Interesting Statistics

[gil]

Ok this is funny.

https://twitter.com/RealJamesWoods/stat ... 94/photo/1

Agree. I'm going to show this to my left wing friends who are worried about what they see as "deliberate" slowdowns in mail delivery.

[gil]

A contributing factor is how people who have more to lose than others tend to be incognito and assert their political influence [even if it's just 1 vote] inconspicuously and from the shadows. They don't want anything to do with risking safety in order to be counted. They don't have the time or energy because they are too busy at their jobs, trying to support the government dependent Losers who are THE BACKBONE of the Democrat Party...via taxes and special programs.
The bitter, self entitled Losers who have NOTHING to lose tend to be more vociferous.

I have randomly met a few dozen people here in fairly liberal MA and 100% have been conservative. The silent majority will be heard with jet engine decibel levels in 67 days.

Is MA going for Trump? Let's say that the "received view" is that 2/3 of MA voters will vote for Biden. If you have randomly met 36 people, the chance that they are all for Trump would be (1/3)^36 or 0.0000000000000006662%

[ddraig]

A new Zogby poll:

"A Zogby poll of 901 “likely voters” (not just people who hold a registration card) published on Wednesday reported Trump’s approval at 52 percent, with a 47 percent disapproval rating. More narrowly, the president had a 36 percent approval rating with blacks, a 37 percent approval rating with Hispanics, and a 35 percent approval rating with Asians.

Ten percent of black and Hispanic voters changed their support from former Vice President Joe Biden to Trump over a two-month period, according to a CNN poll published on August 16."

[ddraig]

Is MA going for Trump? Let's say that the "received view" is that 2/3 of MA voters will vote for Biden. If you have randomly met 36 people, the chance that they are all for Trump would be (1/3)^36 or 0.0000000000000006662%

Well, Ed Markey could be in trouble soon. He's reputed to be a Catholic.

"Boston Cardinal Sean O’Malley issued a statement reiterating the Catholic Church’s position on abortion after one of his priests came out as pro-choice on Facebook this week."

In his statement, O’Malley said that the Catholic community “has the right to expect the priests of the Archdiocese and those entrusted with handing on the faith to be clear and unequivocal on the Church’s teaching concerning respect and protection for life from the first moment of conception to natural death.”

[gil]

A new Zogby poll:

"A Zogby poll of 901 “likely voters” (not just people who hold a registration card) published on Wednesday reported Trump’s approval at 52 percent, with a 47 percent disapproval rating. More narrowly, the president had a 36 percent approval rating with blacks, a 37 percent approval rating with Hispanics, and a 35 percent approval rating with Asians.

Ten percent of black and Hispanic voters changed their support from former Vice President Joe Biden to Trump over a two-month period, according to a CNN poll published on August 16."

Interesting. I saw that on the 538 list of polls and it looks like an outlier. https://projects.fivethirtyeight.com/polls/ (poll conducted Aug 17-19). Is Zogby picking up something that all the other pollsters are missing?

[ddraig]

Well, for one thing, they look at likely voters. Most polls just look for registered voters. Big difference. many registered voters don't vote.

[gil]

Well, for one thing, they look at likely voters. Most polls just look for registered voters. Big difference. many registered voters don't vote.

The site I linked to indicates if the poll is of registered voters (RV), likely voters (LV), or adults (A). A lot of the polls they show are of likely voters, including the one Zogby poll. https://projects.fivethirtyeight.com/polls/

[ddraig]

Then too, most polls over sample Dems by at least 6% and correspondingly under sample Republicans and independents. I can only speculate as to why they do this. Make up your own mind. On top of that is the huge, 10% to 12% of Republicans and Independents that distrust the polls even when they say they won't turn over who they talk to. So if I were a betting man, which i'm not, I would say that the polls under count Republicans by at least 6% and more probably 8%.

[gil]

Most of the polling organizations get and maintain their credibility from being accurate, isn't that true? Why would they NOT want to make adjustments?

Historically, the 2018 midterm polls were pretty accurate.

Going way back to 2016 the national polls clearly were tightening right before the presidential election, and generally showed Clinton winning by 2-6%. She won the popular vote by 2.1%. The polls in Michigan, Pennsylvania, and Wisconsin were WAY off (by 5-7% as I recall). But polls of Texas (and some other states if I recall) overestimated Trumps' support.

Yes, the polls missed in 2016, but many were well within their margin of error. And certainly they didn't miss the national result by 6% or 8%. Do you think something is different in 2020?

[bpj]

Saw this, looks familiar