Ranked choice national poll has Biden and Sanders within the MOE of each other.

[TwilightZone]

Last edited Sat Feb 29, 2020, 01:18 PM - Edit history (1)

"The two candidates are within the margin of error of one another."

"Sen. Bernie Sanders is now leading the Democratic field in both a standard poll — and a “ranked-choice” survey designed to identify who a majority of people ultimately prefer in a crowded field.

According to an online national poll sponsored by FairVote and conducted by SurveyUSA, Sanders comes in first in both surveys. Former Vice President Joe Biden, meanwhile, comes in second and trails him narrowly in the two polls."

"The ranked-choice result is a bit closer than the standard poll — and within the margin of error — indicating that both candidates are popular second and third choices among those who back other candidates."

Details and more here:

https://www.vox.com/2020/2/28/21156380/bernie-sanders-joe-biden-ranked-choice-poll

[Sloumeau]

1. A statistical tie? Hope I'm not steppin' into the Twilight Zone. :) nt

[TwilightZone]

4. Could be a preview of the convention.

If so, all hell is going to break loose, though there are probably still a few plot twists to come.

It certainly throws the narrative that we should just coronate Sanders if he comes in with a plurality on its ear, though I always thought that was a silly assertion anyway.

[Andy823]

2. I don't trust polls

especially "online" polls.

[TwilightZone]

3. SurveyUSA is near the top of the list.

538 has them rated an A with a very-slight D bias. Most of their polls are mixed-mode, meaning some live on both home phone and mobiles, some IVR, some online. Their methodology is explained at the bottom of the poll:

https://fairvote.app.box.com/s/0kj942q70g9ozvvc74n1x5kxnsew5ege

[squirecam]

5. National polls are irrelevant

You can have one candidate get millions more votes in CA and NY, only to lose the election.

We have state by state results. Tell me who performs better in the purple states we need to win the election.

[TwilightZone]

7. While I agree with that from a high-level...

this more indicates that the insistence by some that Sanders should be simply coronated at the convention with a plurality because "that's what the people want" might be a bit overoptimistic.

But, you're right. In November, all that matters are the few states that will determine the election.

[DanTex]

6. LOL that you changed the title because it said Bernie was ahead.

Yet another poll showing that, by any conceivable measure, Bernie is the most popular candidate.

[TwilightZone]

8. He's not ahead.

Still don't get how polls work, huh?

Within the margin of error, as the article notes.

It also pokes a hole in your insistent claim that Sanders should just be coronated at the convention with a plurality because "that's clearly what the voters want". Oops.

[DanTex]

9. Title: "Bernie Sanders narrowly leads Joe Biden in a national ranked-choice poll"

You should send Vox a furious email with your theory of "how polls work" and try to get them to change their headline.

[ehrnst]

15. "conceivable" You keep using that word...

I don't think it means what you think it means....

[thesquanderer]

10. Bad headline. Within MoE does NOT mean statistical tie.

The higher performing result is still more likely than the lesser performing result. Think of the figures as the center points of two bell curves.

Also see: https://www.motherjones.com/kevin-drum/2012/10/one-last-encore-great-statistical-tie-fallacy/

[TwilightZone]

12. Pew has a different take, but I'll change it for clarity.

Pew Research has a different take, but I'll change it. In their example, because the MOE is applicable to each candidate individually, a result would have to be outside double the reported MOE to indicate a race that isn't too close to call.

"News reports about polling will often say that a candidate’s lead is “outside the margin of error” to indicate that a candidate’s lead is greater than what we would expect from sampling error, or that a race is “a statistical tie” if it’s too close to call. It is not enough for one candidate to be ahead by more than the margin of error that is reported for individual candidates (i.e., ahead by more than 3 points, in our example). To determine whether or not the race is too close to call, we need to calculate a new margin of error for the difference between the two candidates’ levels of support. The size of this margin is generally about twice that of the margin for an individual candidate. The larger margin of error is due to the fact that if the Republican share is too high by chance, it follows that the Democratic share is likely too low, and vice versa."

https://www.pewresearch.org/fact-tank/2016/09/08/understanding-the-margin-of-error-in-election-polls/

[Skidmore]

11. Statistical ties

are rw and media wet dreams.

I don't trust polls.

[Squinch]

13. Are they counting the republicans who will vote for one of them?

Or the effect of Russian interference?

[David__77]

14. Somehow I think the MOE reference is more associated with a certain ranking order.

...