US presidential election minus 14 weeks

The first Harris-versus-Trump forecast model to open for business suggests a significant but by no means decisive advantage to Trump.

With The Conversation keeping this site’s regular US correspondent busy, a quick post of my own on the US election campaign, and a forum for the discussion thereof. The big news from my perspective is that, after all the noteworthy forecasters closed for refurbishment following Joe Biden’s withdrawal, Nate Silver has lifted the lid on his Harris-versus-Trump model.

The model launches with a 61.3% win probability for Trump and 38.1% for Harris, the balance presumably reflecting the possibility that one candidate or the other doesn’t survive until November. These numbers suggest a model with a judiciously wide zone of uncertainty around projections that superficially look very encouraging for Trump. Silver’s model records an essentially dead head on the national popular vote, and doesn’t credit Harris with a better than even chance unless she lands at least two points clear. State-level projections find Trump more likely than not to flip Wisconsin (just), Michigan and Pennsylvania (a little further ahead), Nevada (a little further again) and Arizona and Georgia (both about as strong for Trump as North Carolina, which he carried in 2020).

The Economist’s model is still on ice, but its page explaining its methodology is well worth reading. Its charts comparing the predictiveness of its poll-based and “fundamentals” models going back to 1948 are particularly interesting in finding that the latter have the superior record – certainly at predicting the result 150 days out, but even unto election day itself. However, one of its parameters does not seem to me to be quite as fundamental as all that, being a poll-based measure of presidential approval.

The question of polls-versus-fundamentals was the focus of a critique by Nate Silver of the new model developed by FiveThirtyEight, the enterprise formerly synonymous with Silver but now bought out, LucasFilm style, by Disney. Its new incarnation is overseen by G. Elliott Morris, of whom Silver says he is “not a fan”. This was producing remarkably bullish results for Biden up until it was put on ice, which evidently wasn’t persuading too many senior Democrats. As well as criticising a lack of transparency, Silver observes that the model seems to be overwhelmingly favouring fundamentals, despite its supporting data suggesting that fundamentals should in fact be viewed as less predictive than polls. Its thesis, Silver argues, is that – as of July 21 – “Joe Biden is a reasonably clear favorite to win the popular vote because he’s an incumbent, and it’s too early to really update that assumption based on the polling or anything else”.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics.

1,063 comments on “US presidential election minus 14 weeks”

Comments Page 22 of 22
1 21 22
  1. BTSays

    Have you ever had anything positive to say about trends to the Democrats/ Kamala Harris?

    There is always a factor missing, like Trumps rapid downhill slide.

    Paint it how you like, Trump is done and dusted, his party is over!

    Sorry to disappoint you.

  2. The Biden / Harris administration was the first since Carter / Mondale where neither attended an Ivy League university. Biden / Walz would be the second.

    Trump and Vance both attended Ivy League schools.

    Just an observation.

  3. @Been there

    I’m not sure of the point of your question. I’m also not sure how to answer it. I’m 99% sure the answer is ‘yes’.

    It may be lost on partisan posters, but I post psephologically regardless of my opinion. I’m not sure how or why you think you know who I support or why it might be the remotest bit relevant whether I’ve said something ‘positive’ to say on a specific point. It’s not mine or anybody else’s job on here to be ‘positive’ or ‘negative’ about trends in polling, merely to try and add something to the psephology of it all.

    In my dreams, there’s a PB world where no-one can even work out who different posters support because that’s not the point of their posts.

    Unfortunately, because the opposite is largely true, people jump to an awful lot of conclusions – frequently wrong ones – about other posters. Often they’re responded to according to who posters believe they support rather than what they say, for instance.

    Sometimes it feels like treading on eggshells typing a post on here, needing lots of thought so it’s not misconstrued.

    In spite of that, I still get completely irrelevant questions like yours above instead of talking about polling and likely impacts on it. Big waste of time but I’m the old-fashioned type that thinks it’s a bit rude not to reply unless people are being outright abusive.

    Thanks.

    PS FWIW unlike most posters I don’t have a blind loyalty to a particular ‘side’. Regarding the USA, I’m old enough to have supported Bush then Bush then McCain then Obama then H. Clinton then Biden. I’ve given no indication who I’m supporting this time, see if you can work it out by election day and LMK. I am a big fan of the USA as a country so their politics interests me greatly.

  4. It may be lost on partisan posters, but I post psephologically ..
    Psephologically or Pseud-ologically?
    Whatever, no one cares who you’re voting for, so why bring it up?

  5. The former Minnesota chair of the GOP (Hand) has criticised Walz for providing free lunch to school children. If that’s the best they’ve
    got… Walz seems to be the type who won’t get too upset about being called a Liberal; indeed he’ll probably revel in it. When JFK
    invited Johnson to be his running mate, his Texas drawl arguably got the Dems over the line. I think Walz may have a similar effect.

  6. BTSays,

    Your posting says where you are politically based. You can postulate all you like but the evidence is clear.

    Someone with a just a little of intelligence can see that.

    Harris is better than Trump, Walz is better the JD Vance. The choice is simple.

Comments Page 22 of 22
1 21 22

Leave a Reply

Your email address will not be published. Required fields are marked *