At this time of the pre-primary season, the most important pollster is J. Ann Selzer. Her surveys for the Des Moines Register/CNN are considered the most accurate of anyone who polls Iowa. Over the weekend, her latest poll dropped, and it’s the clearest indication yet that we have a four person race for the nomination.
Nationally, Joe Biden is still the front-runner. New surveys from Nevada and South Carolina are reinforcing the idea he’s still very strong in the two early states which actually have voters of color. A ton of voters are still very much open to the idea of voting for Elizabeth Warren. Bernie Sanders has firmed up his base, and is in striking distance of the lead in the first three states.
The story right now is Pete Buttigieg. He’s up 9 points on Warren, 10 on Biden/Sanders in the new Selzer poll. When the same survey was done two months ago, he was at 9%. Now he’s at 25%. If the caucus were today, and this survey was accurate, we’d have a whole different contest.
This prompts a couple of questions.
Was this foreseeable? Yes. Previous DMR/CNN surveys from June and September showed this was very possible.
If Mayor Pete should stumble before the finish line (these next 11 weeks will be an eternity for him), who besides Warren, Biden, or Sanders could benefit? Is there another sleeper hiding in plain sight?
Let’s go through the data to see.
NOTE: The DMR/CNN poll is the strongest so far for Buttigieg. Just because it’s usually the most accurate doesn’t mean we should ignore all the others. He is leading in the Real Clear Politics average too, and is over 20% in all three November surveys, with a clear upward trend line. This ain’t an outlier. We’re going through it in more depth because I like the format and track record, but don’t assume he’s necessarily up by this much right now. It really doesn’t matter if he’s ahead by 9 points or three.
We’re going to start with what gets published in the headlines and then gradually peel back the layers.
First Choice Standings:
Not Sure 5%
*Deval Patrick was not surveyed. Entered the race after the poll was in progress.
**Among the zero percent candidates, Delaney was the only one without a single supporter. The others were between 0.1-0.4%
Viewed this way, the standings seem clear. Mayor Pete is at the top, with a big target on his back. The three established national candidates are in a cluster waiting for him to drop. Amy Klobuchar is on deck, but has a lot of work to do. Once promising Kamala Harris is clustered with never really launched Cory Booker, and the three Honey Badgers.
Mike Bloomberg is off to a slow start. Michael Bennet is the strongest of what was once a full pack of centrist-ish middle aged white guys, but that’s not saying much. The others are buried.
Now, let’s add in second choices. Remember, in Iowa, where a candidate must reach 15% support in a given caucus precinct to be able to officially count it, second choices matter, even if they don’t become first choices ahead of Caucus Day. Then there’s the obvious part of a candidate having more than two months to move up from second to first in a voter’s mind.
First Choice + Second Choice Standings
This isn’t a terribly different reading, but you quickly notice Warren closes the gap with Mayor Pete. He’s had a very good few weeks in Iowa, with a couple successful bus tours. She’s had a tougher few weeks, with the difficult rollout of her health care plan. Wouldn’t take much to tip the scales again.
Though her team and donors clearly had higher hopes a few months ago, Harris does distance herself from the pack she was in for first choice options and mostly catches up to Klobuchar here. It’s a reminder there is interest in her candidacy, even if she’s doing a poor job capitalizing on it.
You can also clearly see anyone at the bottom is at best somebody’s third, fourth, or fifth option. It’s not just a matter of Biden imploding to get them a shot.
We’re going to reveal another layer now. The candidates voters are considering. There’s no limit here. A few voters didn’t even have a first choice. Another few didn’t have a second. Some won’t have anyone else to consider, while others might have another 5 to 7 in mind.
With this many candidates, and this far into the process, if a candidate isn’t even under consideration yet, their odds of being a first choice on Caucus Day for that voter are slim. It’s also a stretch to figure they’ll get the second choice nod from a voter who’s first choice didn’t reach 15%.
This is where we find out who could pull an inside straight and still do something, and who is likely fooling themselves.
First Choice + Second Choice + Considering Standings
Suddenly, the actions of a number of candidates begin making sense. We still have Buttigieg and Warren at the top, with two-thirds of Iowans being open to each of them. This is an enviable position to say the least. It means they both have room to stumble and recover before Caucus Day.
Don’t believe anyone who tells you Biden or Sanders can’t easily win Iowa. If more than half of voters would at least consider you, and 25-30% is likely plenty to win, you’re in good shape.
As tempting as it is to suggest Kamala Harris should just slink back to California and forget this whole presidential campaign thing ever happened, she’s not about to give up with 46% of Iowans at least willing to still think about her. As you’ll see later, this is a very soft 46%. I’d rather be Klobuchar right now.
But it does remind us if Harris could somehow put a good few weeks together, she’d matter again. When her advisors cite John Kerry’s 2004 comeback, this data shows they’re not crazy, though he did have less competition.
Yang has progressed further in Iowa than it looks. On the top line, he only moved from 1% in June to 3% in November. However, the voters considering him jumped from 15% to 33%. Of all the non-traditional candidates, he’s the highest.
Castro needs to exit. This isn’t going to happen for him. He’s not in this week’s debate. He won’t make the next one. He’s holding on to that 18% consideration number as a reason to stay in for now.
One way to test the strength of consideration numbers is to compare to favorability ratings, most specifically, very favorable. In this poll, DMR/CNN was kind enough to separate very from mostly favorable and very from mostly unfavorable. In a field this large, a candidate needs a very favorable rating from most voters who are going to pick them as a first choice.
