Blue wall cracks. Even 538 showing Trump winning in the past five polls in Pennsylvania

They have not included a NYT/Sienna poll that will show Harris in the lead, but that is NYT/Sienna, a historically biased poll

Granted, none of these are big leads, but this is the first time in three elections that Trump has actually started to show polling breaking for him in any of the blue wall states. Both RCP and the Hill show him winning two of the three states right now.

He never led in 2016 in Pennsylvania, Wisconsin, or Michigan. In fact, at this same time in the election cycle Trump was behind by 7.3 and 9.2 in Pennsylvania against Biden and Clinton. He was behind 7.0 and 9.6 in Michigan, and behind 6.3 and 6.0 in Wisconsin on this same day in 2020 and 2016. He was down an average of 7.6% for those three states over the past two Presidential cycles. Today he is ahead by 0.23 points. This is a polling improvement of nearly 8 points over the past two cycles, something that 538 and Nate Silver can no longer ignore even as their models stubbornly keep Harris ahead.

What does it mean when Trump ended up competing closely in all of these states in both years, winning all three in 2016, coming within a few thousand votes in Wisconsin and within just over a point in Pennsylvania in 2020. It tells me he overperformed those polls by over seven points. It tells me that most of these polls have a built-in bias against Trump.

Keep in mind that Pennsylvania has probably the biggest disconnect between the public polls and what we are hearing from the campaigns and other insiders. The "Pennsylvania problem" for Harris has been a thing for some time, and both camps seem to believe that Trump is winning there and that the margin is not just a few tenths of a point. This why the Harris campaign has been pushing so hard in the South with Georgia and North Carolina. They truly believe that they might need to pick up one of those states to offset a likely loss in Pennsylvania. It seems that more and more polling is reflecting it.

Now in all three states, the pollsters that will keep Harris in the mix are NYT/Sienna, Quinnipiac, Bloomberg (and possibly FOX news). These four pollsters have consistently been outliers to the left in state polling, with the first three being more obvious than FOX. As long as two or three of them are showing Harris up 3-5 points, it will at least appear that Harris is running close or even ahead by a bit. It might be wise to toss those three pollsters all together to see what a more accurate polling average might be.