Tuesday, October 20, 2020

HOW Justice Barrett can give the election to Trump

SUMMARY:   The debate about the Supreme Court and the Presidential Election just became real.  Judge Barrett will be the deciding vote on Pennsylvania's Electoral College, and Pennsylvania may very well be the final swing state to report results of the election.  That means that Judge Barrett will cast the deciding vote that may determine the next president.  In that case, should 5% of the mail in ballot be disqualified, Biden could lose PA and the election.


I don't like these kinds of super-hypothetical conversations normally, but we need to look at the following.

1)    As my previous post showed, we will know the results on Nov 4., unless AZ, GA, FL and NC all swing to Trump.  Since he won them in the past election and they are traditionally Republican, this is a very possible scenario.  According to that earlier post, PA becomes the last relevant reporting state in that case, because their law requires late counting of mail in ballots and the state has only limited experience with mail in voting in the past.  Well, things just got worse.  A federal court ruled recently that the PA elections commission may, in the interest of maximizing fairness and avoiding postal problems, accept all ballots postmarked by Nov. 3 rather than following the PA statute of ballots being received by Nov. 3.  The lawsuit quite legitimately challenges that ruling in favor of the statue.

2)   Yesterday the US Supreme Court ruled 4-4 NOT to take up the case, thereby leaving the lower ruling in place as a result of the tie caused by Justice's Ginsberg's empty seat. Voting to leave the lower ruling in place, and thereby expand the number of mail in ballots that will count, were the 3 liberals and Chief Justice John Roberts.  The four Conservatives on the bench all wished to receive the case with the understanding that they would enforce the PA statute and forbid the counting of ballots received after Nov. 3.

3)    As my last post showed, mail in ballots are strongly in favor of Biden nationwide, and it is no different in PA.  Increasing or decreasing the number of mail in ballots counted will increase or decrease the vote for Biden.  If the Conservatives were to enforce the statute, the Democrats will lose ground in PA for sure.

4)    Justice Barrett will be sworn in before Election Day.  

5)    So, when, not if, Justice Barrett joins the court and the lawsuit arrives, she will certainly vote to overturn the lower court.  All mail in ballots received after Nov 3 will be disallowed, and Biden's numbers will go down.

6)   Will it matter?  Absolutely yes. In 2016, Trump received 2,970,733 votes in PA and Clinton had 2,926,441, or a difference of 44,292.  5% of the mail in ballots breaks down to 87,797 Democrats versus 33,619 Republicans, or an advantage of +54,178 for Biden. If the Barrett breaks the tie, and 5% of the mail in ballots are disqualified, Biden could lose over 50,000 votes in PA, which is more than the margin of victory for Trump in 2016.

Here's the breakdown:

 Democrats hold an overall lead in registered voters and nearly a 3:1 lead in mail in ballots requested.   Of the 2,7 million mail in ballots requested, nearly a quarter had been received by Oct 16.  Will people wait until the last minute?

Total voter registration per PA Dept. of State on Oct.16

  • Dem = 4,199,412
  • Rep = 3,494,255
  • No Aff = 890,756
  • Other = 398,765
  • Total = 8,983,188

Total approved mail-in and absentee applications statewide

  • Dem = 1,755,940
  • Rep = 672,381
  • Other = 299,946
  • Total = 2,728,267

Monday, October 19, 2020

How The Polls Miss Trump Voters: Is this 2016 all over again?

 How The Polls Miss Trump Voters

The mainstream news headlines, national and state polls this week look nearly identical to the 2016 polls.  That should be encouraging for Republicans, and frightening for Democrats.  Can Trump run the table, win all the coin flips, and emerge victorious again in 2020?  It could happen.

First, look at the similarities: In the polls Democrats are leading by +12 points nationwide, are secure in their traditional states by several points in PA, MI, WI and MN.  In addition, they are ahead in purple and red states like CO, NM, AZ, FL, NC and they have put IA, GA and TX into play.

With 2 weeks to go in 2016, CNN was projecting 290 of the needed 270 electoral college votes for Hillary Clinton, and showing that Donald Trump would have to win back all the typical red states as well as a few blue states, and that it was hard to see any realistic scenario when that could happen in reality.  According to 270towin.com's "concensus map" of polling averages, they are projecting that Biden has that exact same 290 of the needed 270 to expect a victory.  Deja vu, all over again.

2020 = 2016 in most polling at the surface.  To argue against that, Larry Sabato and others have written extensively about (1) how they have fixed polling compared to 2016 and (2) how the persistent nature of the Biden lead is qualitatively different from the Clinton campaign in 2016.  In other words, "yes" we blew it last time, but honestly we really do know what we're talking about this time.

I believe that the essential issue is voter turnout.  In recent history, especially 2008 and 2012, the least likely demographic to turn out to vote was men with High School (or less) educations.

In addition, since 1980 there has been a steady and constant increase in the percentage of female voters overall, so that they are a majority of the active electorate.  This year's polls show Biden enjoying a larger lead with women than Hillary Clinton had in 2016.

Polls typically measure likely voters with questions like "did you vote in the last (last two) presidential elections, etc.  Accepting that past non-voters probably won't vote is statistically valid if the patterns remain constant.  With Donald Trump, this was not true.  Trump raised voter turnout among less educated white males tremendously, especially in Florida, PA, MI, and Wisconsin - all of which he won by narrow margins within the polling "margin of error."  By getting more than 50% of previous non-voters in his base to vote, he didn't need to take voters away from Hillary in order to win.  He could raise the total participation by previously ignored loyal Trump voters.  In those 4 states it was enough to squeek a victory.  The signs everyone missed were the enthusiasm ratings, and the packed throngs of followers are rallies throughout the midwest in the final weeks of the 2016 campaign.  This is exactly why he is trying to keep the rallies going - they are his best path to victory again.

As a result there are several questions, whose answers will only be proven on November 3:

1)    Are the polls catching those previous non-voting, less educated white males this time around as likely voters?  After all, now they would say "yes" to recent participation.  If so, then the polls are more reliable this time than last.

2)    Will the Biden coalition recreate the 2008 enthusiasm for the Obama/Biden ticket, when over 57% of eligible registered voters turned out, or will it look like 2016 when only 55.5% of the voters showed up?  Early voting seems to indicate tremendous enthusiasm for voting this year, with some predicting record turnout of over 60% of eligible registered voters participating.  That should help the Democrats.

3)    Will the missing 1.5% in 2016, as compared to 2008, show up this year?  If they do, will they counteract the Trump bump in white males enough to hold him back from the close wins he enjoyed in 2016?

The bottom line is this: voter turnout will drive the day.  Trump voters will repeat their numbers, but there is little or no indication that he has any "new" followers compared to last time.  For the Democrats 57% voter turnout has to be an absolute minimum goal.  It is logical to presume that the closer the country gets to 59% or 60%, the better it will be for Joe Biden.

15 days and counting.  Turn out the vote.