If you haven’t yet, please read:
Welcome to the end of the road. We began with the polling leaders. We conclude with a few candidates who need their cards to come up almost perfectly. This group is quantitatively different from the contestants we aren’t mentioning at all. Those individuals need even more to happen, so we’re skipping them until there are dramatic changes in the race.
The debate made Castro a topic of conversation. For 24 hours. Then Harris v. Biden wiped him away as though his duel with Beto never happened. Real Clear Politics lists five national polls taken completely after the debate nights. Castro has a single 3% result. Plus four 1% readings.
This rousing 1.4% average is actually good for 8th place, meaning Castro is in the top third of all candidates. Like I said, he needs a few breaks. One number set speaks to his opportunity and challenge.
Going back to the CNN poll we’ve used in this series, respondents who saw the debate viewed Castro favorably by a 63/12 margin. That’s a solid ratio of over 5:1, with 75% of respondents having an opinion. It’s very similar to Kamala Harris’s favorability in the overall survey, and stronger name recognition than Pete Buttigieg.
The catch is voters who didn’t watch gave Castro 24/16 marks. So a full 60% didn’t know enough about him to know if they like him. Those who think they do were only 3:2 favorable.
You can think of his task as first making more voters aware of him, and second, having those who are aware move him up their list, as he still only wound up with 1% support in this poll. Which voters can he target to do this, and with what issues?
His debate moment was centered on immigration, not one of the issues CNN surveyed on. It’s reasonable to think this could be his lead, though I’ve yet to find a poll that suggests this is the top issue for most Democratic voters.
On the economy, health care, and climate, Castro was favored by 1% of the audience each, equal to his overall support. He received 2% on race relations. At this point, we can’t point to issue categories as a leading indicator of future receptiveness.
With 1% support, the sample size is too small to start dividing by demographics. Instead we’ll look to the question asked about which candidates the respondents wanted to know more about. A full 16% wanted more info on Julian. This was the 5th highest total, trailing Harris, Warren, Buttigieg, Booker.
CNN asked the same question two months prior. Castro was only named 3% of the time. His increase is by far the largest. We’re going to use interest level as a demographic proxy to analyze the way we did with the front-runners’ actual support.
Perhaps due to his stance in favor of decriminalizing illegal border crossing, Castro is more interesting to liberals than moderate/conservatives by a 19/12 margin. This places him squarely on the competitive left flank of the contest.
Women (21%), Men (13%)
Under 45-years-old (19%), 45 and over (14%), 65 and over (11%)
Non-white (21%), White (13%)
White College-educated (19%), White no college (9%)
It’s important to remember the difference between white voters with and without degrees is often larger than the split between different ethnic groups. Castro is an example of the norm in the 2020 Democratic contest.
He’s one of the few non-Bernie candidates to appeal noticeably more to younger voters. His path likely involves Sanders slipping, with Warren not serving as an adequate replacement for some of his supporters. Castro would need to combine these young, liberal voters with some upscale suburbanites to build a coalition.
Beto had a rough debate. He only got 3% support in the CNN poll, but that was 1% among watchers, 5% among non-watchers. He’s slipped noticeably since entering the race and did nothing on stage to indicate this is a fluke.
His most likely outcome is dropping out after Iowa. Running against a generally popular primary field is proving very different than challenging the polarizing Ted Cruz. It’s just that, welllllll, you never know. Particularly when the lagging candidate has previous poll support and previous fundraising support.
With Castro, we substituted interest for support to measure demographics. For O’Rourke, we’re going back in time. To a Quinnipiac survey from late March. When Beto was at 12%, and a top-tier contender. In a world where he’s seen as viable, where is his foundation?
Quinnipiac didn’t ask about specific issues. He didn’t register anywhere in particular in the CNN survey either. Beto’s campaign is lighter than average on policy detail. For now, we’re going to assume the restoration of his appeal wouldn’t be strongly issue-based.
