How to predict the 2020 presidential election

A few weeks ago, I shared a piece that suggested we might have to make a prediction about the presidential election.

It was a bit of a joke at the time, and not because I was surprised that it was true.

After all, I’d done the same thing back in March.

But in the years since, I’ve become increasingly convinced that we might be in for some rather wild predictions.

As I wrote then, “I think there is a good chance we’ll be seeing more and more predictions in the coming months, and that the future will be filled with predictions that are literally unverifiable.”

And so, to help me figure out just what these predictions will be, I spoke to Nate Silver, a Stanford University professor and author of Predictably Irrational: A Memoir of the Year in Sports.

In the process of sharing this article, I thought I’d share a little bit about my experience as an analyst.

I’m no stranger to predicting the election, and I’ve done it a lot.

But I’ve never been so surprised by a prediction as I was by this one.

And I’m not just talking about the number of days until the election.

I was working for CBS News as a reporter when I first saw a story about the first presidential debate in 2000.

It turned out to be a great time to do predictions.

The debate was held on October 5, 2000, and the audience was nearly as big as the one that watched the previous presidential debate.

It’s the kind of show where you can make predictions about everything.

But it was also the first time I was working with a major television network.

The networks were not interested in predictions; they were interested in ratings.

And as a result, the ratings were quite low.

What was interesting was that CBS was very much a public company, so they didn’t have to worry about the impact of the debate.

But since they were the one broadcasting it, I was able to get an accurate forecast for what the audience would be and how many people were watching it.

And, I’m sure you can guess what I found out.

CBS predicted that the audience for the second presidential debate, held in Miami, would be 8 million people, or 8.4 million viewers.

The audience for that debate in Denver was just over 8 million.

So there was a very, very, large audience watching that debate, but the number was a little off.

What CBS didn’t do was predict the election outcome.

That was up to them.

But they predicted a big number.

I didn’t get a chance to sit in the audience that night, but I remember hearing that the first debate had an audience of about 5 million people.

The network then went ahead and did a poll, which predicted that if they got 10 percent of the votes, they’d win.

So CBS got 10.5 million viewers and beat the networks.

The network’s prediction, which they published in its own magazine, was correct.

So the audience had a little over 7 million viewers, and there were over 7.5 and 8 million in the general population.

The question now is whether the network got its results right.

And here’s what I think: First, they did get the numbers right.

The general population has about 6.6 million people in the U.S., which is less than half of the number that tuned in to watch the debates.

Second, the audience is pretty small.

The average age of the viewers was 32.3, which is a bit older than the audience in the 2000 debate.

And third, the average age was more than 6.4.

All of which suggests that there was probably a lot of people who were watching, but that CBS got the data right.

In fact, there is some evidence that the network’s poll overestimated the audience.

In fact, the poll was almost twice as accurate as the actual audience, at about 9.5 percent.

That means that even if the poll had gotten the exact numbers it had predicted, the network would still have gotten the audience it expected.

In other words, if the audience actually came out to watch, it would have been less than 6 million people; in fact, it probably would have looked much larger.

And I think this is a pretty important distinction.

A lot of these predictions are made by people who are working for networks.

And if you think about it, networks are not just going to do polls and give their estimates; they’re going to run campaigns and hire people.

They’re going get to know the demographics of people, and then they’re then going to make predictions.

And so they’re doing surveys and polls and polls, and so they know things about demographics and demographics, and they can make a pretty good prediction about how many votes they think the electorate will give to the candidates.

But if you look at the actual