A regular mostly favorable might do as a second choice swap at the caucus precinct, but you need that 15% first to qualify to grab excess voters. So I’m listing them in order of very favorable rating. In parentheses, you’ll see their +/- between very favorable and very unfavorable.
Candidates with negative ratios are in very bad shape.
Very Favorable Standings
Buttigieg 42% (+36)
Warren 36% (+25)
Sanders 31% (+19)
Biden 24% (+10)
Klobuchar 20% (+12)
Harris 16% (+1)
Booker 14% (+5)
Yang 11% (+1)
Steyer 10% (-5)
Castro 8% (-3)
Gabbard 7% (-10)
Bennet 5% (-3)
Bullock 5% (-3)
Bloomberg 5% (-25)
Williamson 4% (-21)
Delaney 3% (-9)
Sestak 2% (-5)
The last table showed Biden’s upside. This one is the reverse. His popularity numbers are similar to Klobuchar’s. They both primarily appeal to older voters. They’re both stronger with less liberal voters. He’s well ahead of her mostly because people think he has a better chance of beating Donald Trump.
While more voters are considering Harris, and a similar amount are thinking about Booker, Klobuchar is better positioned. She also benefits from having her support condensed into older, more moderate voters, so she can target a particular place on the board. She’s the candidate most capable of taking a leap.
Nobody is positioned as well as Buttigieg was in September. Even then, with a first choice of 9% and first + second of 18%, he was at least being considered by 55%, and had a 33% (+28) very favorable rating.
His top line support dropped from 15% in June to that 9% in September, but all of the underlying numbers were steady or improved. That was the indication that any bit of momentum could launch him forward again. Among the lower tier candidates, only Klobuchar and Yang have universally improving data.
It’s more likely Buttigieg will take a step back due to more scrutiny before he takes another one forward. Keep in mind, he’s not at his ceiling yet. He leads in every measure we looked at here. If voters start thinking he’s as good a bet against Trump as Biden, he’ll climb further. Selzer’s data already shows him as the best ideological match for Iowans. One last measure to take a look at:
Only 7% of Iowans have a very favorable opinion of Tulsi Gabbard. But 6% of them have her as their first or second choice. If someone likes her, she’s at or near the top of their list. Conversely, 14% of Iowans are very favorable on Cory Booker. With the same 6% first or second choice rating.
Booker is stuck because the majority of people who really like him still have him third or lower on their preference list. Gabbard is stuck because most Iowan Democrats can’t stand her. If she had Bernie’s favorable numbers, with her conversion percentage, she’d be a strong contender.
This gives us the Booker-Gabbard Index to measure conversion of support. It’s a simple calculation. First place support + second place support, divided by very favorable rating. Booker is 0.43, Gabbard is 0.86. In parenthesis is the index number if you’re just comparing first place against very favorable.
Here’s the field:
Biden 1.16 (0.63)
Warren 1.00 (0.44)
Buttigieg 0.93 (0.60)
Sanders 0.90 (0.54)
Gabbard 0.86 (0.43)
Harris 0.63 (0.19)
Bloomberg 0.60 (0.40)
Klobuchar 0.60 (0.30)
Steyer 0.60 (0.30)
Yang 0.45 (0.27)
Booker 0.43 (0.21)
Williamson 0.25 (0.00)
Bennet 0.20 (0.20)
Castro 0.13 (0.00)
Bullock 0.00 (0.00)
Delaney 0.00 (0.00)
Sestak 0.00 (0.00)
We don’t care about the two groups at the bottom. They combine limited support with terrible conversion. On top, Biden is dipping into the pool of voters who only see him as mostly favorable for first or second place support. This indicates he either needs to raise Iowan opinion, or continue convincing them he’s the most suited to defeat Trump.
In between, Warren, Buttigieg, and Sanders are all doing a good job of converting, without needing to rely on voters who see them as less than very favorable. They also have a batch of voters who see them as mostly favorable, who are at least willing to consider them. Figure that’s the pool of voters who would go to them in a 15% Caucus Day scenario.
Of the four leading candidates, the overall scope of data has Biden in the worst position, and Buttigieg in the best, with Warren barely trailing him. Do not ignore Bernie. The DMR/CNN results are plenty good for him. Plus a new CBS/YouGov survey has him (22%) in an effective 3-way tie with Biden (22%) and Buttigieg (21%.)
Yang and Booker have room for growth, though likely too far to go. After spending a ton of money, Steyer has done well to convert interest to support, but hasn’t built enough interest to truly contend. Bloomberg is following in his footsteps, with good conversion, just not enough support.
Harris and Klobuchar seem similar here, but they’re going in opposite directions.
June: 31% (+26)
Sept: 23% (+14)
Nov: 16% (+1)
June: 12% (+8)
Sept: 17% (+11)
Nov: 20% (+12)
June: 0.65 (0.19)
Sept: 0.70 (0.38)
Nov: 0.63 (0.19)
June: 0.25 (0.17)
Sept: 0.41 (0.18)
Nov: 0.60 (0.30)
Klobuchar’s very favorable number has almost doubled, while her conversion percentage has more than doubled. Meanwhile, Harris’s very favorable number has dropped in half, while her conversion rate to first place choice is still weak after having some temporary improvement. At this point, Klobuchar is the clear #5 option, only needing a Buttigieg or Biden stumble to become an actual Iowa contender.
Tomorrow we’ll do our debate preview. Everything we’ve discussed here has already occurred to the campaigns. There’s a direct link between these numbers and what they’ll attempt to accomplish on stage.