9/8/16

I intend to write a longer post on the elections during the next few weeks.
For now three quick notes:

A. Trump: Trump gave two public appearances yesterday. In the morning in Philadelphia he gave a pre-prepared speech on national security and foreign policy, which was a very good speech. In recent weeks, he consistently gives very good speeches, which are all pre-prepared. Yesterday evening he appeared in a question and answer forum on the same subject and was the usual Trump of old; ducking, weaving, and not really saying anything by talking from both sides of his mouth. While not insulting anyone this time, he did make unnecessary mistakes by refusing to accept that Putin is not a positive influence on the world stage (although he is right in saying that Putin is much more of a leader than Obama is). The bottom line is this: “Teleprompter Trump” is pretty impressive and right on most points. “Unscripted Trump” is pretty bad. To me, this means that his performance in the debates is likely to be terrible and that he is at risk of losing the elections (if he has not already lost in these debates).

B. Polls

  1. Polls in General: My scathing view on polls is well established. I made a big deal of the inadequacy and inaccuracy of polls before the 2012 elections only to be faced with my own failing as a human being when the polls were proven right. Since then, there were three major elections where the polls were badly wrong. Two were in the UK. While the story of the Brexit vote and the polls is now famous, the exact same screw up in the polls for the general elections in the UK just 12 months prior is already forgotten. The polls totally failed to predict the impressive win by the Conservative party in the 2015 elections. How come? Polling, after all, is supposedly based on well-known science where polling methods and statistical sampling are the same. Indeed, some of the polls are being conducted by the same organizations both here and in the UK. The UK is a much smaller country and much more homogenous. So how is it that these polls got it so wrong? In addition, there is the case of the polls in the US for the mid-term in 2014. They, too, failed miserably to predict the extent of the Republican victory in both House and Senate. So what does this tell us about the polls? I come back to my original contention that polls are widely inaccurate, not an effective tool, and really have no clue. Of course, by pure statistics, sometimes they will get it right too. The bottom line: no one really knows.
  2. Clinton Polls: Having just concluded that the polls are worthless and not predictive, there is one element that is so consistent in so many polls that it seems to correctly reflect the public opinion. It also makes sense. Clinton has not been able to garner 50+% in ANY of the head-to-head match-ups in the polls, with one exception. So, literarily, in dozens of head-to-head polls in the last 6 weeks or so since the nomination became official, she is failing to cross the 50% line needed in order to win. How is that possible? How does it make sense for a politician that the same polls and all the pundits are saying is a shoe-in to be elected and is so well known, having been on the political stage for 15 years plus? That cannot be a good indicator for her.