Vin Suprynowicz

First, before our major post of the day, a note on presidential polling:

Most major polls in the Trump-Clinton race “weight” their results. They generally admit this, though the TV talking heads rarely take the time or trouble to disclose how each poll is “weighted” as they announce the results, giving the impression that the numbers they’re parroting have a scientific precision -– as though they were reporting that the Dallas Cowboys are trailing by six points, second down and goal, on the six yard line with 58 seconds to play and no time-outs remaining. Very precise.

(Football and baseball are popular in America in part because they lend themselves so well to precise statistical analysis -– batting averages, percent of forward passes completed. This carries over to the sad notion that TV viewers can somehow tell whether our armed forces are winning some foreign war based on “body counts” (a lot more Americans than Germans died on D-Day; that doesn’t mean we lost), as well as the notion that political candidates are best judged by who’s “leading in the polls,” or who has “committed a gaffe” . . . rather than whether the candidate believes in, say, heavy government regulation with punitive taxation, versus the prosperity generated by a relatively free market. Much too complicated.)

But different polls are weighted differently, reflecting different presumptions about what percentage of Democrats, Republicans, and independents are likely to actually vote. This makes it even more misleading to simply “take an average” of all the available major polls. Since we know, by an actual count of their minutes of positive versus hostile coverage (another statistic!) and by records of their political donations, that those in the Mainstream Media today tend to favor Clinton over Trump by a ratio of 91-to-9 (yes, really!), it’s quite believable that some polls are “over-weighted” precisely to skew those averages.

Most major polls today over-weight Democrats in their samples by 4 to 12 (yes, 12 — way to go, ABC!) percent, supposedly based on the fact that Democrats have been more likely to actually go vote in recent elections, particularly in the year 2008.

But we all remember the excitement among liberals and particularly among African-Americans at having their first chance to vote for a black presidential candidate in 2008 (sufficient to get them to ignore the fact Barack Obama is the child of two communists, that his mentor Frank Marshall Davis was a Communist, that he was registered for school in Indonesia as a Muslim, that his black heritage comes from an African national rather than from anyone who was ever a slave in America, that he was a do-nothing senator who won office because “someone” illegally leaked his opponent’s confidential divorce records, that he hadn’t even served out a single term, that his political experience was limited to teaching ACORN activists how to do things like dump garbage in the City Hall chambers to intimidate elected officials -– stuff out of the Saul Alinsky socialist playbook.)

This year, on the other hand, it’s Republicans and particularly newly registered, white, blue-collar Americans who are incredibly enthused to get to the polls next week to vote for Donald Trump and his vision for a restored and prosperous America. Yet the pollsters still “weight” their samples on the assumption that vast masses of Democrats will brave storm and flood to get out and vote for . . . Hillary Clinton?

Based on turnout in this spring’s primaries alone, they should be weighting their polls based on an expectation of a 2016 Republican “turnout” advantage of at least 1 to 2 percent. That’s why I say a current “weighted” poll that shows Hillary Clinton leading in some battleground state “by two points” really predicts a Trump victory by at least 2 points -– probably more.

Are they doing this to “help” Mrs. Clinton and discourage Trump voters, or are they merely “wrong”?

That calls for a judgment of intent. Nonetheless, I believe it’s acceptable — based on their positively poisonous and unbroken character assassination of Mr. Trump, while their investigative reporters all seem to have been issued dark glasses and white canes this year when it comes to investigating the outright solicitation of bribes through the Clinton Foundation as well as Mrs. Clinton’s obstruction of justice as revealed in the Wikileaks emails — to assume the former.

(What a shame. If they were still curious reporters, instead of axe-wielding leftist berserkers, it might have occurred to some real journalist out there to write a prize-worthy story asking “How has Trump done it, when in many states he spent only 20 percent of the usual moneys hiring 20 percent as many ‘people on the ground’? Has he just demonstrated the difference between the typical government bureaucrat’s approach — ‘How many friends do we have whose nephews need jobs?’ and the approach of a free-market businessman, asking ‘What do we need done and what’s the least we can spend to get it done?’ Might it be possible to extend that approach to GOVERNMENT?”)

It may also be worth noting that, when modern polling was in its heyday 40 or 50 years ago, pollsters reported about 60 percent success in getting people to answer their phones and respond to questions at length. Americans actually felt privileged to be asked their opinions.

Today, with all kinds of systems and devices in place to screen out “robot sales calls” (some disguised as “opinion surveys”) and a general suspicion of talking to strangers on the phone in an era of “identity theft” — plus an outright hatred of the liberal media and political class by the kind of people who tend to support Donald Trump — these outfits quietly acknowledge only 8 percent of the people they try to call are actually willing to complete an interview. Are the eight percent of Americans who still have a “land-line” phone and are willing to talk to strangers really a “representative sample”?

And those are the most RIGOROUS outfits. The reliability of polls where folks are allowed to “self-select” and type in answers on their home computers is, pardon me, laughable.

On poll “weighting,” see: http://www.trump-conservative.com/news/monmouth-poll-showing-trump-2-in-ohio-amended-to-give-clinton-4-lead/ , . . . and . . . https://www.youtube.com/watch?v=dpaiuXNEWUE .

THE 80-20 SPLIT

Meantime, and even more important, the ignorance of most of our network TV talking heads (one lovely simpleton, responding to Donald Trump’s proposal to impose congressional term limits, contended on-air this week that congressmen are already term-limited) about the arcana of polling even extends to the point where they will report that a poll showing Hillary Clinton with 46 percent support, Donald Trump with 44 percent report, and 10 percent undecided, as showing “Hillary Clinton leading by 2 points.”

But that’s not what those numbers show, at all. That would be based on the “common-sense” assumption that the “undecideds” will break 50-50. In fact, since Mrs. Clinton forthrightly states she’s running to give us a “third Obama term” -– not to bring change but to supposedly “save the progress we’ve made” -– Hillary Clinton has voluntarily assumed the role of the “incumbent,” while Donald Trump is clearly the “change” candidate, the “challenger.” And undecideds tend to break 80-20 for the challenger, which means a 46-44-10 “advantage” for Hillary in a pre-election poll is actually likely to translate to a 52-48 victory for TRUMP on election day. (Add 8 out of 10 “undecideds” to Trump’s 44, but only 2 out of 10 “undecideds” to Mrs. Clinton’s 46.)

Note I’m not saying the 46-44-10 finding is in any way WRONG. I’m saying it INDICATES (to anyone who knows the 80-20 rule) a 52-48 election-day victory for a challenger now “scoring” 44 percent.

For the original documentation of this standard 80-20 break, see: http://www.pollingreport.com/incumbent.htm .

This is why the (in-the-tank-for-Hillary) major media tell us this race is “tightening,” when in fact Donald Trump has been surging ahead in the popular vote since about Oct. 25, and (barring massive fraud involving the electronic computers that capture and tabulate the votes, most likely in a few crucial states including Pennsylvania and Virginia) as of Oct. 27 was likely to win by 9 points or more.

Why do I say “as of Oct. 27”? Because it appears likely the Hillary Clinton campaign is about to implode, because of something the public started to find out about only on Oct. 28. I believe the desperate efforts of the corrupt, lame-duck Barack Obama and Loretta Lynch (and of course the all-in-the-tank-for-Hillary Legacy Media) to suppress the real revelations of “Weinergate” are about to fail.