• jordanlund@lemmy.worldM
    link
    fedilink
    arrow-up
    98
    arrow-down
    4
    ·
    edit-2
    3 months ago

    (Sigh) Again, national polls are meaningless since we don’t have national elections.

    Last time I did this, Harris was in the hole in the swing states where it counted, let’s see how it looks now…

    Arizona - Trump +2 to +5
    https://projects.fivethirtyeight.com/polls/president-general/2024/arizona/

    Nevada - Tossup, Tie, Harris +2, Trump +1
    https://projects.fivethirtyeight.com/polls/president-general/2024/nevada/

    New Mexico - No polling data for Harris.
    https://projects.fivethirtyeight.com/polls/president-general/2024/new-mexico/

    Georgia - Tossup, Tie, Harris +1, Trump +2
    https://projects.fivethirtyeight.com/polls/president-general/2024/georgia/

    North Carolina - Trump +2 to +3
    https://projects.fivethirtyeight.com/polls/president-general/2024/north-carolina/

    Virginia - Trump +1 to +4
    https://projects.fivethirtyeight.com/polls/president-general/2024/virginia/

    Pennsylvania - Trump +2, +3, +4, Harris +3
    https://projects.fivethirtyeight.com/polls/president-general/2024/pennsylvania/

    Michigan - Tie, Harris +11, +12 https://projects.fivethirtyeight.com/polls/president-general/2024/michigan/

    Wisconsin - Harris +1, +2
    https://projects.fivethirtyeight.com/polls/president-general/2024/wisconsin/

    Minnesota - Harris +3 to +10
    https://projects.fivethirtyeight.com/polls/president-general/2024/minnesota/

    So, previously, Harris was behind everywhere. She’s now moved Nevada and Georgia to tossup territory, which is good, but the notable news are the HUGE gains in Michigan and Minnesota.

    HOWEVER… Plotted on the electoral map:

    The Triumverate is Pennsylvania, Michigan, Wisconsin. Losing any one of those three tanks her and she’s down in PA.

    She could win all the tossup states and still lose without PA.

    • usernamesAreTricky@lemmy.ml
      link
      fedilink
      arrow-up
      25
      ·
      edit-2
      3 months ago

      The electoral college is still close, yes, but it is not what you are saying in some of those states. For instance, the Virginia polling you cite stood out to me as not matching most of the polling even when Biden was running, so I looked into that:

      1. There aren’t polls in 538 for Virginia that were run after Biden dropped out

      2. The range your writing to me looks though you are looking at only a results of the singlular FAU/Mainstreet USA Poll from not only before Biden dropped out, but the days after Trump was shot. This poll also has a 4-5% margin of error due to low sample rates. The reason I suspect that this was the only poll being looked at is the poll has a ton of entries in 538 because it was broken down quite a bit. It asked likely voters and register voters about Harris as a hypothetical both with third parties and without (this creates 4 entries in there). The difference between likely and registered voters created a trump +1 and a trump +4.

      3. Most of the other hypothetical Harris polls in Virginia before Biden dropped showed Harris up. Considering Harris has generally improved compared to hypothetical polling before Biden dropped in most states, it’s unlikely to be Trump +1 to +4 at the moment


      538’s moving averages show a bit of a different picture for some of those states as well. For instance, Pennsylvania shows a narrow +0.6% Harris average (note though that they do weight register voter polls different than likely voter polls)


      Election is still close, but there has been movement in swing states. Enough that some election models are now changing who is most likely to win if only slightly. For instance, Nate Silver’s model puts Harris just a bit above in win % including electoral college in it

      • jordanlund@lemmy.worldM
        link
        fedilink
        arrow-up
        8
        ·
        3 months ago

        Just going with what we have, and like I say, there has been positive movement for her since the last time I ran these stats.

        The concern for Virginia was that it had been solid for Biden, with him out, it’s definitely at risk.

        What will be interesting is to see how all this changes once she picks a VP and they start team campaigning.

    • nifty@lemmy.world
      link
      fedilink
      arrow-up
      3
      ·
      3 months ago

      Some newer YouGov and RealClearPolling reports: https://ijr.com/trump-harris-sit-on-razors-edge-with-voters-in-battleground-states-poll-reveals/

      In the rust belt, Harris is tied with Trump at 48% in Michigan, 50% in Pennsylvania and is just one point behind in Wisconsin at 49%, according to the poll. Previously, Trump was leading Biden in Michigan by 2.1 points, in Pennsylvania by 4.5 points and in Wisconsin by 2.9 points, according to polling averages from RealClearPolling

      Harris has also gained ground in the south and the sun belt, tied with Trump at 49% in Arizona, trailing by three points in Georgia and North Carolina and leading by two points in Nevada. Before Harris became the presumptive nominee, Trump was leading Biden in Arizona by 5.8 points, in Georgia by 3.8 points, in North Carolina by 5.4 points and in Nevada by 5.6 points, according to RealClearPolling averages.

      • jordanlund@lemmy.worldM
        link
        fedilink
        arrow-up
        4
        ·
        3 months ago

        Yup, the momentum has swung back away from Trump, there should be another bump when the VP tour starts and another for the Convention.

        The question then becomes if it can hold through November.

        I really look forward to re-running these stats in September/October.

    • Socsa@sh.itjust.works
      link
      fedilink
      arrow-up
      2
      ·
      edit-2
      3 months ago

      There is literally zero percent chance that Trump is up in Virginia. If anything you can calibrate the other polls from this.

    • whotookkarl@lemmy.world
      link
      fedilink
      arrow-up
      2
      ·
      3 months ago

      What are the error bounds on those polls? Last I checked they were all within the margin of error for swing states.