Each poll was analyzed separately, and then both polls were analyzed together, producing three separate reports.
Below are those reports.
Analyze the linked topline poll and the attached crosstab spreadsheet for biases and hidden trends. Republican Primary Poll for Kentucky House District 4 | BIG DATA POLL
Here is a thorough analysis of both the published topline results and the attached crosstab file, organized by methodological concerns, identified biases, and the hidden structural trends the data surface.
Methodology: What's Solid
The poll was conducted April 3–4 and 6–7, 2026 by BIG DATA POLL's Public Polling Project, among 433 registered Republican likely primary voters (n=600 districtwide) . The margin of error is ±4.0% at 95% confidence for the full sample, but subgroup margins are meaningfully higher — a critical caveat the poll acknowledges . Weighting was applied for sex, age, race/ethnicity, education, and geography, per the crosstab header. Interviews used a mixed-mode design: online panel (60 respondents via Lucid/CINT), live-agent phone (P2P SMS and text-to-online) sourced from the L2 voter file . Easter Sunday was excluded, which was methodologically appropriate but should be noted as a minor contact-rate constraint.
Core validity concerns:
- The online panel (Lucid/CINT) introduces a self-selection component. Panel respondents systematically skew toward more engaged, digitally literate, and issue-aware respondents — a known source of enthusiasm inflation.
- P2P/text-to-online respondents were given 24 hours to complete. This "slow response" window can introduce social desirability effects or allow respondents to look up information before answering candidate favorability questions.
- Non-white respondents number only 11 in the crosstab (n=11 out of 433), making any race/ethnicity breakdowns statistically unreliable. The poll wisely does not headline these comparisons, but they remain in the crosstab and could be misread.
The Age Faultline: The Poll's Core Structural Story
The single most important hidden trend is the nearly perfect inverse relationship between age and Massie support, which the topline understates.
Among voters 18–29, Massie leads 78.5%–21.5%; among 30–44 it's 64.9%–35.1%; among 45–64 it narrows to 53.3%–46.7%; and among 65+ Gallrein leads 65.0%–35.0% . This is not a gradual fade — it is a structural coalition split. Gallrein's entire competitive viability rests on a single demographic bucket.
The Turnout Trap: Where Gallrein's Lead Could Matter
The crosstab reveals a critical compounding dynamic: the age cohorts most favorable to Gallrein are also the most certain to vote.
Among voters 65+, 87.5% say they are "certain to vote," compared to 67.1% for ages 18–29 and 72.0% for ages 30–44. This is the central electoral tension in the race. Massie's younger coalition is structurally less reliable in a low-turnout closed primary environment. The topline headline — "Massie leads by 2 points initially, 52.4%–47.6% on leaners" — is a turnout-weighted figure, but the raw enthusiasm gap for that 65+ cohort creates a scenario where even a modest overperformance in senior turnout could flip the result.
Trump Approval: A Below-Average District for the Endorser
The poll's most politically significant finding — that Trump's disapproval among district Republicans is above the national average — is buried in the narrative and deserves sharper emphasis.
Trump's "strongly approve" rate among 65+ voters is 66.2%, but it falls to just 24.2% among 30–44 year olds and 26.9% among 18–29 year olds. Overall, 32.8% of Registered Republicans in the district disapprove of Trump's performance — higher than the national average — driven by his handling of inflation/cost of living (53.9% disapproval), the Iran War (59.1% disapproval), and the Epstein files (underwater at 40.9% disapprove vs. 50.0% approve) . The endorsement's efficacy is therefore structurally age-gated: it reinforces Gallrein's existing base (65+) but has limited penetration where Massie needs to be weakened (18–44).
