A Fourth Fake Polling Firm Has Been Unmasked

I told you the polls were entirely cooked this year

FiftyPlusOne has outed another fake polling firm in two days. The details of their investigation are published at their Substack. SoCal Strategies appears to be connected to Wedgewood Polls, which FiftyPlus One revealed as fake yesterday.

Altogether, four fraudulent polling firms have been credibly accused of publishing false information in this cycle. Median Strategies and The Public Sentiment Institute (TPSI) were exposed one week apart in August.

SoCal Strategies apparently took real money from real candidates to conduct real surveys, but they also produced two surveys in May that “are materially different than other releases on their page”, FiftyPlusOne report.

“For example, they do not include a link to a full set of crosstabs, as the firm typically publishes, and the format of the releases is different from their usual style.” FiftyPlusOne says their investigation into SoCal Strategies began a week ago.

The company appears to be an amateur startup, possibly created by a Floral City, Florida man named Jack Sims, and possibly run by a Boston University student named Wilson Rochford.

Whoever was running SoCal Strategies, that person or persons took down the website and X account this morning. The timing was not a coincidence. Rochford, or someone claiming to be Rochford, had agreed to give FiftyPlusOne “a statement and additional information” this week, and then the Wedgewood story dropped.

Two things in the FiftyPlusOne reporting stand out to me. First is the attestation of Mr. Sims earlier this year that “ITS TIME FOR THIS TO END AND FOR US TO MOVE ON. I JUST WANT TO PUT THIS ALL BEHIND”. Sims was shutting down the limited liability corporation at about the same time SoCal strategies apparently faked a set of poll results.

Second, someone broke a cardinal rule of tradecraft by doing just one apparently fake poll with SoCal Strategies, which had a track record of actual, real, paid survey work. If the plan was to publish both fake and real poll results, the bifurcation of those two lines of business, real and shady, should have been total.

As I said above, this looks like an amateur operation by someone who understands the new world of prediction markets. If you can convince the world that Candidate A will win the race by at least six points while betting on Kalshi that Candidate B will overperform their polls, you can make a bucket of cash on Election Day.

These fake pollsters apparently took Democratic voters for suckers. Nearly the entire polling industry exhibits a bias towards Democrats, in no small part because media consumes more polls that make Democrats look good, especially in the time of Trump.

Whoever did this understood opinion surveys. They leaned into the low response rates from red, rural voters, the normative assumption of a massive youth turnout, and so on. Online betting sites created the opportunity for arbitrage. They did not have to be rocket scientists, or even real political scientists, to profit on the margin call.