Four years ago today I had fear in the pit of my stomach. I had spent the entire year watching people who I thought were sensible dissolve into a puddle of insanity over almost 40 years of dedicated Republican propaganda prepare to vote away our Democracy.

The people I fought with insisted that Independent voters, Greens, and Libertarians could exercise their right to an opinion expressed at the ballot box, that Bernie Sanders wasn’t the boss of them because they “thought for themselves,” or that Trump wouldn’t be that bad.

By the middle of September, 2016 I suspected Silver’s site was wrong. By the end of October I was sure. I said so then. People ignored the outcome that showed a win for Trump, choosing instead to believe that Clinton had the race locked up. The people were wrong, and I hold polling aggregator sites like FiveThirtyEight responsible for failing to make that outcome clearer.

I don’t trust them. I won’t even link to their site.

Too many red states are still a crap shoot. This is a theoretical second term election, which throws a monkey wrench into the primary results for prediction purposes. There is literally no way to know how many people who voted early were counted four years ago when they voted on Election Day. Just based on my own situation (mailed ballot cast on 10/1), I’m not in the Early Voting stats because I voted before the polling centers opened. That’s why I won’t predict anything. The numbers look hopeful, but then so did Hillary Clinton’s.

Nope. I’m in it for the long haul.

What I’m most worried about is a hack of the power grid, a take down of polling places because dangerous, armed lunatics try to break down access, or some other wag-the-dog event.

Too many jurisdictions still use touch-screen voting machines, which have proved to be hackable.

This year, while I’m not making predictions about any of the races, I will tell you there’s only one site that does which passes my muster: Sabato’s Crystal Ball (a non-partisan project of the UVA). The Associated Press has already declared they will not call a single race until they are absolutely certain beyond the shadow of a doubt that the outcome they declare will be the final one. Sabato’s uses some of the same data I do. Of all the possible models, their map looked the closest to mine.

I’m not willing to get excited or count anything until I see actual returns. I don’t start to fill in numbers on my Lists until the returns are at least at 75% or better.

There is zero chance I will call any race before it gets to at least 85% counted across the board. And for any race closer than 100k votes, there’s no chance at all.
We’ll see how long I manage to stay awake. This is when polls close in all the states:

State Poll Opening and Closing Times (2020): https://ballotpedia.org/State_Poll_Opening_and_Closing_Times_(2020)

The Green Papers (http://www.thegreenpapers.com/) has additional information for the territories. This is a major flaw in Ballotpedia’s site. Eventually, they will have the definitive numbers for the vast majority of races.

Four years ago, I watched the returns using The Guardian’s online tracking system for incoming poll results. https://www.theguardian.com/us

Not sure whether I will do the same this year. It depends on what data they deliver for the House and Senate races. I need to see all the data, not just the so-called “important” races. Numbers are everything.

I will breathe easier when we get to the 6th. We will know, I hope, by my birthday, that Trump is history.

Or not.

The only thing I will predict is that November 3, 2020 is going to be a very long night.

[ETA] The Guardian backs me up with regard to my concerns about the state of voting:

How many Americans have voted in the presidential election – and how?

Early voting accounts for two-thirds of all 2016 turnout but an uneven system and Republican tactics suggest trouble may lie ahead