Harris in the crucial battleground states. But that is 2% in polls with 600 respondents so well within the margins.”
I’ve just looked at 538 for all 7 states and just can’t see the evidence of that. I mean, one could cherry-pick individual polls but it’s basically swings and roundabouts all over the place. Perhaps Trump would benefit slightly more in Pennsylvania than elsewhere, but maybe Harris somewhere else. But it’s too little anywhere, and fraught with too many caveats and not enough polls, to suggest a clear advantage for either purely through Kennedy dropping out.
Dropping out. . . AND endorsing Trump. . .might make a little more difference. But frankly he will be largely forgotten a month after he drops out. Still it is likely to be another election of fine margins so any little difference it does make is not to be sniffed at I suppose.
___________
Agreed. If RFK decides to pull out, timing is everything 😉
BTSays:
Wednesday, August 21, 2024 at 9:17 pm
[‘Not a problem! – was just a friendly reminder tbh.’]
Cheers.
Most families have a black sheep; the Kennedys had/have two: Ted & RFK. I doubt Robert will be invited to the Kennedy compound in Hyannis Port at XMAS. Anyway, if he does accept Trump’s kind offer of a cabinet position,
it’ll have minimal impact on the election. Harris/Walz by a country mile.
Summary of some of the poll analysis websites. Not really showing the tightening that the betting markets are:
538 – Harris +2.9
270towin – Harris +1.5
The Economist – Harris +3.2
Raceto theWH – Harris +2.6
RealClearPolling – Harris +1.5
Silver Bulletin – Harris +2.5
The Hill – Harris +3.0
Evads
But aren’t those numbers simply the raw vote lead nationally? – which would imply, frankly, a Trump narrow win if taken by themselves.
BTSays
You can do the analysis either way. Taking away the state with the largest delta or the state with the largest populations. Either way, its basically another way of showing that Democratic votes are distributed very unevenly, which we’ve already known. But whether or not that means Harris has to lead by 4 points nationally to be favoured for a win, or less (say 2%) is something we don’t know at this point. But the point stands, to do the analysis properly, one should remove from both sides of the equation. And if you’re getting smarter about it, you can weight by number electoral votes, to show how much excess or “wasted” votes each party has per EV. If I get sufficiently bored I might try and do that later.
BTSayssays:
Wednesday, August 21, 2024 at 10:16 pm
I think you’re probably right. I was making no commentary or analysis, just posting the averages as they are on each website tonight. The only comment I would make is that the margins have not closed up like the betting averages have.
New thread.