On Monday, April 13, David Trone posted to X that he’s leading April McClain Delaney by 21 points.
“Big news: A new, independent poll just dropped and we’re winning by 21 points!” Trone wrote, slapping the word “independent” onto a poll conducted by an organization that did not exist three months ago that surveyed fewer people than attended a Frederick County zoning hearing, and was funded for less than the cost of a decent set of tires.
Within hours, the Trone campaign blasted the same numbers to its email list.
The email, sent from info@davidtrone.com and paid for by David Trone for Maryland, told supporters that “independent polling released yesterday shows us with a 21-point lead over our opponent.”
No methodology.
No sample size.
No margin of error.
Just the number: 21 points.
Problem: David Trone’s poll is garbage – and the person who funded it uses David Trone’s face and campaign slogan as their X profile banner.
Who Has Ever Heard of The Public Sentiment Institute?
The poll was conducted by The Public Sentiment Institute, an organization that operates under the handle @TPSIOfficial on X and publishes through a Substack account that launched in February 2026.
TPSI describes itself as “a start-up organization trying to gain polling reputation and experience.” Those are TPSI’s own words, written in a direct email to A Miner Detail on the evening of April 13.
TPSI’s Substack mission statement describes the organization’s goal as building a “national voter and public sentiment panel” using Pollfish. This mobile app-based survey tool delivers questionnaires via smartphone apps in exchange for non-monetary rewards such as extra lives in games or access to premium content.
Pollfish is a legitimate platform for consumer research and market surveys, but it’s not one that any credible U.S.-based political polling firm relies on as its primary means of measuring voter intent in a competitive congressional primary.
TPSI has no track record.
It has no FiveThirtyEight pollster rating.
It’s not a member of the American Association for Public Opinion Research.
It has no published history of accurate predictions, and it didn’t exist when voters in Maryland’s 6th Congressional District went to the polls in November 2024.
154 People Is Not A Serious Number
The TPSI poll surveyed 309 respondents in total.
But the “Trone +21” number that David Trone blasted to his email list and posted on X comes from a subsample of just 154 likely Democratic voters.
One hundred and fifty-four.
To understand how absurd that number is, consider the standard. Hart Research Associates, the firm that has conducted the NBC News national poll since 1989 and holds a 2.6-star accuracy rating from FiveThirtyEight across fifty-three evaluated polls, conducted the most recent survey of Maryland’s 6th Congressional Democratic primary for the McClain Delaney campaign.
That April 13 poll shows incumbent Congresswoman April McClain Delaney leading David Trone 49 percent to 37 percent.
Hart Research’s December 2025 survey of the same race polled 600 likely Democratic primary voters via landline, cell phone, and text-to-web, with a margin of error of plus or minus 4.1 percentage points. That poll showed McClain Delaney leading 50 to 33.
TPSI acknowledged a margin of error of ±7.9 percent for its 154-person Democratic primary subsample.
That means, by TPSI’s own math, the poll’s worst-case scenario still shows Trone leading by 5.2 points. But a 7.9 percent margin of error for a 154-person subsample drawn via a mobile app survey tool is not a measure of voter sentiment. It’s nothing more than a measurement of statistical noise.
For context, a properly constructed poll of a Democratic primary electorate in a single congressional district typically requires 400 to 600 respondents in the target universe to achieve a margin of error of four to five percentage points. TPSI’s Democratic subsample is roughly one-quarter to one-third the size of the floor that professional firms consider.
No legitimate polling firm in the United States would release a congressional district primary poll with 154 respondents.
No credible news organization would report it as meaningful data.
No serious candidate would blast it to a fundraising list and call it “independent polling” unless the purpose was not information but narrative construction.
TPSI’s Maryland County Problem
The crosstabs published by TPSI tell an even more damaging story than the sample size.
TPSI released the county-level breakdown of its Democratic primary subsample. Of the 154 Democratic likely voters who produced the “Trone +21” headline, the weighted data show approximately 52 from Allegany County, 72 from Frederick County, 8 from Garrett County, 22 from Montgomery County, and 0 from Washington County.
That distribution bears no relationship to the actual Democratic primary electorate in Maryland’s 6th District.
Montgomery County is the most populous and most heavily Democratic portion of the 6th District. In any real Democratic primary, it will cast the largest share of ballots. In the TPSI poll, it accounts for roughly fourteen percent of the Democratic sample. Allegany County, one of the smallest and most conservative counties in the District, accounts for roughly thirty-four percent.
In what universe does Allegany County outnumber Montgomery County by more than two to one in a Democratic primary electorate? And in what universe does Washington County, home to Hagerstown and a meaningful chunk of the District’s registered Democrats, contribute zero voters to a congressional primary sample?
The answer: in a universe where the pollster draws the sample through random mobile app surveys without calibrating geographic controls to the actual voter file, and then processes the data through what TPSI’s methodology section describes as a ‘score-favoring sigmoid model.’
