When CNN released a new poll this morning showing President Trump behind by 14 points to Joe Biden, the question became not whether Trump would take it badly, but just how badly.
The answer is, pretty badly. After a brief, angry morning tweet, Trump this afternoon announced he has retained Republican pollster John McLaughlin to write a memo, which Trump shared, denouncing CNN’s poll:
It is true that CNN’s result of a 14-point Biden lead is far higher than the leads produced by other polls. A slew of polling finds Biden’s lead ranging from a low of 7 to CNN’s high of 14. Trump’s staffers backhandedly acknowledge this fact by the rote denials they have been giving reporters whenever asked about Trump’s polling: “Our internal data consistently shows the president running strong against a defined Joe Biden in all of our key states.” This line might sound confident, until you realize “defined” means “after pollsters have given respondents one-sided attack messages against him” and “running strong” means “still losing.”
In order to fulfill Trump’s demand to discredit CNN’s poll, he needed to bring in an outsider capable of generating even more outrageous claims. Hence the appearance of “highly respected pollster” John McLaughlin, who is, in fact, a laughingstock within the profession.
McLaughlin is notorious for producing rosy polling data on behalf of his clients. (Or, at least, purportedly on behalf of his clients, who often find themselves unpleasantly surprised on election night.) A lengthy stream of overly optimistic polls culminated in McLaughlin convincing his client, Eric Cantor, that he was leading primary challenger David Brat by 34 points in 2014. Cantor lost by 11, at which point leading Republicans began begging their party not to hire him. “Nearly a dozen Republican strategists who’ve worked with McLaughlin over the years say they try to steer their clients elsewhere and increasingly don’t trust his polling,” reported The Hill.
McLaughlin somehow managed to retain a piece of the GOP polling business. In 2018, one of his big House clients, John Faso, only performed 5 points worse than McLaughlin predicted. The other, Barbara Comstock, lost by 12 after McLaughlin churned out polling for weeks showing her ahead.
A record that was for most Republicans a disqualification was precisely the thing that endeared McLaughlin to Trump, who, a Republican source told Cameron Joseph last year, “likes to surround himself with people who like to tell him he wants to hear.”
Trump of course reportedly threatened to sue campaign manager Brad Parscale after being shown bad polling in April.
Earlier this year, when polls showed strong support for impeachment, Trump got McLaughlin to yield the opposite report by loading up his questions. (Actual McLaughlin poll question: “Do you agree or disagree with the following statement? ‘Impeaching President Trump is a waste of time and tax dollars and it will ultimately go nowhere, so the Democrats should focus on working with Republicans to solve our nation’s problems rather than focusing on trying to impeach President Trump.’”)
Mclaughlin’s memo dutifully mixes methodological criticisms of CNN’s sample size with ludicrous Trumpian claims that CNN deliberately skewed its findings “to manufacture an anti-Trump outcome.” Obviously, it would be totally self-defeating for a reputable pollster to deliberately produce inaccurate outcomes. Pollsters’ livelihoods depend on the most accurate possible predictions, and their polls will ultimately be measured against actual votes. Pollsters that produce bad findings are sacrificing the credibility that is their source of income.
If you look at McLaughlin’s career, though, this paranoid claim actually makes some sense.