Fake Polling Firm Behind LA Mayor's Race Hoax Shuts Down, Identified as 21-Year-Old College Grad

Since Median Strategies first posted its fake Los Angeles mayoral poll on August 12 showing Karen Bass up 12 points, the fallout has moved fast: the company admitted the fraud, deleted its site, and got unmasked as the work of one 21-year-old.

The Guardian identified him as Rahil Prakash, a recent college graduate, according to reporting cited by NPR. Prakash posted a statement saying Median Strategies was a short-term social experiment to test how fast fabricated numbers could spread through the political information ecosystem without anyone checking. He said he wasn't trying to profit off it.

NPR reported that Kalshi, the prediction market where political bets flow by the billions each week, dug into Prakash's account after the hoax became public. Two people with direct knowledge of his trading, who were not authorized to speak publicly, told NPR the company found no evidence he placed money on any of the races tied to his fake polls.

The timing looked suspicious on its face. Prediction markets have had a rough year for insider dealing. NPR noted regulators and platform operators have already caught a special forces soldier betting on the Maduro capture operation, former Congressman George Santos betting against his own State of the Union attendance, and even President Trump's teleprompter operator trading off advance knowledge of the president's remarks. Against that backdrop, a stranger flooding three state races with fake polling data two months before the midterms looked like it could be another manipulation play. The sources reviewed provided no evidence supporting that theory for this specific case.

How the hoax actually worked

Median Strategies published three fabricated surveys this month: Los Angeles's mayoral primary, the Wisconsin Democratic gubernatorial primary, and a Nevada governor's race. The Los Angeles poll claimed a survey of 560 voters, a 4.1-point margin of error, and a 12-point Bass lead over councilmember Nithya Raman, plus a fabricated claim that 85% of Republican Spencer Pratt's primary voters would back Bass in the general, according to Breitbart's account of the released memo.

Bass beat Raman by five points in the actual primary. The Wisconsin poll showed Rep. Francesca Hong up 23 points on Milwaukee County Executive David Crowley; Crowley won by under a point, according to Fox News and Breitbart.

The Bass campaign shared the fake Los Angeles poll on social media as evidence of "gaining momentum," then deleted the post once the fraud surfaced, according to the Associated Press coverage carried by PBS and WDBO. Campaign spokesman Alex Stack defended the decision by noting the poll "was reported on by multiple news outlets," and said "any bad faith attempts to influence elections should be investigated and prosecuted to the fullest extent of the law."

Raman's campaign, without directly commenting, told the Los Angeles Times that Bass should have vetted the firm before amplifying it, according to the Epoch Times. Bass's campaign had separately paid $67,250 to legitimate pollster David Binder Research, a longtime strategist for Kamala Harris and Barack Obama, per campaign finance records cited by the Epoch Times.

Neither the New York Times' polling database, RealClearPolitics, nor FiftyPlusOne included Median's numbers, according to the Associated Press. Those aggregators noted the releases lacked basic disclosure, like who ran the firm, and Median never responded when they asked.

A fair question about intent

Skeptics have reason to doubt a stranger's self-description as a harmless researcher. Fabricating polling data in three competitive statewide and mayoral races, timed right before primaries, is a strange way to run an academic experiment, and Prakash has declined interviews and won't name anyone else involved. Media outlets, campaigns and betting markets all got fooled at once, which is exactly the outcome a bad actor with a financial or political motive would want.

But the direct evidence available so far cuts against a profit motive. Kalshi's review, per NPR, found no bets tied to the races Prakash lied about. And unlike the Wisconsin race, where State Navigate's independently conducted poll also showed Hong with a similar lead, Prakash never got picked up by any campaign or PAC as a paid or coordinated effort.

Where it stands now

The American Association for Public Opinion Research called the fake polls "a serious breach of ethics" that "undermines legitimate polling and public opinion work," according to CalMatters. Mark Baldassare of the Public Policy Institute of California, who has surveyed voters for decades, called the episode "a wake-up call for all of us."

No law enforcement investigation has been announced against Prakash, and no charges have been filed as of this reporting. Amanda Fischer of Better Markets told NPR she expects more attempts to game prediction markets before the midterms, given how easily this one slipped past editors, aggregators and a sitting mayor's own campaign. Whether platforms like Kalshi and Polymarket tighten verification before November, or whether the next fake pollster gets caught before a campaign amplifies bad numbers again, remains an open question.