We are swimming in a sea of lies now — more than ever before.
Readers, a quick thought: Have you noticed that primary polls seem to be garbage lately? Remember Abdul El-Sayed’s double digit lead over Haley Stevens in Michigan? Remember how Francesca Hong was going to leg drop Milwaukee County Executive Jiffrenson “Jiff” Ramsey from the top rope in Wisconsin? You’re not alone in thinking that a lot of these polls might be fishy, or at the very least incompetent. And nor are you entirely correct — there is still such a thing as decent quality general election polling.
But for now, especially during primary season, you are right to be skeptical of primary polls, because as it turns out, several of them are . . . well, entirely fraudulent. (This is a very different assertion from casting doubt on big-name firms with track records of actually doing the work.) Just a week ago, it was revealed that a fly-by-night company called “Median Strategies” had released multiple fake polls — on the governor’s race in Wisconsin and the Los Angeles mayoral race in California, among others. When their spectacularly wrong Wisconsin poll (showing Hong winning by 23 points) was noticed after the fact, they revealed that they were not a polling firm at all — merely a bunch of anons who were “conducting a social experiment.”
My bet is actually that whoever made these fake polls — nothing but invented numbers, sans methodology — was attempting to see if he could move the political prediction markets, a topic about which more later. Luckily nobody bit on the bad data — except, fittingly enough, for Karen Bass’s campaign, which is just that stupid.
But even when these lesser-known pollsters — with their generically coded, anonymous names like “Florida Values” or “Progress America, Etc.” — aren’t just making up the numbers wholesale in a vain attempt to slant expectations, they’re usually just chumps: bad at polling, bad with statistics, cheaply run, and little better than a front for all the same impulses that inspire the pure frauds. Witness the tale of The Public Sentiment Institute (TPSI), a startup pollster who rather clearly entered the race for ideological reasons: as friends of antisemitic bugman candidate James Fishback.
And when their online panels didn’t see enough Fishback voters in their results — this despite their “profiles,” which TPSI thought made them certain Fishback votes — TPSI simply decided that their respondents had answered the wrong way, and altered their data, switching votes from Collins (who ended drawing a surprisingly strong 25 percent of the vote) to Fishback’s column.
To say that this is spectacularly corrupt (in fact, exactly the sort of corrupt polling practice that Trump falsely accused Ann Selzer of) is to waste breath on the obvious: Nobody informed will ever listen to their polls again. But phony pollsters aren’t going after the smart money in the room — they’re going after the stupid money, and there is an enormous amount of it out there, willing to believe whatever happy story someone wants to sell them. (See again: the prediction markets.)
This is not a new story. Those of us old enough to have white in our beards can remember the Daily Kos/Research 2000 fiasco, from all the way back in the early Obama era. But it’s easier to maliciously spread false data than ever before, and the bad actors are now usually acting with specific financial or partisan motives. We are swimming in a sea of lies now — more than ever before.