There is an interesting interview with Nate Silver of FiveThirtyEight at Adweek that includes his prognostications concerning the fast approaching election.
I was surprised that this hadn’t already been reprinted on DKos so I took the time to search to see if I was reprinting old news. Nothing was found so I am dairying this now.
Nate Silver was asked about Trump’s chances to win the GOP nomination, Nate Silver still did not see him carrying his current support to victory:
You've made it clear that you don't believe Donald Trump has a legitimate chance to win the GOP nomination. He's leading the polls … why are you so confident in his demise?
I don't think his chances are zero. You have to be very careful about saying they're zero, but I think they're lower than 20 [percent] or 25 percent. Maybe they're 10 percent. Maybe they're 8 percent. I'm not sure, somewhere in that range. You've never had a candidate like Donald Trump win a nomination before, at least not in the modern era, which is 1972 or so, onward, which is when people started voting in primaries and caucuses. Usually the party is able to find ways to shuffle candidates who are openly running against the party to prevent them from winning their party's nomination. A second reason is that I think people are paying too much attention to the polls. Historically, polls, at this point, are not very predictive at all.
Asked about Clinton’s chances of not winning the nomination, Nate replied that in essence, Sanders had less of a chance than Trump at this stage:
Is there any chance Hillary isn't the Democratic nominee?
I think you would have to have some type of renewed scandal or health problem or something like that. I could see Bernie Sanders winning a few states. New Hampshire is still very close. But her chances have to be in the range of 90 [percent] to 95 percent. Trump has more of a chance than Bernie.
www.adweek.com/...
According to Nate, Trump has an 8 to 25 percent chance of securing the Republican nomination and Sanders has a 5 to 10 percent chance of nabbing the Democratic nomination.
However, he was careful to point out that predictions are useless until nominees are chosen, so we will see how close to accurate his numbers are as the primaries progress.
The general rule of thumb is that predictions are fairly useless until the nominees are chosen. Maybe if Ted Cruz or Marco Rubio sweeps through Iowa and New Hampshire, and that nomination is effectively decided by March, then maybe by April we'll be ready. If it turns out to take longer than that, which it very easily could, then we might wait longer before launching. The thing people don't realize is that the reason why I get to look smart is because we wait until we are pretty confident. Never mind getting 50 states right—we'd be happy with 47 or 48. Sooner or later it's inevitable that you come up on the wrong side of a 60/40 bet. We put probabilities around things for a reason.