DU Primaries Report, August 2019: I'll Just Get My Coat

[EarlG]

Welcome to the fifth monthly DU Primaries Report, based on the ongoing candidate preferences of DU members right here in the Democratic Primaries forum!

How the Candidate Rankings board works:

• Displayed percentages are rounded to the nearest whole number, so a candidate displayed at (for example) 10% will have an actual percentage in the range of 9.50% to 10.49%.


• Candidates below 0.5% are displayed at 0%. Candidates with no supporters at all are marked with a dash.


• For the purposes of the DU Primaries Report, only candidates who have an actual percentage of at least 1% are included.

Previously on the DU Primaries Report

July 2019: Look, a Squirrel!
June 2019: Warren Takes Charge
May 2019: Steady As She Goes
April 2019: Let's Get This Party Started!

Where the candidates stand this month

The big news this month is that of the 25 candidates who were merrily sprinting along the campaign trail at the start of August, five are no longer standing, and there are now only six candidates who have more than 1% actual support on DU. A moment of silence please for Mike Gravel, John Hickenlooper, Jay Inslee, Seth Moulton, and Kirsten Gillibrand.



Gone but not forgotten. Except for that first guy with the stache, who was he again? Anyway, the five candidates who dropped out were all obviously underperforming, so their disappearance from the race didn't affect too much here at DU.

Candidates above 1% actual support, including Undecided, August 2019



The biggest name of the bunch -- who had been featured as one of the top candidates in the Primaries Report every month so far -- was Jay Inslee. Inslee was holding down 1.52% support at DU when he dropped out, and you can see the small bump in Undecideds as his supporters decided where to go next.

Removing Undecideds from the graph, we can see more clearly that Elizabeth Warren continued to consolidate her first place position, thwarting my prediction from a couple months back that she must surely be about to peak. She moved from 17.74% at the beginning of August to 19.26% at the end as more and more DUers piled aboard the Elizabeth Warren bandwagon (Bethwagon? Bandwarren?)

Candidates above 1% actual support, excluding Undecided, August 2019



Most of the other top candidates remained steady. Joe Biden (13.32% to 13.62%), Bernie Sanders (7.8% to 8%), and Beto O'Rourke (1.89% to 2.01%) all ticked up slightly, while Pete Buttigieg (6.66% to 6.57%) edged downwards but otherwise remained very stable throughout the month, arresting his June/July slide.

Meanwhile Kamala Harris, after remaining buoyant throughout July, continued her downward trajectory, dropping more than half a percentage point during August from 10.03% to 9.38%. Harris is still in third place, but is currently at her lowest share of support on DU since we began tracking the numbers back in March.

A graph of these candidates' share of support over the last three months gives a broader view of how they are trending:

Candidates above 1% actual support, excluding Undecided, last three months



And for a complete view of where we are so far, here's the all-time graph, where that Inslee bump to Undecided at the end of August is really noticeable:

Candidates above 1% actual support, including Undecided, all-time



Next month's predictions

Last month I predicted that the July debates could have an impact on the DU candidate rankings (I was wrong!) and that a number of lower-tier candidates would drop out during August (I was right!). We do have another round of debates coming up in September, this time winnowed down to a field of 10, which leaves half of our candidates out in the cold. Of the 10 candidates participating in the upcoming debates, all but one match the DU top 10 (if the debate was based on DU rankings, Tulsi Gabbard would replace Julian Castro). Going on past experience I'd have to guess that the next debates won't make too much of a difference to the rankings here on DU. At this point, most DUers who have chosen a candidate seem to be sticking with their choice.

I am, however, going to make the not-bold-in-the-slightest prediction that even more lower-tier candidates -- and maybe even a top 10 candidate or two -- will drop out during September. Come at me!

The next report will take place on Monday, September 30. Thanks for reading!

[CaliforniaPeggy]

1. Thanks for posting this stuff--it's fascinating!

[Politicub]

6. Thank you for taking the time to do this!

It's really informative. It's interesting to see how the DU primary community is trending.

[FakeNoose]

2. It looks like Liz Warren got the bump as others have dropped out

Actually they're all doing well, considering it's still in the early months of the race. I do believe the front-runners will continue to distinguish themselves as time goes on. It's a long haul where momentum and fundraising are important factors. But for now we can anticipate the next debate and root for our favorites.

