The three data: polls, coronavirus and weather

Okay, this has been a LONG week with a lot of tension. This was the week of voting.

The few days before the final vote started, the opinion news from mainstream media got, to me, giddy. For a long time the pundits were cautious due to the polling trauma from 2016, but as we got closer to the actual final voting day when any conceivable “October” surprises could no longer impact the voting decisions (the Hunter Biden story did not appear to work), the opinion makers decided to buy into what the polls were saying: Biden would decisively win. I don’t know if the polls were saying a landslide but they were saying that Biden would win.

Then Tuesday rolled around and the results were coming in WAY closer than anticipated.

So, the polls appeared to have been off.

I was afraid of that. While the news pundits were seemingly giddy at the prospect of an overwhelming Biden win due to the number of polls predicting his win, I was thinking, “Oh, no, Trump could still win.” Why did I think that?

I wasn’t positive Biden would lose but I wasn’t positive he would win either.

There were a few articles about how a lot of communities were overrun with Trump signs, how some Biden supporters had to remain silent. There was an uneasy feeling that polls were missing this dynamic.

Then there was this article about the shy Trump supporters who would lie to the pollsters about their support because of the social undesirability of admitting such support. This theory postulating that most polls were underestimating Trump’s support due to social undesirability of such support made a lot of sense to me. I think most people are aware that what Trump was doing was cruel, corrupt and criminal. You can’t separate children from parents and come off smelling like a rose. You can’t say that the Nazis and white supremacists are “fine people” and look like an angel. You can’t tell the American people that the virus is going to go away miraculously when you have knowledge that it was contagious and potentially lethal. And supporting such a person says a lot about the supporter.

These “shy Trump voters” were aware about his undesirability in the social context so they lied to the pollsters. To me, that says a lot about these silent supporters. They are okay with his cruelty, his corruption, and his criminality. They put aside these 3 C’s for something else.

They knew what they were doing and supporting.

The reasoning behind the article made a lot of sense. I had another theory which at this point, I don’t think pans out. I thought maybe these “shy Trump voters” were deliberating lying to the pollsters not because of social undesirability but to prank them as a way to “own the libs”, a la those Tik Tok teens who fooled the Trump campaign when they signed up for the Oklahoma rally. I thought that other theory might have been a possibility, but if so, we would have heard about it by now.

The polling guys have a lot of work ahead of them to figure out why, yet again, their polling failed dramatically. It appears that polling have mostly worked up until 2016. There is something about this era that might requires a different approach to polling to capture the actual sentiment of those being polled. Or it could be that polling just does not work when it comes to someone like Trump.

The number of daily cases are rising rapidly, especially in the Midwest, but it is impacting all regions of the US, and I now see deaths are rising. Hospitalization and positivity rate are going up too. We are deep in a new surge.

The latest news are warning that hospitals are running out of capacity. Cases around the world are also skyrocketing, especially in Europe.

And yet, it looks like people are not settling down for the next four months. People just want to get together. I suspect that once hospitals are inundated with sick and dying people, that’s when people will do what’s needed. Right now, we just went through a series of Trump rallies, we opened up schools and loosened restrictions for businesses, and today, it looks like there were some Biden victory rallies.

So, yeah, we’re not serious about this yet.

Sources for charts:

Wikipedia and Covid-Tracking Project via creative commons license CC BY-NC-4.0.

https://en.wikipedia.org/wiki/Template:2019–20_coronavirus_pandemic_data/United_States_medical_cases

https://covidtracking.com/api/v1/states/daily.csv

You can see how the US is increasing very steeply and this increase is driven by the Midwest. Illinois is really driving the Midwest.

Boooo! Yet another storm to worry about. When I look at the data for hurricanes hitting the US, either from the Atlantic or the Gulf Coast (Hawaii and Virgin Islands), the records from Wikipedia show only three and the last one took place in 1985 hitting Florida as a Cat 2 and Georgia as Cat 1. The latest weather data shows the tropical storm, rather than a hurricane hitting Florida on Monday and then bouncing off and churn in the Gulf Tuesday and Wednesday.

I hope that Eta stays a tropical storm and I hope this is the last one.

Boy, this has been a LOOOONNNNNGGGGG year!