Today’s Politico headline story is this: “President Donald Trump is in a deeper hole heading into Election Day than he was four years ago, when he stunned Hillary Clinton and most of America.” Which is exactly what I said, more or less, yesterday.

              Except there’s only one little problem, which is why I decided to write another column on the election outcome. And the problem is that both my column yesterday and Politico’s story today is based only on the analysis from Nate Silver’s 538 website, and he ain’t the only one out there calling the race.

              The other aggregator is RCP. And their numbers show a race that is not only closer in the national polls, but is closer in the battleground states to the point that Trump could actually win.

              How is this possible?  Don’t these guys aggregate the same polls? They do and they don’t. I’ll try to explain.oHow

              First and most important, the aggregators are aggregating different polls. Nate’s group posted 12 different polls yesterday which questioned voters up through October 1^st; RCP posted only 4 new polls. Even within the polls on each website there are crucial differences in terms of how each polling organization developed its poll.

              So, for example, for Pennsylvania, both aggregators posted the most recent poll from Susquehanna which gives Trump a 49%-48% edge. RCP also posted a poll from some outfit called Insider Advantage/Center for American Greatness, which gave Trump a 2-point lead, 49% for Trump, 47% for Joe. But Susquehanna queried 500 voters, of whom 79% described themselves as White; Insider/Advantage also queried 500 voters, and 81% said they were White. 

According to the Census, Pennsylvania’s White population is estimated to  be 75.7%. When a poll gets down to differences of less than 1 percent, asking what could turn out to be an over-sample of a certain racial group can produce results that may not turn out to be correct.

  Another problem with the 538 predictions is that whatever number you see for a particular day, that number doesn’t represent just an average of the most recent polls. 538 claims to run “the most sophisticated daily rolling average” but I can’t find exactly how long the average rolls. Is it 7 days? Is It 14 days? Either way, if a poll comes in today that represents a significant shift from the previous polls, the degree of shift will be understated in terms of predicting how the race will turn out.

Dixville Notch just announced the first official vote count for the 2020 Presidential campaign. Dixville Notch holds its voting in the lobby of the old Balsams Grand Hotel in New Hampshire and it has always been the first place to report. Well they used to vote in the hotel lobby, but the hotel has been torn down.

Last night, right after the stroke of midnight, Joe got all 5 votes in Dixville Notch. There’s only one problem. Hillary got 4 votes to Trump’s 2 votes in 2016. So, as we used to say in corporate, WTFK? If you don’t know what that means….

  As for all of Trump’s threats about how the election’s a ‘fake;’ how he’s going to declare victory as soon as the polls close; how he’s going to make sure that the National Guard is ready to be called out if those terrorists and radicals try to burn down the cities after he wins; I don’t think he’s going to do anything at all.

  This is a guy who has spent the last four years announcing one plan after another which never took place.  Remember the wall he was building in Mexico? Remember how he was going to propose a term limit on members of Congress? Remember how he was going to ban White House officials from lobbying Congress until they were off his payroll for at least five years? Remember how he was going to cancel all federal funds going to sanctuary cities?

  This is a guy who in the 1,173 days that he has been President has gone golfing at least 306 times. That’s what his Presidency has been all about.