[PatrickforO]

3. Thank you for this. I always like these updates.

I find it very interesting that the national polling is different than what we see here on DU.

I went into Real Clear Politics and averaged their late August polls. There, we have Biden at 29%, Warren and Sanders in a tie at 17%, Harris at 7% and Buttigieg at 5%.

What I find fascinating that Warren leads on here, Sanders has been steadily dropping, and Biden a steadily further-back second place. I guess this group (DU members) is generally pretty well-informed about the issues and the candidates.

I love the discussions that can be had on here, and am supporting Warren for what I as an economist think are very sound reasons.

I cannot help but wonder as time goes by what will happen with Sanders and Warren moving forward in terms of the wider national polling groups, and Biden as well. Thinking of the bitterly fought '16 primary battle between Sanders and Clinton, I don't recall how they polled on here. I seem to remember it was awfully close - I was a big Bernie supporter at that point.

It would be fun to see how DU has 'tracked' over the last few elections in terms of who people on here predominantly support versus credible national polling averages. I wonder if we are somehow prescient, or maybe not so much. Did your numbers actually come to mirror the national polling averages, or has DU pretty consistently marched to the beat of a different drum?

[EarlG]

5. Good question

Just going from memory here, the most popular candidates on DU in past presidential election cycles were:

2016: Sanders
2008: Kucinich, then Edwards, then Obama
2004: Dean

So DU does not have the best track record in predicting the nominee Traditionally our members are further left than the electorate at large, and better informed (there are no name recognition problems when polling DUers, for example).

But who knows what will happen this time around? The world is a very different place, and we have many excellent candidates to choose from. At the end of the day I’m confident we’ll end up with a great nominee that everyone can behind, who will beat Trump.

[PatrickforO]

7. Yeah, this primary definitely has a different feel.

Much less sniping between supporters of the various candidates.

It's definitely going to be an anybody-but-Trump kind of scenario come 2020. Way too much at stake to even think otherwise.

[Aussie105]

10. Sure beats

the anybody-but-Hillary thing of the 2016 election.

We now have good reasons to think anybody-but-Trump will be an improvement. No outside propaganda or fake news needed.

[calimary]

4. There are some lower-tier folks who should have folded their tents already.

Some folks like John Delaney and Tim Ryan and Marianne Williamson are realistically and in all practicality non-starters. Bill DeBlasio too.

John Delaney rightfully should have given up after Elizabeth Warren ate his lunch and dinner at that debate. He should have closed things down the next morning. Could have stepped away with head held reasonably high. By now, though? He’s nothing but damaged goods. And he has less charisma than a used kitchen sponge.

[joost5]

8. Would it be OK to reveal to us the total member numbers that participate in this?

...or would you prefer that to remain secret? Just wondering about the statistical significance/weight.

[EndGOPPropaganda]

12. Can be inferred

Last month Harris showed a 0.03% change, so if that was one person difference, it would mean there are approximately 3000 voters participating.

[joost5]

16. Smart! +1 Thanks

[KY_EnviroGuy]

9. Thanks for posting this, EarlG. Great presentation!

KY...........

[EndGOPPropaganda]

11. Thank you. Comments below

Q:
Given that you can measure a 0.03% change (Harris last month), that must mean there are at least 3000 voters participating here?

Suggestion:
For the graphs that span only one month on the X axis, change the aspect ratio to make them thin and tall. The current plots are almost impossible to interpret because they're stretched out so much.

[RazBerryBeret]

13. this is super interesting

Warren is a formidable force. I believe there is a definite skew here tho, one candidate who is not liked on this site probably is a good percentage of the "undecided" category. just my guess.

[RO4Warren]

14. There is a definite skew to the data

Warren voters tend to be super politically engaged and there is more enthusiasm around her campaign as evidenced by crowd sizes, so we are more likely to be on this site (or any online pol forum).

This only bodes well for her as more and more voters tune in as the primaries approach. It means the people who are paying the most attention are skewing in her direction, which means we like what we see.

Granted, we are a little bit different demographically than those who are less engaged at this point in the process, but we are also a big chunk of the base.

[James48]

15. I'm withBernie.

He’s the real deal, and has nationwide support.