CNN's God awful poll (voters, interviews, illegal, Democrats)

[btownboss4]

So, CNN/Opinion research released a poll yesterday, and its so bad for the Republicans, its laughable.
RealClearPolitics - Election 2016 Presidential Polls
It has Clinton hitting 60% in a National General election. A non-incumbent presidential Candidate has never hit 60% since Harding (The only one to break 55%since WWII is Ike) ever in an election, and hitting 60% in polling, with some undecided voters to boot (some of which would pick Clinton).
And in an environment like today the floor for either party is likely 45-46%. Even in Massachusetts, or Alabama, the loser gets close to 40% of the vote.

[Bureaucat]

Originally Posted by btownboss4

So, CNN/Opinion research released a poll yesterday, and its so bad for the Republicans, its laughable.
RealClearPolitics - Election 2016 Presidential Polls
It has Clinton hitting 60% in a National General election. A non-incumbent presidential Candidate has never hit 60% since Harding (The only one to break 55%since WWII is Ike) ever in an election, and hitting 60% in polling, with some undecided voters to boot (some of which would pick Clinton).
And in an environment like today the floor for either party is likely 45-46%. Even in Massachusetts, or Alabama, the loser gets close to 40% of the vote.

It's certainly an outlier, but the only matchups in that poll that has Hilldog reaching 60% are versus Cruz and Carson. If one of that duo were nominated, it would likely be a bloodbath, not 60% but a bloodbath.

[Spartacus713]

Well, that clinches it then.

Of course as everybody knows, CNN has a flawless record in the predictions department, and since they have never mislead the American people with their sensationalism, partisan spin or half-baked "journalism," clearly the only sensible thing for the Republicans to do is to disband as a party and concede permanent one-party rule to the Democrats from this point forward.

Amiright?

[Ponderosa]

I think 60% versus Cruz or Carson is certainly within the realm of possibility. Bush, though, will be the candidate for the Republicans.

[victimofGM]

Only poll that counts is the one on Election Day. Taking a poll more than a year and a half from election is on par with push polling.

[btownboss4]

Originally Posted by Ponderosa

I think 60% versus Cruz or Carson is certainly within the realm of possibility. Bush, though, will be the candidate for the Republicans.

I don't think so, even in places like Michigan, Pennsylvainia, New Jersey, Connecticut it's hard to break 60/40 the general floor even it moderately-highly blue states is 42-45% or so Unless it is exposed Cruz beats babies with severed puppy legs.

[drishmael]

Originally Posted by btownboss4

"I don't think so, even in places like Michigan, Pennsylvainia, New Jersey, Connecticut it's hard to break 60/40 the general floor even it moderately-highly blue states is 42-45% or so Unless it is exposed Cruz beats babies with severed puppy legs."

What is Ted Cruz best known for among the general public? His role in the government shutdown. How popular was the government shutdown?

[LauraC]

Who is the person in this Election forum who told me just last week that all election polls are registered voters?

Here's who CNN polled:

"Interviews with 1,018 adult Americans conducted by telephone by ORC International on April 16 - 19, 015. The margin of sampling error for results based on the total sample is plus or minus 3 percentage points. This sample includes 615 interviews among landline respondents and 403 interviews among cell phone respondents."

http://i2.cdn.turner.com/cnn/2015/im...ll04202015.pdf

That's right, they polled any adult who picked up the phone, registered or not registered, eligible to vote or not eligible to vote. If you go further down into the poll you'll see from the questions they just had to describe themselves as Republican or Democrat.

Now look at the question for which this topic is based:

"I'm going to read a list of people who may be running in the Republican primaries for president in 2016. After I read all the names, please tell me which of those candidates you would be most likely to support for the Republican nomination for president in 2016, or if you would support someone else."

You'll notice they asked who they would support not who they would vote for. They most likely did that because they had absolutely no idea if the person they were talking to on the phone was registered to vote, would be likely to vote or was even eligible to vote. For all we know, and for all they know, half of the people they polled were illegal, ex-cons or not registered to vote so their opinion means squat.

This is also why Real Clear Politics averaging polls is garbage. You shouldn't average polls with different methodology.

[drishmael]

Originally Posted by LauraC

"Who is the person in this Election forum who told me just last week that all election polls are registered voters?"

I stand corrected -- good catch! However, take a look at this screenshot from the RCP average of Clinton vs. Bush:



No, not every polls uses exclusively RV's, but CNN/ORC is more or less the exception. Good to know -- I'll be more skeptical of its findings going forward.

"This is also why Real Clear Politics averaging polls is garbage. You shouldn't average polls with different methodology."

That's the whole point of aggregating polls: some polls are good, some polls -- even well established polls (Gallup) -- are garbage, but we don't always know which is which, so the average is generally the best indicator. (Unless we're talking about something like the Selzer poll in Iowa, in which case I'd trust it over the aggregate.)

[Arjay51]

Poll results tend to reflect the outcome that the poll maker wants them to reflect. It is about how the questions are worded and who the selected responders are.

That is why polls are useless.