Trump supporters have repeatedly questioned not just the findings, but the motives of pollsters.

Earlier this week, Thomas Lifson expressed skepticism about a Hart Research Associates poll showing Donald Trump down by 11 points.  As Conservative Treehouse revealed, the organization has been a big contributor to Hillary Clinton, handing over $220,500 in September alone.

But Hart is not taken seriously by those who follow the polling industry.  Peter Hart is always identified as a “Democratic pollster,” and Nate Silver, in his detailed analysis of the 2012 presidential race, didn’t bother looking at Hart’s data.

In fact, the major polls have a strong incentive to get accurate results.  As Frank Newport of Gallup can tell you, their reputation and their bottom line suffer when they misforecast an important election.

And pollsters are well aware that they are under more intense scrutiny that they were a decade ago.  Call it the Nate Silver Effect.

Gallup, you’ll recall, showed Mitt Romney comfortably ahead for all of October, with the race tightening in the final days.  In its last poll before the election, Gallup still showed Romney in the lead, 49% to 48%.  But the president wound up with 51.06% of the vote, and Mitt with 47.2%

Frank Newport ordered an exhaustive review of the organization’s polling methods.  The results were released six months later.

A key problem, according to the report, was with the model for determining “likely voters.”  Gallup overweighed the respondents’ past voting history, and, revealingly, “the thought respondents were giving to the election.”

Though they don’t phrase it this way, the venerable polling organization concluded that Democratic voters do not give much thought to the election and aren’t likely to have voted in the past.  Who are these people?

Dead people do not think about upcoming elections, nor do patients with Alzheimer’s and dementia.

But though Democrats still harvest votes from cemeteries and assisted living facilities, Richard Daley’s modus operandi has been passé for more than two decades.

Today, voters who haven’t voted in the past or thought much about the election are more likely to be illegal immigrants.

The floodgates were opened by the National Voter Registration Act of 1993.  Section 5 requires DMV personnel to issue voter registration forms to everyone applying for a license.  You just check the box on the form that says you’re a citizen, and you’re good to go.  Attempts by states to verify the claim bring Soros-funded lawsuits.

Early voting further compounded the difficulty of eliminating ineligible voters.  States trying simply to limit the window for early voting have been successfully targeted by leftist law firms.  (Wikileaks, Hillary’s health problems, and growing anger over Obamacare make early voting especially important for the Dems this year, Ed Lasky argues.)

After early voting came “out of precinct voting” and, worse, “instant registration,” where individuals can register and vote at the same time, eliminating the possibility of any verification of their citizenship. 

No one has been more diligent in exposing voter fraud than former Dept. of Justice attorney J. Christian Adams, now president of the Public Interest Legal Foundation.  A couple of articles worth reading on the way voting is being rigged are here and here.  Another article documents the situation in Virginia and Pennsylvania, and Michelle Malkin looks at illegal registration in Colorado.

A key takeaway from Adams: while Republicans have focused on policy, Democrats have focused on process.

Knowing which party illegals, ex-cons, and the Lofos and Nofos vote for, big leftist law firms have worked around the clock to gut the system’s safeguards.

It’s no accident, as Marxists used to say, that the National Voter Registration Act was one of the first pieces of legislation pushed by Bill Clinton, or that Barack Obama for years worked closely with ACORN and funneled hundreds of thousands to the organization.

Back to the polls.  Again, the major firms need to get the election right if they are going to trusted and hired.  So this year they’re overcounting Democrats.

Buried deep in Reuters’s article on the latest Reuters/Ipsos nation-wide poll, which showed Donald Trump slipping to an 8-point deficit after the tape furor and the debate, was the following information:

The poll of 2,386 American adults included 1,839 people who watched the debates, 1,605 people who were considered likely voters due to their registration status, voting history and stated intention to vote in the election. Among the likely voters, the poll counted 798 Democrats and 586 Republicans.

Why the sampling imbalance?

Why do Democrats represent 57.6% of likely voters affiliated with a major party, while Republicans represent just 36.5% of this group?

Remember, in the last election, the parties split the vote 51% to 47%.

So as a percentage of Democratic votes, the GOP had just over 92% in 2012.  A survey of voter affiliation by Gallup put the percentage at 80.6%, and a more recent survey by Pew claims it’s 79.59%.

But the Reuters/Ipsos poll estimates the percentage at only 73.4%.

What’s the reason for assuming that the GOP has fewer than 3 voters to every 4 Democrats, rather than 4 to every 5, or 9 to every 10?   Could it be that the poll is trying to factor in votes that are coming in under the radar?

If so, this is good news for Donald Trump.

1. The polls may be overcompensating for their failures in 2012.  Gallup was not alone in its Republican bias.  According to Nate Silver, 19 of 23 polls overestimated Romney’s vote totals.  The pollsters may be swinging too far in the other direction, despite massive voter fraud being perpetrated by the Dems.

2. The projected easy victory for Hillary is likely to dissuade unmotivated voters to turn out, or even mail in their absentee ballots.

This is something that worries the DNC, according to Reuters.  The perception that the race is not competitive will inevitably depress the party’s totals.  And the Dems can’t count on much affection for their nominee.  According to Reuters/Ipsos, just 12.6% of Hillary’s supporters are voting for her because they like her personally, and this figure sounds suspiciously high.

So it looks as if the major pollsters are putting their thumbs on the scale.  But it’s not because they’re in cahoots with the DNC, but because they’re trying to factor in Democratic vote fraud.  And they may be pressing too hard.

Jay Michaels (as Josh Michaels) is the author of the novel Outlaws.