First, the “news”: the two latest national polls, from YouGov and Monmouth, both have Elizabeth Warren ahead of Bernie Sanders in Democratic primary voter intention:
Candidate |
Monmouth:
current |
Monmouth:
Change
(1 month) |
Monmouth:
perceived
electability |
YouGov:
current
|
YOUGOV:
CHANGE
(1 week) |
YouGov:
consider-
dissapoint |
Two new polls
Biden |
32 |
-1 |
7.7 |
26 |
0 |
52-19=33 |
Warren |
15 |
+5 |
6.4 |
14 |
-2 |
45-11=34 |
Sanders |
14 |
-1 |
6.5 |
13 |
+1 |
35-19=26 |
Harris |
7 |
-4 |
6.0 |
7 |
+1 |
38-10=28 |
Buttigieg |
9 |
+3 |
5.6 |
4 |
+1 |
34-6=28 |
FiveThirtyEight’s pollster ratings give YouGov a B, and Monmouth an A+.
Does this mean that Warren has now clearly taken second place from Sanders? Well… not really. Since my last polling diary, there have been 6 polls, and only these two have Warren ahead of Sanders. So yes, I’m cherry-picking here.
In exactly two months from today, if all goes well, I will defend my thesis to get a PhD in statistics. I know perfectly well that the rational thing to do is follow polling averages, where Warren hasn’t yet passed Sanders, not focus on just the polls where she has. So why am I writing this diary?
Because perceived electability matters. It matters because it’s a decisionmaking shortcut, but also because our crappy voting methods make it matter more than it should. And so focusing on the polls where Warren does well actually is useful evidence that she’s electable.
...
The first reason that electability matters is that when making complex choices, we often fall back to heuristics; cognitive shortcuts. One important heuristic for Democrats in this election is perceived “electability”: who we think can beat Trump. The Monmouth poll explicitly asked voters to rate electability on a 0-10 scale, as shown in the third column above. On that measure Sanders is perceived as more electable than Warren, and Biden beats both of them by more than a point.
And I believe that perception of relative electability is wrong. As a Warren supporter, I believe that the data show that Warren is in fact the most electable candidate in this race. Look at the last column above: the percentage of Democrats who say they’re considering each candidate, minus the percentage who say they’d be disappointed if that candidate wins. By that measure, Warren is actually the most electable, despite still having lower name recognition than either Biden or Sanders.
And frankly, I believe that both Biden and Sanders have weaknesses that don’t entirely show in the polling numbers.
For Biden, it’s obvious. I watched how the top four candidates talked with the Poor People’s Forum in Washington DC last week, and frankly, Biden was a hot mess. As just one example of his long string of gaffes and nonsequitors, he talked about “celebrating” the deaths at “Mother Emmanuel church” 5 years ago. The guy I saw in that video simply does not have what it takes to become president.
With Sanders, I’m less certain about it. He certainly did a great job at the forum, as did both Warren and Harris. But I think that his age is still a valid concern. I do not want to run an election against the oldest first-term president ever with a candidate who, on the day of his inauguration, would be older than any previous president was when they left office.
…
Second, perceived electability matters because our current voting method creates strategic voting incentives that turn it into a self-perpetuating cycle. When your vote — in either the primary or the general election — is artificially limited to choosing just one candidate, that nearly forces you to pay more attention to “electability” than you should. Thus “electability” becomes just a cycle of people chasing the polls—polls of other people doing the same thing.
There are better ways to run elections!
If we were talking about electing a legislature, I’d explain proportional representation, which can ensure that almost every vote matters, that almost everyone has a true representative in Congress. Gerrymandering, the practice of drawing lines to ensure the other party wastes their votes, would simply become impossible, because almost no votes would be wasted. In modern versions, you can even keep simple ballots, broader individual voter choice, and direct local representation.
But that doesn’t work when you’re talking about a single-winner election like a party nomination or presidential election. So, I’ll explain 3 voting methods that would fix the problem in that case: Ranked Choice Voting, Approval voting, and STAR voting. That’s in order from the best-known to the best actual solution to the problem.
First, RCV. Voters rank the candidates; first choices are tallied; the last-place candidate is eliminated; their votes are transferred if possible; and that continues until some candidate has a majority of remaining votes. This system can sometimes prematurely eliminate compromise options, but other than that, it does a good job. It would certainly reduce the strategic pressure to only vote for frontrunners, so that we could see who is really electable, not just who everyone else thinks everyone thinks is electable.
Second, approval voting. You can vote for as many candidates as you want, and most votes wins. Simplest possible reform, and better than RCV in several ways. For instance, it would be very simple for Daily Kos to have an option to run user polls (like the one below) with approval voting; unlike RCV, tallying and displaying the results would still be easy. But the downside is, out of the candidates you approve, you can’t show which one is your favorite.
Finally, STAR voting. You rate each candidate 0-5 stars; the two with the highest ratings are finalists; and the finalist with a higher rating on the majority of ballots wins. Like approval, it’s easy to tally and display results, but like RCV you can make distinctions between the candidates you like. It’s really the best of both worlds.
I know, explaining voting methods is a bit of a digression from a “Warren is getting some good polling numbers” diary. But please forgive me. I’m a Warren fan, and Warren fans love wonky plans.
…
Anyway, here’s the moral of this diary:
Warren is getting some good polling numbers. She hasn’t passed Sanders (yet) in the averages, but there is an important lesson here: she’s electable. And the more people hear that lesson, the more I think her numbers will continue to rise.
Want to see that happen? Contribute:
Let’s donate to Senator Warren’s campaign before June 30, the end of the reporting quarter.
Please donate to Senator Warren’s campaign by using the ActBlue page of the Daily Kos group EWarren2020.
Our new group has raised almost $1000 so far for her campaign. If you helped — Thank you! Our current goal is $2000
Kos Media, LLC, has not endorsed any candidate for President. The Daily Kos group EWarren2020 arranged for our ActBlue thermometer.
Authors and Commenters: Please feel free to copy this comment to any diary about Elizabeth Warren or its comment section. Please Republish all diaries about Elizabeth Warren’s campaign to EWarren2020
Beginners: Please Follow the Daily Kos group EWarren2020 (click the word Follow on that page), and watch your Stream (your Daily Kos Inbox) for her campaign diaries.