October 27, 2015
Polling has some variance, that whole margin of error thing. It’s a snapshot in time, in this case months before anyone votes. At this point the majority of voters say they haven’t made their final decision. Even some who think they have might think differently later as events change. Sampling methods vary. Live phone interview is considered more credible, but many voters won’t answer calls from unrecognized numbers.
Still, with all those caveats, there’s still some overall consistency. Donald Trump led everywhere until Ben Carson passed him in Iowa and now in one national poll. You aren’t going to see a Huckabee Leads in Iowa headline tomorrow. Speaking of the Hawkeye State, reading polls of their voters over the past few months was usually very straightforward, and remains so on the GOP side.
Though the margins may have differed slightly, they moved in the same direction. Constructing a narrative was easy. Ben Carson catching and passing Trump was visible weeks ago. The underlying data showed him with more upside, higher approval rating, high amount of second-choice mentions, etc. Eventually the inevitable happened. Whether Carson can retain the lead is another matter, but he was going to have his moment in the sun.
Democratic voters are a bit harder to read. The trend line is fairly consistent. Between May and Labor Day, Bernie Sanders made up a lot of ground. Polls from the spring showed him trailing by 40 to 50 points, with Hillary having quadruple his support (think 60% to 15% for example). He spent part of September in a virtual tie with Clinton in the Real Clear Politics average for Iowa.
Buried within that time of September stasis was some inconsistency. Within a week of each other, a CBS/YouGov poll put Bernie up 10, while a PPP survey had Hillary ahead by 21. Huh? The same thing happened in late August. The Des Moines Register, via Ann Selzer had Clinton +7, sandwiched between three other polls with Hillary +25, +28 and +34.
By any standard Bernie was in better shape in September than August and has regressed a bit in October. That much we know. But he’s either still very much in contention (CBS/YouGov -3, Des Moines Register/Bloomberg -7) or in the process of being obliterated (Loras College -38, Monmouth -41). It’s a huge difference.
The Loras and Monmouth polls are slightly newer, though there is overlap between the CBS/YouGov poll and Loras. All were post-debate. Only Monmouth was post-Biden announcement. Given the similarity between Loras and Monmouth, that’s not the reason for the different readings.
For three distinct polling cycles (mid-late August, mid-September, mid-late October) we can observe a 20-30 point gap in polling results. CBS/YouGov and Des Moines Register are Bernie-friendly. Loras is pro-Hillary, as are PPP and Monmouth. Quinnipiac and NBC/Marist like Sanders. Suffolk University and Gravis Marketing like Clinton.
Most of the pollsters have participated in two of the three cycles, and each time there are multiple entrants from each side, so I’m very comfortable saying this isn’t a fluke. It’s also unique to Iowa. In New Hampshire, the variance is 10 points at most among polls being taken in the same window. Given a 5 point margin of error, if one is high by that amount, while the other is low, it can happen.
Nationally, same story. South Carolina, nothing to interpret. In all those instances, using the RCP average works, smoothing out slight outliers and making it clear Bernie is still ahead in New Hampshire, but not by so much that Hillary couldn’t catch him. She’s safely ahead nationally, but given enough time and some luck, maybe Bernie can make it interesting, but probably not. South Carolina is hers. If Biden was around to siphon votes, maybe some drama, but now a done deal.
So why don’t we just ignore Iowa? We know the overall trend is favoring Hillary. We know Bernie’s task got harder when Joe said no. We know Sanders has retained his base and that mainstream Democrats have rallied around the front-runner. We know the math has never worked to push a candidate like Bernie past one like Hillary.
It only works when liberal white voters are combined with at least some minority voters as Barack Obama successfully managed in 2008. Minority voters in Iowa are sparse, so why do we care which polls are accurate? Shouldn’t we worry about national numbers or if he can close the gap a little in South Carolina?
No. Iowa is the only polling that matters. Hillary is a clear favorite, but isn’t invincible if she loses in Iowa and New Hampshire. If she wins even one out of two, it’s almost unimaginable to see how he catches her later (at this point we should not count on an indictment taking Hillary down). We know Bernie is practically on home turf in New Hampshire. He has a decent chance of winning there even if Hillary wraps this up early. He needs Iowa.
All polling is showing Democratic voters like both candidates. Each have extremely favorable approval margins among potential primary voters. However, many voters aren’t paying that much attention yet (GOP voters are more engaged so far). Right now, the immediate narrative is pro-Hillary. She’s surging, she had a great debate, she made Republicans look stupid in the Benghazi hearing, she scared away Biden, she won several new union endorsements.
If you’re a semi-to-not focused Democrat, you’re thinking she has this sewn up and aren’t going to give Sanders much thought as your choice unless you are in one of his target groups or your pals are feeling the Bern. Speaking of those partisans, we know younger voters are inconsistent participators. If you think your guy has a chance, much more likely to vote, organize, pull others along. This impacts left-leaning Independents too, as many primaries are open to non-Democrats.
If the anti-Bernie scenario is correct, and he’s now trailing Hillary 65/25, after never having caught her at her weakest, he’s not going to win Iowa and we are not going to have a Democratic race to talk about next spring. Proceed directly to coronation.
If the pro-Bernie view is accurate, even now, when he’s received nothing but shaky press for a couple weeks and Hillary is flying high, he’s still within range, running much closer than Obama was in late October 2007. Sanders has a LOT of unspent money to blanket the Iowa airwaves over the holiday season. 10 points is nothing. Carson went from 5 behind Trump to 10 ahead in a couple of weeks.
So who’s right? I think the Bernie-friendly view is more accurate. Ann Selzer has a really good track record, and aside from her, the pollsters favoring Bernie have a better record than those favoring Hillary. Perhaps more importantly, it doesn’t make sense that candidates with roughly even approval ratings would have wildly disparate support. Demographically, Iowa is pretty Sanders-friendly too.
By January if not much sooner, journalists and pundits will find themselves very bored with Inevitable Hillary. The next debate (November 14) will feature the main candidates more with Jim Webb and Lincoln Chafee out of the picture. Sanders will find plenty of opportunity to attack, having already (by loudly objecting to discussing her emails) made it clear it’s all about the issues.
Clinton is the clear favorite to win the nomination. Her present odds are at least 85-90%. However, I also think we get to talk about the Democrats a bit longer. Bernie is probably trailing in Iowa on the order of 52/38 today, a margin he is likely to close at least once in the next three months. At this point I would say Hillary is a 3 to 2 favorite to win Iowa. If that happens, race over. But there’s a 40% chance the nomination fight continues for a bit longer. At a minimum we should have some drama on caucus day.