Five Days to Iowa: Second Choice, Organization, & Three Ways to Win

In an effort to better predict the results and impact of Monday’s Iowa caucuses, today I plan on taking a closer look at the contest’s unique dynamics. Three factors in particular need a great deal of examination:

1. Voters’ “second choice” candidates are not important in your normal primary, but in a caucus those can be huge. If after the first round of caucus voting a voter’s “first choice” is deemed unviable — which usually means short of 15% support in a precinct — then that voter must re-align with a viable candidate or leave the caucus. Those realignments make a big difference in overall caucus results.
2. A campaign’s organization is particularly important in a caucus. Though primary results are certainly impacted by what demographics turn out to vote, predicting the turnout of those demographics in a primary is much easier than predicting caucus turnout. Primary polls are open all day and often have early voting, but caucuses are in a tight window on a cold night. A campaign’s ability to get its supporters to turn out to a caucus and then coordinate them while in there is more paramount in a caucus than a primary.
3. Finally, for the first time ever, there will be three different winning categories made public. A candidate might sweep all three — but not necessarily. Considering how the Expectations Game and Spinnability can impact momentum after Iowa, this new paradigm deserves a closer look as well.

Let’s do it.

---

1. Voters’ “second choice” candidates

I think too many media outlets make a mistake when reporting out “second choice” statistics. Usually, they rank the “second choice” candidates from top to bottom without the necessary context. The New York Times, though its poll with Siena College reported out that Elizabeth Warren leads in Iowa’s “second choice” support, also pointed out the most obvious way that’s a misleading category: “Ms. Warren, according to the poll, is the top second-choice preference of caucusgoers, which could lift her candidacy after the initial vote. But that is in part because she is the preferred alternative for those who support Mr. Sanders.”

Because Iowa progressives are so highly motivated, it makes sense than Warren is a popular second choice. However, being the second choice of the many Sanders supporters means nothing, because Sanders is going to clear 15% viability in every last precinct across the state. If we remove all the Sanders voters out of the equation, Warren’s “second choice” numbers would plummet.

No, instead, we must consider the “second choice” leaders of each first-choice candidate.

Of the flurry of Iowa polls to come out in the last week, I like what Emerson College did with this category best, so we’re going to take a closer look at it. (Emerson also has the benefit of being tied with a USA Today/Suffolk survey as the most recent poll. Both were completed on Sunday.) Like the New York Times/Siena poll, it also found Warren comfortably the leader in “second choice” (SC):

		SC Frequency	SC Percent
Valid	Joe Biden	71	15.8
	Bernie Sanders	70	15.6
	Elizabeth Warren	103	22.9
	Pete Buttigieg	64	14.3
	Amy Klobuchar	50	11.1
	I would leave	34	7.5
	Total	392	87.1

But now let’s consider if that matters at all. I went deep into the crosstabs and deployed my not inconsiderable Microsoft Excel skills to give you the following table:

Though Warren leads the field with 23% second choice support, half of that comes from Sanders first-choicers. That significantly reduces her relevant second-choice strength to more of a weakness. Relevantly, half of Warren’s supporters’ second choice is Sanders. No other candidates are as strongly correlated to another.

Meanwhile, most of Buttigieg’s supporters sooner break to Biden and Klobuchar, while Klobuchar’s supporters mostly break toward Biden and Buttigieg. In other words, lots of Biden-Buttigieg-Klobuchar coalitions are possible, but Warren will be out of luck in the second-choice category thanks to Sanders consolidating the left. Therefore, Warren leading the second-choice category overall is nearly meaningless.

Let’s now pair what we’ve learned with a reminder of the last seven polls conducted in the state:

Warren is falling short of 15% statewide, and she’s trending down. She’ll surely get to 15% in her stronger precincts, but there’s a decent chance she’ll just as often fall short. Though she will often be able to piece together a 15% coalition from supporters of unviable candidates, we may see some supporters desert her altogether. And where will they go go? To Sanders, who’s trending up up up.

Based on top-line numbers and this second choice analysis, I think we’re headed toward a big Iowa win for Bernie Sanders.

As for the order underneath him, the moderate lane still has some sorting to do between Biden, Buttigieg, and Klobuchar. Biden can compete with Sanders if Buttigieg and Klobuchar fall short of 15% thresholds and their voters realign behind Biden, but if they’re more competitive it’ll look like a Sanders runaway.

Perhaps today’s second big factor can help clarify…

---

2. Campaign organization

Top-line and second choice statistics aren’t the only factors that can help us divine Iowa’s results. Four years ago, my perfect Iowa caucuses prediction of the similarly crowded Republican Primary noted that “Trump has led 11 of the last 12 Iowa polls, including the last nine in a row.” Yet I still went with Cruz to win the caucuses. Why?

Organization!

“[Cruz’s] ground game is second to none, not just in this cycle, but perhaps in the history of the caucus. One Iowa insider called it “one of the most sophisticated, if not the most sophisticated, organizational efforts this state has ever seen.” His network will get people out to the caucuses. Check out this graphic from the Washington Post, citing the latest Monmouth poll. What percentage of Iowans had personally been contacted by the campaigns?”

