Monday, October 24, 2016

Polls and three kinds of lies: Today's news for October 24th

Donald Trump offered his own Contract with America over the weekend, but that deserves its own post. Feel free to read it here though.

In other news from this "worst election in the history of mankind"...

Fox News:
The political organization of Virginia Gov. Terry McAuliffe, an influential Democrat with longstanding ties to Bill and Hillary Clinton, gave nearly $500,000 to the election campaign of the wife of an official at the Federal Bureau of Investigation who later helped oversee the investigation into Mrs. Clinton’s email use.

Campaign finance records show Mr. McAuliffe’s political-action committee donated $467,500 to the 2015 state Senate campaign of Dr. Jill McCabe, who is married to Andrew McCabe, now the deputy director of the FBI.

The Virginia Democratic Party, over which Mr. McAuliffe exerts considerable control, donated an additional $207,788 worth of support to Dr. McCabe’s campaign in the form of mailers, according to the records. That adds up to slightly more than $675,000 to her candidacy from entities either directly under Mr. McAuliffe’s control or strongly influenced by him. The figure represents more than a third of all the campaign funds Dr. McCabe raised in the effort.
Before I get into this, how come political action committees can donate 6 figure sums without a problem, but if you or I do it as individuals, it's against the law?

Back to the story, sure this stinks to high heaven. But if you were Terry McAuliffe, wouldn't you want to provide as much financial support as possible to your state party's senate candidate?

Add this to the story:
The FBI said in a statement that during his wife’s campaign Mr. McCabe “played no role, attended no events, and did not participate in fundraising or support of any kind. Months after the completion of her campaign, then-Associate Deputy Director McCabe was promoted to Deputy, where, in that position, he assumed for the first time, an oversight role in the investigation into Secretary Clinton’s emails.”
My conspiracy radar isn't going off on this story. I have more problems with a legal 6 figure campaign donation from a PAC than I do with any connection to Hillary Clinton.

Back to the election, already in progress...

Investor's Business Daily:
With 16 days to go until November 8, Donald Trump leads Hillary Clinton by 2 percentage points — 43% to 41% — in a four-way race, according to the latest IBD/TIPP presidential tracking poll.

Libertarian Party candidate Gary Johnson saw his support held steady at 7%, while the Green Party's Jill Stein dipped to 3%.

Unrounded, Trump leads 42.6% to 40.8% — a 1.8-point edge — with Johnson at 7.2% and Stein at 3.3%.

Trump also holds a nearly one-point lead in a two-way matchup — 43.2% to 42.3%.
Looks like a tight race, right?

Maybe, maybe not...

ABC News:
Hillary Clinton has vaulted to a double-digit advantage in the inaugural ABC News 2016 election tracking poll, boosted by broad disapproval of Donald Trump on two controversial issues: His treatment of women and his reluctance to endorse the election’s legitimacy. 
...All told, Clinton leads Trump by 12 percentage points among likely voters, 50 to 38 percent, in the national survey, her highest support and his lowest to date in ABC News and ABC News/Washington Post polls. Gary Johnson has 5 percent support, Jill Stein 2 percent.
Huh? One of these two polls is seriously off.

Fortunately, we have Wikileaks to point us to the answer:

Zero Hedge:
Now, for all of you out there who still aren't convinced that the polls are "adjusted", we present to you the following Podesta email, leaked earlier today, that conveniently spells out, in detail, exactly how to "manufacture" the desired data. The email starts out with a request for recommendations on "oversamples for polling" in order to "maximize what we get out of our media polling."  
Some of the oversampling tips in the email include for Arizona:
-  Over-sample Hispanics
-  Use Spanish language interviewing. (Monolingual Spanish-speaking voters are among the lowest turnout Democratic targets)
-  Over-sample the Native American population
For Florida:
-  Consistently monitor the sample to ensure it is not too old, and that it has enough African American and Hispanic voters to reflect the state.
-  On Independents: Tampa and Orlando are better persuasion targets than north or south Florida (check your polls before concluding this). If there are budget questions or oversamples, make sure that Tampa and Orlando are included first.
For national polling:
-  General election benchmark, 800 sample, with potential over samples in key districts/regions
-  Benchmark polling in targeted races, with ethnic over samples as needed
-  Targeting tracking polls in key races, with ethnic over samples as needed
Back to the 2 polls, how do they get such radical differences? One part is the breakdown between Democrats-Republicans-Independents. For the ABC News poll, the breakdown was 36-27-31 percent, as opposed to the IBD poll, which was 36-28-33. It may not seem like much, but it certainly contributes to a 14 point swing.

Then again, there is no telling what ABC News is doing with the poll "behind the curtains". As Tyler Durden describes it:
As a quick example, the ABC / WaPo poll found that Hillary enjoys a 79-point advantage over Trump with black voters.  Therefore, even a small "oversample" of black voters of 5% could swing the overall poll by 3 full points.  Moreover, the pollsters don't provide data on the demographic mix of their polls which makes it impossible to "fact check" the bias...convenient.
So why do Democrats do this? Because you political binaries out there have to vote for the winner. First, you limit your choices to two parties, thereby giving you a 50-50 shot at nailing the winner. Then you look at the polls, see who will win there, and then find a way to logically support that candidate.

Mind you, not everyone does this. But there are enough "joiners" out there to make this a worthwhile practice for Democrats.

As Mark Twain once said, "There are three kinds of lies: lies, damned lies, and statistics."

No comments:

Post a Comment