Nate Silver’s primary election forecasts highlight just how uncertain the outcomes are this time. A boom of polling in Iowa and New Hampshire shows some very close races for both parties. And most Republicans are okay with Ted Cruz’s Canadian birth, but many still doubt Obama’s eligibility. This is HuffPollster for Friday, January 15, 2016
DESPITE CHALLENGES, FIVETHIRTYEIGHT IS FORECASTING THE PRIMARIES - Nate Silver: “Forecasting primaries and caucuses is challenging, much more so than general elections. Polls shift rapidly and often prove to be fairly inaccurate, even on the eve of the election….And races with several viable candidates, like the one the Republicans are contesting this year, are especially hard to predict. Nonetheless, we’ve developed a pair of relatively simple statistical models that we hope can shed some light on the upcoming primaries. These models don’t claim to have a lot of precision, sometimes showing very wide potential ranges of results for each candidate.” [538]
Two different models sometimes predict very different outcomes. - Silver describes two models: One using only state-level polling data -- “polls only” -- and the other using state polling data plus endorsements and national polling data -- “polls plus.” According to the polls-only forecast in Iowa, Donald Trump and Ted Cruz are neck-and-neck with a 44 percent and 42 percent chance of winning the caucus. Incorporating the other factors bumps Cruz to a 51 percent chance of winning and drops Trump to 29 percent. In New Hampshire, Trump has either a 57 percent chance of winning in the polls only model or a 39 percent chance in the polls plus. On the Democratic side, Hillary Clinton goes from a 66 percent chance of winning the Iowa caucuses in the polls only model to an 82 percent chance in the polls plus model. Bernie Sanders has either a 72 percent chance (polls only) of winning New Hampshire or a 43 percent chance (polls-plus). [538]
What to make of widely varying forecasts? The uncertainty in these forecasts highlights Silver’s point that forecasting primary elections is very difficult. Relying on polling data is much more problematic in primaries than in general elections because primary polls are more volatile and less accurate than general election polls. Additionally, there are widely varying numbers of polls across different states -- and some states’ primary elections have no polling at all. In the 2012 GOP primary, only 21 states had polling data available. Nine of those had fewer than 10 polls. And in Iowa, South Carolina, Ohio, Pennsylvania and Texas the polls pointed toward the wrong outcome. In 2016 it’s likely that many states will again have little or no polling, and there will probably be at least a few polling misses. Combine the volatility or absence of state-level polls with the striking differences in the two models Silver has created, and it’s clear no one knows what to expect.
HUFFPOLLSTER VIA EMAIL! - Get the latest polling updates in your inbox every Friday morning! Just click here, enter your email address and click "sign up." That's all there is to it (and you can unsubscribe anytime).
GAP BETWEEN SANDERS AND CLINTON CLOSING IN IOWA - Pollsters have been hard at work in the post-holiday period, and although there’s variation in the actual numbers, it appears that Hillary Clinton is losing a grip on her lead while Bernie Sanders is moving on up. Polls released by NBC/WSJ, Bloomberg/Selzer/DMR and PPP show Clinton maintaining a slight lead of 2-4 points. At the same time, releases from ARG and Quinnipiac show Sanders up. The variance is likely partly explained by differences in methodology. But overall, the race has closed: Clinton is holding on to about 46 percent of the vote and Sanders has surged to 42 percent, according to the HuffPost Pollster aggregate.