His potential supporters are a bit more centrist on average:
Very Liberal: 9%
Somewhat Liberal: 12%
This indicates that he’s a possible backup for Biden voters who are wary of how far left the field is moving and worry Uncle Joe isn’t up to the task.
In March, he gathered 13% of men, 11% of women. In June he retained 1% of men, 4% of women. So either there really isn’t much of a gender split, or his more solid supporters are female, but as his audience expands, the increase is weighted towards men. You probably shouldn’t expect a single gender to cause a Beto Restoration.
Younger is better. At least for Beto. In March, he drew 15% of voters under 50, only 9% on the other side of the line. In June, he was at 4% under 45, 2% over. He shares more in common with Bernie and Castro, than Buttigieg, Booker, and Harris.
In March, Beto got the support of 16% of black voters, 9% of whites. In June, he was down to 1% among white voters, while keeping 5% of non-whites. This is consistent with other surveys. Whether measuring specific groups within voters of color, or as an entire set, O’Rourke always does better than with white voters. Any successful coalition will include a high-percentage of non-white voters.
He doesn’t have a clear split across income or education. In March, he was at 12% under $50k, 12% over $100k, 14% in between. The June numbers show him better (4%) under $50k than over (2%). I think this is like the gender issue, where there’s a difference when his support is very low, but not in a world where he’s contenting.
At the moment, Beto registers 2% as the candidate most likely to defeat Trump. The same as Booker and Buttigieg, noticeably trailing the lead group. In order for him to contend, this will need to change. He doesn’t get points for being a “first” at something, and is going to get outflanked on the left, so electability is key. His suitability to voters of color makes him an option for those who are worried a more groundbreaking candidate could lose in November 2020.
Beto needs to improve, or none of this matters. But if he does, there’s a path.
There’s less separating the Minnesotan from the herd of unmentionables than anyone else we’re looking at. She’s at 1% in the RCP national average. Right between Tulsi Gabbard and Andrew Yang.
So why do we care about her? A few reasons:
Iowa. Mind you, it’s not helping much yet. She’s averaging 2.7%, with a high of 4% in an early June CBS/YouGov poll. Proximity is not a guarantee of results. It doesn’t hurt though. It’s easier for her to camp out and do heavy retail campaigning. If she does gain a bit of traction, it will leverage faster. The YouGov poll showed 23% of Iowans are considering her as an option.
Biden. We don’t know yet if he’s going to recover completely, stop the slide where he’s at, or bleed out. If he falls off, there are limited alternatives for his audience of more centrist, older, white voters. His remaining African American support may go to several others before Klobuchar, but besides a version of Beto that doesn’t exist yet, there aren’t that many alternatives.
Different data set. I’m planning a separate post about this. Writer/statistician Bill James has created a new approach to polling. He’s tested this over the past few months. The idea is to compare four candidates at a time, to get a better idea of how they actually stack up. Right now, most of the field is in the same blob of nothingness. When they get compared against each other, gaps appear.
James doesn’t claim his Twitter polling audience is representative of the national or democratic primary electorate as a whole. He and I are well aware of the potential biases. But his surveying has proven predictive over the past couple months, catching Warren’s rise before it happened in the traditional polls.
With all of the above caveats in mind, Klobuchar currently places 5th, with the equivalent of 5% true support (Warren is leading at 19%), on his measurement. If Klobuchar is the third or fourth choice of many voters, she could score well here, while performing badly in regular polling. Some candidates ahead of her will falter over the next few months.
Beyond the possibility that James caught something here—his track record in other fields is excellent, Klobuchar has one more advantage.
She’s not up for re-election until 2024. If Booker, Gabbard, or Jay Inslee want to remain in office, they’ll need to run for their current posts in 2020. O’Rourke and Castro are both being urged to challenge John Cornyn’s Senate seat in Texas. John Hickenlooper is passing up a chance to contest Cory Gardner’s seat in Colorado.
Odds are we’ll look back on these paragraphs with a chuckle. However, if Klobuchar makes things interesting in Iowa, and perhaps winds up with a VP nomination, the reasoning was contained herein.