Whaaaat Is Happening in Iowa?

In the Battle of Iowa, it’s as if the world has turned upside down.

Over the last month or so, a narrative congealed as we neared the voting portion of the 2020 Democratic Primary. The following seemed increasingly likely:

1. Heading into voting, Joe Biden would remain the national polling leader, maintain a polling lead in Nevada, and have a pretty unbreakable wall in the fourth voting state, South Carolina.
2. Bernie Sanders steadily siphoned enough Elizabeth Warren voters nationally and in early states to emerge as the clear progressive candidate to do battle with Biden across the country.
3. Sanders also took Iowa and New Hampshire leads. If he won the first, he was a lock to win the second. And if he won the first and second, he was close enough to Biden to go toe-to-toe in the third, Nevada, and at that point likely win the state.
4. And then it came down to whether 2 and 3 was enough to overcome 1 in March.

Both polling and analysis suggested the above was likely the plot for the next six weeks.

But the polling has changed in Iowa. Therefore, analysis will have to change as well.

---

Let’s get into this surprising turn of polls before we consider the new pecking order in Iowa. Here’s a graph of the candidates rolling Real Clear Politics polling averages in the state since October 29:

It’s first worth noting the reversed momentum of the previous three leading candidates. Warren had her turn at the top, then faded. Then Buttigieg did, and then he faded. Finally (or perhaps not “finally”) it was Sanders‘s long ascent peaking a bit too soon.

The other main notes are A) Klobuchar‘s slow climb perhaps is finally picking up in pace, and B) More importantly, we have a new Iowa polling leader in Joe Biden.

Another way to depict this turn of events, rather than a graph, is a list of recent polls. I’ll go back as far as I can before doubling up on a polling firm:

It’s worth noting that a poll released on Sunday but not included by RCP (though FiveThirtyEight has included it) is a survey from Neighborhood Research and Media. It contains comparable findings to the recent polls documented by RCP:

Thus, it appears Joe Biden, who we all kind of ruled out winning white and liberal Iowa Democrats, is, for the moment anyway, in the lead. The last handful of polls also say Sanders is fading: 23, 20, 18, 14, and 10. And again, for what it’s worth, it appears Amy Klobuchar is making a late push into the top four, hitting double digits in the last two polls and even passing Sanders in one of them. And then there’s Warren and Buttigieg, who keep trucking along in the mid-to-high teens.

So just two weeks before Iowa, topline polling suggests the following Iowa results and their possible implications:

• Klobuchar is looking for a late surge into the top tier. If she finishes in the top three, she dominates in the hugely important “expectations game.” And, importantly, she’ll knock one of the Big Four out of the Big Four. That would be devastating for that candidate’s expectations game.
• Sanders is hemorrhaging support and might be headed toward fourth or fifth. Why the dramatic turnaround? The most obvious answer Warren’s Women Strategy in the days leading up to and including the last debate. (I thought it was the debate’s most effective moment… but was it really that effective?)
• Warren and Buttigieg will be angling for second place, but unless Buttigieg does a lot better than her there, I would suspect Warren picks up Sanders deserters to win the New Hampshire Primary, and then she’s the last candidate standing with…
• Joe Biden, who, with an Iowa win, would likely run away with the Democratic nomination.

So… um… we’ve done it? We’ve figured out the primary?

---

Of course not! We thought we had figured out the narrative before, remember? And what did that get us?

No, the last 13 days before Iowa can be just as tumultuous as the prior 13 weeks.

Let’s return to that Neighborhood Research and Media poll again — the one that had Biden six points up on the field and Sanders all the way back in fifth. Its survey did something clever: it separated the results for us, teasing out the first day of the survey — which was January 14, the day of the debate — and then the subsequent three days after the debate. The Biden difference is stunning. After his (bad) debate performance, here’s what happened:

Whoa! NRM offered a bit of analysis to these numbers, and I found its strong wording pretty remarkable: “Joe Biden is essentially in freefall and despite his current lead, it’s highly unlikely he will win the caucus unless something changes in his favor rapidly in the last few days. . . . Look for Biden to lose his lead over the next few days.“[1]

Meanwhile, Klobuchar was up four points based on the debate, giving further evidence that she could well knock down a member of the Big Four and really throw this primary into a free-for-all, particularly if NRM is right and Biden’s polling peak is behind him, just like the three candidates that led before.

In totality, all of today’s numbers make it really hard to predict Iowa — something I did with almost divine accuracy four years ago. Biden’s in the lead but he’ll likely succumb to the same gravity as the other Iowa leaders, so who the heck wins this thing?

It appears we’re back to Warren and Buttigieg, both of whom just keep scoring in the high teens; saw benefits, according to NRM, from the debate; and have two of the best ground games and footprints in Iowa, which means they’re organized and ready to caucus. So is the Sanders Campaign, who I think will perform closer to his Des Moines Register numbers than the three polls since. I doubt we’ll see the same organizational potency from Biden, who has been more concerned with his southern strategy, or Klobuchar, who simply hasn’t had much cash for most of this primary. And yet, it’s the latter two with the most upward momentum right now according to an average of polls.

The lesson of today: we have no idea what’s going on in Iowa, and I absolutely love it.

---

FOOTNOTE:

[1]When I saw that injected punditry, I followed up on the pollster writing the memo, Rick Shaftan. It turns out he’s a Republican consultant, with a Twitter handle that reads #AllSocialistsAreTheSame. It’d be reasonable to think he’s both A) very anti-Sanders, for obvious reasons, and B) anti-Biden, largely seen as the most electable candidate against President Trump.

Though that doesn’t mean he fudged the numbers, it does appear his sample can be criticized:

Only 56 voters of the 300-voter sample were under 50! And 175, or 58%, was 65 or older. By comparison, Iowa exit polls from 2016 had only 28% of caucus-goers as 65 or older. That’s nothing short of egregious sampling.

In other words, our review of this poll should sound a little like: BOOOOOOOOOOO! Now we know why Real Clear Politics didn’t include it. (The gender split is better; this poll had 183 women, or 61%, while the 2016 exits said women made up 57%.) My takeaway is that Sanders is stronger — and Biden weaker — than this poll suggests, though the poll does fit with trend-lines for both candidates.

All that said, the decline in Biden’s support from pre- to post-debate would be hard to mess around with. It’s something to keep an eye on.