Bernie Sanders maintains lead in New Hampshire: Sanders holds an advantage over Clinton in all four of the polls released this week, although that advantage ranges from as little as 3 points in an ARG poll to 14 points in a Monmouth poll. According to the HuffPost Pollster aggregate, Sanders has 50 percent of the vote while Clinton trails with 44 percent. [NBC/WSJ/Marist, Fox, Monmouth, ARG]
GOP DEAD HEAT IN IOWA - Ted Cruz and Donald Trump are battling it out for the lead in the Iowa polls. According to the HuffPost Pollster aggregate, the two are tied at about 28 percent of the vote. This week’s poll releases ranged from a 3-4 point lead for Cruz to a similar margin in Trump's favor. Marco Rubio is the only other candidate averaging double digits with about 13 percent. A majority of Iowa caucus goers are still making up their minds, so there could be further upheaval in the next two weeks. [NBC/WSJ/Marist, Fox, Quinnipiac, ARG, Bloomberg/Selzer/DMR, PPP]
Trump continues to dominate New Hampshire polls. In the five polls released this week, Trump has an advantage that ranges from 11 points to 20 points. According to the HuffPost Pollster aggregate, he has 29 percent of support in the state. Marco Rubio is second with 14 percent, Ted Cruz is in third with 12 percent and Chris Christie and John Kasich are stuck just behind with 11 percent. [PPP, NBC/WSJ/Marist, Fox, Monmouth, ARG]
REPUBLICANS STILL AREN'T CONVINCED OBAMA IS AMERICAN, BUT HAVE NO QUALMS WITH TED CRUZ'S CANADIAN CITIZENSHIP - HuffPollster: "Donald Trump famously questioned President Barack Obama's citizenship in 2011 and demanded he publicly release his birth certificate. Now, Trump's at it again, this time questioning the citizenship of Canadian-born Republican presidential rival Ted Cruz. A new HuffPost/YouGov poll finds that 53 percent of Republicans still doubt Obama's citizenship. At the same time, an overwhelming 70 percent don't have any doubt Cruz is American and eligible to be president. More interestingly, most of the Republicans who don't think Obama was born in America aren't concerned about Cruz's citizenship. Sixty-four percent think Cruz is eligible to become president, while just 18 percent think he's not. Another 18 percent are not sure. [HuffPost]
WHY EARLY POLLING STILL MATTERS - Lee Miringoff: "Poll Watcher Season is upon us big time. And, with it comes both the good and the bad. Each election cycle resurrects some oldies about the failings of public polls and typically ushers in a few new critiques. Expect 2016 to follow the same pattern….Pre-election polls are not predictive even though many continue to treat them that way. Common sense tells us that a poll conducted substantially before voting cannot be predictive. Without pre-election polls, we would be clueless about the surprising and lasting electoral appeal of Donald Trump….Public polls help us understand the emergence and decline of different candidates and also let the public in on the secret that campaign pollsters and strategists see in their private poll data. [HuffPost]
MORE OF THIS WEEK'S POLLS
-The public's worries about terrorism are starting to subside [Gallup]
-Democrats continue to shift further to the left. [Gallup]
-Republicans are on board for tougher background checks, as long as they don't think it's coming from Obama. [HuffPost]
-National polls show Trump maintaining a nearly 20-point lead over with Cruz. [HuffPost]
-Sanders is making gains in national primary polls, but Clinton still has a solid lead. [HuffPost]
THIS WEEK'S 'OUTLIERS' - Links to the best of news at the intersection of polling, politics and political data:
-Ted Cruz is poll testing attacks on Donald Trump in Iowa. [RCP]
-GOP pollster Dave Sackett warned the NRCC that Ted Cruz's nomination could hurt down ballot candidates. [Politico]
- Amy Walters looks at how turnout could affect Iowa and New Hampshire. [Cook Political Report]
-David Wasserman breaks down the number of delegates in each state. [Cook Political Report]
-The anti-establishment vote is on the rise, but establishment voters are still holding their own in early primary states. [Monmouth]
-Keith Gaddie and Kirby Goidel look at the decline of the working class white vote in American politics. [HuffPost]
-Republican primary voters think of Ted Cruz as a "true conservative," but Donald Trump as "strong." [YouGov]
-Margie Omero (D) argues that partisanship is behind Americans' continued divide over gun laws. [HuffPost]
-Mark Mellman (D) debunks a poll Donald Trump used to justify his attitude towards Muslims. [The Hill]
-Shawn Zeller asks if 2016 could be the final year for the telephone poll. [Roll Call]
-Stephanie Slade looks at why polling has become less accurate and more riskier. [Reason]
-The American Association for Public Opinion Research releases a full report on address-based sampling. [AAPOR]