Getting close to political Armageddon
___________________________________________________
I have no idea who is going to win the election next week. When I look at it I'm 50/50. That does not mean I think each candidate has a 50% chance of winning, it means that I am fifty/fifty. Half the time I think Trump will win half the time I think Biden will win.
When I analyze why I am so conflicted it really does come down to the polls and political punditry. Those definitely point to Biden while nearly everything else points to Trump. It is easy to throw out the pundits, I have virtually no trust in the majority of them. So putting the polls aside for a minute, let's look at the "everything else" side of the equation.
There is little doubt that Trump has a tremendous enthusiasm advantage, but even that can be misleading. Watching months of giant boat parades, miles long car caravans and huge rallies of enthusiastic supporters is impressive, but it is not an election. Don't get me wrong, all those people that show up for such events are solid Trump supporters that will vote for him, but a campaign with millions of enthusiastic supporters can motivate thousands to come out for large events. That, however, does not mean there are more of us than them.
The large crowds are not unimportant, they motivate people to vote, get others involved, probably even turn some undecided to their cause, but again it does not mean they have the most votes come election day. I remember going to a Palin rally in 2008, it was very enthusiastic and large. McCain conceded the election by 11:00 EST, people were still voting in Alaska and Hawaii. There may, however, be something hidden in those rallies that are bigger than the rally themselves, more on that later.
Then I look at other things, like Gallup having a poll that shows that 56% of Americans say they are better off today than they were four years ago. That is amazing given that we are in the middle of a pandemic with tremendous financial difficulties, but it is by far the highest number since they first polled that question in 1984. As soon as you think that number is unbeatable though, Gallup also has a poll that shows only 43% of Americans think that Trump has earned a second term. Ugh. Just to finish off the Gallup thing, every month going back to 2004, Gallup asks people what political party do you identify with. Over the years Democrats have usually been ahead. In the last one prior to the 2016 election the results were Rep 27, Ind 36, Dem 31. The most recent one, I don't know if another will come out before election, is Rep 28, Ind 42, Dem 27. What does that mean? It seems to bode well for Republicans, but who knows? But how can Trump be trailing so badly while political identification is moving his way.
There are however some truly good signs for Trump though, and it has to do with the pandemic. When the pandemic hit, the Democrats basically shut down their "ground game," Trump did not. In fact it is massive. While Biden's team sat in their collective basements, Trump's campaign knocked on a hundred million doors! This is why Republican voter registration, especially in swing states, is greatly outperforming Democrats. In all the key states, Republicans have grown their voter base compared to Democrats. New voters are the most likely to vote. Despite some dubious misreporting by NBC there is something else to factor in when it comes to the pandemic, many colleges are closed. The normal hive of college registration drives is now stuck doing virtual classes rather than peer pressure at the sorority house. Finally, the push by the Democrats for mail in voting means that far more Democrat votes will be tossed out due to errors than if they had just voted in person.
As I said there is something in the Trump rallies that may be transformative and validates the pandemic advantages I just mentioned. It is something that most pollsters are missing and the closer we get to the election the more they miss it. I noticed it awhile back and you would only realize it if you are on Twitter and follow Ronna McDaniel the GOP Chairwoman. During every rally the Trump campaign has attendees link to an app in order to obtain information. I am sure this is primarily for fundraising and volunteers but they also take other information. After every rally, McDaniel releases this information, here is one from the rally in Erie, Pennsylvania the other night.
Forget the signups, notice the Not Republican percentage. Now that does not mean they are all Democrats, though some certainly are, some may be no party affiliation voters. Regardless, if over a quarter of the attendees were not Republicans it shows that Trump is popular outside is "base". Perhaps even more important when it comes to the polling is that more than a fifth of attendees did not vote in 2016. This is important because when pollsters poll "likely voters," one of their most important factors they use to include a person in that category is if they voted in the last election. Pollsters might, and probably are missing, a large number of new voters. Here are two more, one from Sanford Florida rally and the other from a different Pennsylvania rally, both on October 12th.
You can almost be sure that none of those attendees, or anyone else, who has not voted in the last four elections is also not showing up in "likely voter" polls. It seems there is a significant number of first time voters, returning voters and crossover voters attending these rallies. If they are representative of the larger population that is not being picked up in polls, then the entire playing field could have changed. In other words, as he did in 2016, Trump could be creating new voters beyond what would normally be expected. Those increased voter registrations point to this as well and normal polling outfits would not figure this in. They are famous for fighting the last war.
Remember that polls are more an art than a science, though they would have you believe otherwise. They are nothing more than models forecasting a future event. Basically all models are saying that if this happens, this will occur. The "this" which must happen is an accurate assumption of the turn out. If the assumed turnout model is wrong, there will be a different result, plain and simple. You know this to be true due to wide differences in the various polls.
Michigan for example
FOX News* 10/17 - 10/20 1032 LV3.0 52 40 Biden +12
Trafalgar Group (R)* 10/11 - 10/14 1025 LV3.0 46 47 Trump +1
The fact that there is a 13 point difference in two polls where they interviewed basically the same number of likely voters (LV), have the same margin of error (3.0) shows that this is not an exact science. By the way, Trafalgar Group was the only polling outfit that nailed all those upsets in 2016 in individual states. They use a different methodology than other polls.
To accommodate for all these disparities most people rely on the RCP average or the 538 Model which use the average of polls to compensate for these. The problem is that if most polling firms are using the same or very similar methodology to determine their turnout model, it is just a continued repetition of the same "if this happens" forecast.
Is this what is happening? I do not know, but based on the way these media and academic institutions glom onto narratives to push their agendas it would not be a surprise. The fact that they were off in 2016 shows that it could be happening again, only this time it would be worse if they are this far off, both in raw numbers and in consequences. Because there is one thing that is for sure:
Whoever wins, the other side is going to be shocked..... and that is not good
Comentários