I proceeded to show Cruz dominated Trump and other candidates in the category. Twice as many caucusgoers had been contacted by the Cruz Campaign as compared to the Trump Campaign. That contributed to Cruz’s 3.3-point win in the state despite Trump having led by 4.7 points in the RCP average — an 8-point swing.

With that in mind, who’s the most organized this time around? Well, guess what folks… it’s about to get more muddled at the top. Neither Sanders nor Biden has the most field offices. That award goes to…

Pete Buttigieg! His campaign may also employ the most Iowa staff. More offices and more staff usually means more phone calls, more face-to-face recruiting, and more mobilization and organization on caucus day. It also means having sufficient leadership and cohesiveness in a caucus room, a tool that helps hold on to potential deserters while also wooing undecideds and/or caucusgoers from other camps, particularly nonviable ones.

In the fall, evidence of how Buttigieg’s organization helps him could be found at the high-profile Liberty & Justice dinner. The Buttigieg Campaign held the largest rally beforehand (some 2000 supporters partied in the rain) and attracted the most supporters in the building (each armed with matching glow-in-the-dark bracelets).

For a caucus, polls only go so far. The secret sauce is how to ensure those supporters actually show up and stick with the candidate. Organization does that. His organization also looks well primed to to peel supporters from other candidates; look again at the Excel chart from above, and we see Buttigieg is intriguing to supporters of all other candidates. He might be a sort of compromise between older moderates and younger forward-thinkers.

As a result, Buttigieg might therefore thrash polling expectations and join Sanders and Buttigieg near or even at the top of the Iowa results. He could ask for no better result, and he’s smart and wily enough to suppress expectations if that’s within his ability to do so. On Tuesday morning, we might look back at Mayor Pete’s organization as his overlooked weapon.

Meanwhile, a campaign that many are worried will lose the turnout battle is Joe Biden’s. Compared to the other contenders, his Iowa events are few in number, sparsely populated, and unenthusiastic. His organizers are reportedly not as visible. Though his campaign footprint is large, the footsteps reveal a limp; his Iowa campaign has been criticized as too old-school and out-of-touch. Speaking of old-school, that’s precisely where his supporters attended. His aged voters can be counted on in traditional primaries, but Iowa caucuses are cold, loud, and late, three words antithetical to the gerontocracy, a typically dominant voting demographic.

If we consider Buttigieg’s organization strength is Biden’s weakness, it wouldn’t surprise me to see the Buttigieg team more successfully pilfer undecided voters and caucusgoers from nonviable candidates. That could drive Buttigieg into second behind Sanders, whose caucus experiences of four years ago means he knows precisely how to deploy his organization. That’s the scenario toward which I’m now leaning.

---

3. Three winners??

At this point, it’s clear that chaos reigns on caucus night. For the first time ever, however, new precinct reporting rules allow us to see what caucusgoers initial intentions were before the quirky caucus rules catalyze that chaos.

In past years, the only official number reported by Iowa’s Democratic Party was the expected delegate split — called the “delegate equivalent,” a projection of the results from the multi-tiered caucus process that will be ongoing for months at the county, district, and state levels. With 41 pledged delegates out of Iowa, that means we would have had something like Sanders 18, Biden 14, Buttigieg 6, Warren 3, and that would be the end of it.

Now, however, each precinct must report out its initial popular vote and the popular vote after the re-alignment. Iowa will then aggregate each category and report them out alongside the traditional “delegate equivalent” number.

The ramifications for the mid-tier candidates are minor but not irrelevant. There’s a decent chance a Warren or Klobuchar falls short of 15% in most precincts, which could make their “delegate equivalent” a big fat zero. Thanks to realignment, their final popular vote number would be misleadingly small as well. However, under the new rules, they would still be able to claim some measure of potency by pointing to, say, 12% statewide support before the precincts re-aligned.

For the top candidates, these three categories are more important. Say on caucus night, Sanders is as strong before realignment as I think he will be, but Biden ends up inheriting a lot of Buttigeg and Klobuchar support after realignment. Biden might come away with the most delegates, but Sanders might have easily bested Biden in the initial show of support. Sanders could claim a victory, but Biden could as well.

There’s also a far outside chance we see three different winners in these three categories, particularly if support is not distributed evenly across the state. A candidate could run up the score in some areas (either pre- or post-realignment) to increase their popular vote number, but their “delegate equivalent” might not be ideally placed. (Perhaps an analogy could be made to Democrats’ current popular vote advantage in the national election, but Republicans’ better distributed support in the Electoral College.)

Considering the most important effect of the early states is how it determines momentum moving forward into more delegate-rich contests, like the ones on Super Tuesday, how candidates can spin their Iowa results will have a big impact.

---

I’m glad I got all that out of the way! I still hope to have a final Iowa prediction up on Sunday, and I’ll surely reference this post along the way. That will hopefully earn me some column inches when the time comes.

Before then, I hope to find time to make an official PPFA endorsement. Not that you care about such things. See you then, perhaps.