Established firms use turnout models based on validated voter files, prior vote history, and propensity scores calibrated against known election results. A “score-favoring sigmoid model” applied to a Pollfish mobile app sample is opaque jargon dressing up a fundamentally unvalidated approach.
And in that distorted sample, the crosstabs show one candidate receiving 94 percent of the vote in Allegany County.
Ninety-four percent – in a competitive Democratic primary. That’s not a poll finding. What is that? Well, it’s a mathematical artifact of surveying a handful of people in a county where the Democratic primary electorate is tiny and then treating the result as meaningful.
When you build a sample that overweights areas favorable to one candidate and underweights areas favorable to the other, and then report the topline number as though it represents the District, you have not conducted a poll. You’ve manufactured a talking point.
Follow the Money (All $500 of It)
In its email exchange with A Miner Detail, TPSI revealed that ‘Federal Elections Analysis’ funded the poll, which TPSI said ‘paid the 500 dollars to do the poll.’
Federal Elections Analysis operates a GoFundMe campaign, organized out of Poolesville, Maryland, titled “Help Support Our Poll in Maryland’s 6th District!” The fundraiser launched on April 4, 2026, and, as of this writing, has raised $195 toward a $500 goal from 10 donors.
After Trone posted the poll results to his social media, TPSI issued a clarification on X: “Would like to make it clear that we were NOT SPONSORED by the David Trone campaign to poll this race. We received crowdfunding and donations from @FedElects for a public crowdfunding campaign of this race, and some of his earned income in order to poll this race at HIS REQUEST.”
Read that sentence again.
TPSI says the poll was conducted “at HIS REQUEST,” referring to the person behind Federal Elections Analysis.
Let me get this right: A partisan election hobbyist based in Montgomery County requested a poll of the 6th Congressional District Democratic primary, funded it for $500 through a mix of GoFundMe donations and personal money, and the result was a 154-person mobile app survey showing David Trone ahead by 21 points.
And within hours of its release, David Trone’s congressional campaign was calling it “independent.”
The Man Behind the Money
The person who runs Federal Elections Analysis operates the X account @FedElectsNate.
His profile lists his location as Poolesville, Maryland. He joined X in June 2024. He describes himself as a “Mainstream Progressive” interested in “Polling, Data, Elections.” His bio links directly to the GoFundMe that funded the TPSI poll. This individual’s X banner image features David Trone’s likeness alongside the words “FIGHTING FOR US.”
That phrase is not a generic political slogan. That’s David Trone’s campaign slogan, displayed on the official Trone for Congress website.
The person who paid for the poll uses David Trone’s campaign slogan and likeness as his social media banner; he requested the poll himself, according to TPSI’s own statement, and the Trone campaign immediately amplified the results to its fundraising list without disclosing any of this context.
Voters can evaluate those facts for themselves.
The Israel PAC Question
Buried in the TPSI poll is a question that has nothing to do with measuring the state of a congressional primary and everything to do with testing an attack line: whether a candidate accepting donations from a political action committee supporting Israel would make respondents more or less likely to vote for that candidate.
Thirty-one percent said it would make them less likely. Twenty-seven percent said more likely. Forty-two percent said it made no difference.
That’s not a neutral public interest research question.
Guess what that is? Opposition research.
Someone wanted to know whether an Israel PAC attack would land in the 6th District, and the TPSI poll tested it. The question is who wanted to know and why.
What “Independent” Actually Means in Political Polling
In polling, “independent” has a specific meaning.
An independent poll is one conducted without the direction, involvement, or financial sponsorship of a candidate, campaign, party, or affiliated entity. It is conducted by a firm with a verifiable track record, using an established methodology and a sample size sufficient to produce statistically meaningful results.
The TPSI poll meets none of those criteria.
An apparent Trone supporter who uses Mr. Trone’s 6th District campaign slogan as their profile banner requested the poll.
The poll was funded for $500 and conducted by a “firm” with no track record, using a mobile app survey platform, on a subsample so small that the organization itself acknowledges a margin of error that swallows the finding. Its county-level weighting bears no resemblance to the actual electorate.
And the Trone campaign immediately weaponized it as proof of an insurmountable lead.
The word for this is not “independent.”
I think we’re looking for a different word?
Convenient, maybe?
The David Trone Pattern
In November 2025, Inside Elections obtained and published a Trone internal poll testing a primary matchup against McClain Delaney, a poll clearly designed to gauge whether a comeback bid was viable. That poll tested negative messaging against McClain Delaney without testing equivalent attacks on Trone. It was the first public sign that Trone was preparing to run.
In February 2016, NBC Washington reported that Trone campaign workers posed as volunteers inside the campaigns of rivals Jamie Raskin and Kathleen Matthews during the 2016 8th Congressional District Democratic primary. The Washington Post confirmed the details. Trone himself acknowledged the conduct and said he fired the workers involved.
