FiveThirtyEight.com founder Nate Silver speaks with ABC’s George Stephanopoulos about his predictions for Tuesday. " Clinton has about 270, so she's one state away from potentially losing the electoral college," he said.
SILVER: So, Clinton is a lot weaker in the Midwest where four years ago President Obama was leading in Ohio by four points. Clinton's probably a couple of points behind there. Iowa, maybe the best poll in the country, the Des Moines Register Poll showed her down seven points in Iowa, a state she'll probably lose.
So the demographics for Clinton don't actually work as well when you underperform among white noncollege voters. That's a good group...
STEPHANOPOULOS: Even though the national polling is about the same.
SILVER: Even though -- so but it's contesting the electoral college and her electoral college
polling in the swing states is a little bit weaker than Obama's. So in some sense, it's a deceptively large or small lead for Clinton in some ways whereas Obama had a bigger lead electorally than you'd thing.
The other thing, too, is that we see lots of polls that show numbers like Clinton 44 percent, Trump 40 percent. If you only have 44 percent of the vote that means you're vulnerable if most of the undecideds break in a certain way whereas four years ago it was like Obama 49, Romney 46. So in that sense both candidates still need a good turnout on election day and still have their work cut out for them.
STEPHANOPOULOS: This year, we've seen a high early vote, about 41 million Americans
have already voted. You don't pay that much attention to the early vote.
SILVER: Well, look, our model is designed to look at the public polling. And if you want to look at anything else that you want, and the early vote, then that's fine and I think there are definitely some decent signs for Democrats in Nevada, for example.
But Democrats also in 2014 told themselves a lot of stories about how they would be saved by the early vote and got wiped out across the board in the midterms, the polls actually overestimate how
well Democrats would do., So I would be a little bit careful.
With the exception of Nevada, which is a hard state to poll, you know, that should be incorporated in theory in the polling numbers.
STEPHANOPOULOS: Another variability that we've seen here right now. There have been a lot of other forecasts out there, Princeton Election Consortium, Huffington Post, several others -- and The New York Times. Yours is much more bullish for Donald Trump and more cautious on Hillary Clinton than theirs are. Why?
SILVER: Because we think we have a good process and, look, you have some forecasts that show Clinton with a 98 or 99 percent chance of winning. That doesn't pass a commonsense test, which is we've seen lots of elections where there's about a three-point polling error. In 2012, in fact, Obama beat his polls in many states by about three points. If Clinton were to beat her polls by three points and you see something we call a borderline landslide, but if it goes the other way, and all of a sudden Trump could very easily win the electoral college.
I mean, it's all based on history. People have different ways of interpreting history, but, you know, if you think a three-point lead is going to be safe 98 or 99 percent of the time, then you probably didn't design the model in a good way.