It would be funny at this point to simply say, yes, correct and end the article. But that is not The Truth™.
We spend a lot of time looking at polling because there are insights buried in the numbers beyond their ability to predict electoral outcomes. The rise of Donald Trump in 2015 may end up meaning nothing in terms of who the Republicans nominate, but it tells us a lot about the state of the party who chooses the nominee and a little bit about the media, too. As elections near, polls become more predictive, and our attention to them bears more predictive fruit.
As the 2008 and 2012 cycles show, however, even at this point those polls are not super predictive. Consider New Hampshire in 2008 and 2012. In 2008, John McCain pulled out a win over Mitt Romney. In 2012, Romney won easily. We took Real Clear Politics data from those years and figured out how much the eventual winner trailed the person in the lead over the last 60 days. In some cases, like in 2012, that was the same person -- Romney. His line stays at 0 percent because he never trailed. We also tracked the person who was leading 30 and 60 days out, giving us these graphs.
In 2008, McCain was 15 points behind the leader (Romney) 30 days out. He went on to win. In 2012, Romney led easily the whole time. But: 15 points! That's more than Trump's current lead.
Iowa is even messier.
In 2008, Mike Huckabee, the eventual winner, made a surge to pass Romney, who led two months out. In 2012, the leader 60 days out was Herman Cain, who didn't even make it to election day. The eventual winner, Rick Santorum, was still more than five points back right before the election. The leader 30 days out, Newt Gingrich, collapsed.
Part of the weirdness in Iowa, of course, is Iowa’s unusual process. It can be hard to gauge how the caucus itself will go and who will show up. Turnout in general is a question mark, of course; Trump is expecting that his popularity will lead to a groundswell of infrequent voters heading to the caucus and to the polls. If that happens, the polling may not predict it.
In the Democratic race in 2008, we can see similar uncertainty as the election neared.
Iowa was a back-and-forth: Clinton led 60 days out, Obama (barely) led 30 days out, and Clinton then regained the lead -- only to eventually come in third. In New Hampshire, Clinton led the whole time -- until Obama won Iowa and he surged ahead in the New Hampshire polls, only to lose.
We're less than a month out from Iowa and slightly over a month from New Hampshire. Want to see how the polling now (or, really, as of the most recent RCP polling average) compares with polling in the past? We made a tool for that. Click the buttons. Again, the lines indicate how much each candidate trails the leader.
Make whatever predictions you'd like, of course. Just don't be surprised if you're wrong.