During his 2024 Senate bid, the Baltimore Sun documented that Trone funded the campaigns of dozens of state and local officials who subsequently endorsed him.
A Miner Detail’s independent review of MDCRIS campaign finance records found that same pattern extending directly into the 2026 6th District Democratic primary: elected officials who received Trone family contributions either endorsed him or went silent, while those who endorsed McClain Delaney took no Trone money.
A Miner Detail has also that Trone’s television ad claiming he “funded” a women’s health clinic rests on a single $10,000 foundation contribution toward a $1.17 million project, and that Trone’s characterization of his involvement evolved from “honored to attend the groundbreaking” in 2023, to claiming he “opened” the clinic during his 2024 Senate campaign, to asserting he “funded” it in his 2026 advertising.
And in March 2026, a Miner Detail reported that People First, a New York-based political influencer firm, was recruiting paid Instagram influencers in Maryland under the banner of “Maryland Voices for Community Care and Reform,” an entity with no public records footprint and an issue set that mirrors the Trone campaign platform. People First’s CEO declined to identify the client. The Trone campaign did not respond to a request for comment.
The TPSI poll fits the pattern precisely: when David Trone needs a narrative, the narrative appears. When Trone needs endorsements, the checks come first, and when Trone needs grassroots energy, paid influencers materialize.
And when Trone needs a poll showing him up 21 points, a $500 mobile app survey appears on cue, funded by a supporter who plasters Mr. Trone’s face on his social media profile.
McClain Delaney Actually Leads Trone by Double-Digits
For voters who want actual data, the Hart Research Associates poll released the same day by the McClain Delaney campaign tells a very different story: McClain Delaney leads 49 to 37 among likely Democratic primary voters – a 12-point advantage.
The congresswoman leads across every geographic and demographic group in the district. Among the 80% of voters who view both candidates favorably, Mrs. McClain Delaney leads by 13 points.
Hart Research Associates has conducted polling for more than fifty years. The firm has polled for the NBC News national survey since 1989. Hart Research is not a blog that launched two months ago; it’s one of the most established and respected polling operations in the country.
The Hart Research poll is a campaign poll. The McClain Delaney campaign commissioned and paid for it. Campaign polls carry well-documented tendencies to oversample favorable respondents or test favorable messaging. Readers should apply appropriate skepticism.
But a campaign poll conducted by Hart Research Associates falls into a fundamentally different category than a $500 Pollfish survey of 154 people, weighted by an opaque “sigmoid model” that gives Allegany County more than double the representation of Montgomery County.
We can evaluate, challenge, and analyze one within an established methodological framework. We cannot do the same for the other because there is no framework to evaluate.
TPSI’s Response to A Miner Detail
TPSI contacted A Miner Detail directly on the evening of April 13 after this publication posted skepticism about the poll on X. TPSI emailed the A Miner Detail tip line at Tips@AMinerDetail.com from a Gmail address (tpsinstitutecontact@gmail.com).
The full email exchange is published here because TPSI did not request any portion of the conversation be off the record, on background, or on deep background. TPSI contacted a blogger at a news publication’s tip line without establishing any ground rules. Every word is publishable.
In that exchange, TPSI acknowledged the small sample size, attributed it to “budget constraints,” and revealed that Federal Elections Analysis paid $500 to fund the poll. When this publication posted the exchange on X, TPSI responded: “Why would you post our emails, extremely disingenuous behavior.”
A note for TPSI and anyone else who emails a journalist unsolicited: if you want a conversation to be off the record, you say so before you start talking, and the journalist has to agree. You do not email a news outlet’s tip line, make on-the-record statements, and then object when those statements are published. That is how journalism works. I didn’t make the rules.
Sleazy Is as Sleazy Does
David Trone spent more than $60 million of his own money on a Senate race and lost to Angela Alsobrooks by 11 points. He endorsed April McClain Delaney as his successor for the very seat he now wants back.
Mr. Trone announced his challenge in December 2025, saying McClain Delaney was “not doing the job” and “not fighting Trump.” And he is now running the same playbook he has run in every race: buy the narrative, build the coalition with checks, and manufacture the appearance of momentum where the organic kind does not exist.
The TPSI poll is not evidence that David Trone is winning in the 6th Congressional District.
What this is – it’s evidence that Mr. Trone’s operation understands something important about how modern political information flows: most people who see “BREAKING: DAVID TRONE AHEAD +21” on social media will never read the methodology, never check the sample size, never discover that Allegany County outnumbers Montgomery County two to one in the Democratic primary sample, never learn that the person who funded it uses Trone’s campaign slogan as his social media banner, and never find out that the entire exercise cost $500.
That’s the gamble – and that is exactly the bet a man with David Trone’s resources and David Trone’s record would make.
Maryland’s primary takes place on June 23, 2026.
Hi, I’m Ryan. I’m the editor and founder of A Miner Detail. Please reach out to me at Ryan@AMinerDetail.com.