Party ID Irregularity: A Sampling Concern
A significant methodological flag concerns who is in the sample. Kentucky is a closed primary state, meaning only registered Republicans can vote . Yet the crosstab shows that among the 433 Registered Republicans sampled:
- 76.4% self-identify as Republican
- 20.1% self-identify as Independent
- 3.5% self-identify as Democrat
The Independent-identifying registered Republicans are not just an ideological quirk — they are electorally unreliable. The crosstab shows they are only 61.1% "certain to vote," versus 86.7% for self-identified Republicans. This 25-percentage-point certainty gap is statistically significant per the MarketSight flags in the crosstab. If Independent-ID Republicans are being sampled at a rate that doesn't reflect actual primary turnout (because they're less likely to show up), the poll may be underweighting the most reliable, partisan Republican voters — who also happen to lean Gallrein via the Trump connection — or conversely, overweighting the independent streak that favors Massie. The poll uses a likely-voter weight from the L2 voter file, which should partially correct for this, but the magnitude of the certainty gap warrants scrutiny.
The "America First vs. MAGA" Internal Split
One genuinely hidden trend is the distinction BIG DATA POLL draws between "America First" and "MAGA" identifiers. Among the 60.1% of Republican self-identifiers who call themselves "America First/MAGA," when forced to choose between the two labels, 67.8% chose "America First" and 32.2% chose "MAGA" . This is not trivial. Massie explicitly brands himself as a constitutionalist and libertarian-leaning independent, not a MAGA loyalist, and the crosstab suggests the pure-MAGA segment (32.2% of the 60.1%, so roughly 19% of all Republican-ID respondents) is the core of Gallrein's ideological support. America First identifiers, by contrast, skew toward Massie — a nuance entirely obscured when the two labels are collapsed into a single "MAGA" shorthand in most outside commentary on this race.
Geographic Concentration Risk
The crosstab shows Kenton County backing Massie 64.5%–35.5% — a decisive margin . Boone County is essentially a toss-up (Gallrein leads 51.3%–48.7%) . Campbell County and the rural counties are described as split. BIG DATA POLL Director Rich Baris noted that Boone has a higher-than-district-average share of younger voters (18–44 may be 40–45% of Boone's vote) , which limits Gallrein's ceiling even where he is nominally ahead. The crosstab county subgroups have very small n's for many rural counties (single digits in several cases), meaning those county-level figures are essentially illustrative rather than inferential.
Summary Bias Assessment
Potential Bias Direction Severity Online panel (CINT) inclusion Enthusiasm inflation, likely slight Massie inflation Low-moderate Independent-ID voters sampled at high rate May overstate Massie's "independent streak" coalition Moderate Age turnout gap (65+ most certain, backs Gallrein) Poll's leaners may understate Gallrein's actual Election Day number Moderate-high Non-white subgroup (n=11) Unreliable; should not be reported as standalone findings High for subgroup 24-hour text-to-online completion window Social desirability / information-gathering risk Low Small rural county n's Cannot infer county-level results outside major 3 counties High for subgroup The net effect is that the poll likely modestly overestimates Massie's margin — not because of bad faith, but because the structural reliability advantage of the 65+ Gallrein coalition will likely manifest more fully on Election Day (May 19) than current likely-voter screens capture. The Sherlock Holmes framing here points to one conclusion that cannot be eliminated: a race framed as "Massie +4.8 on leaners" is, at the subgroup level, actually a toss-up determined entirely by whether high-school-educated voters over 65 in Boone, Campbell, and the rural counties show up at the rate they claim .
⁂
Analyze the linked topline poll and the attached crosstab spreadsheet for biases and hidden trends. Massie Leads, but the Kentucky Primary Remains Competitive
Review all four tabs in the attached spreadsheet: Overview, Toplines, Crosstabs, Methodology.
The full data is now thoroughly parsed. Here is a comprehensive analysis of the poll, covering methodology, structural biases, topline framing, and the hidden trends revealed by the crosstabs.
Quantus Insights — KY-CD4 Republican Primary Poll: Full Analysis
Methodology & Structural Issues
The survey was conducted April 6–7, 2026, via SMS to mobile phones drawn from a Republican voter file, covering Kentucky's 4th Congressional District. The sample size of 438 "likely voters" was screened from an initial 464 registered Republicans, with likely voters defined as anyone answering "Definitely" or "Probably will vote" on Q1. That screen is extremely permissive — 94.8% of the initial pull qualified — raising an immediate red flag. A genuinely rigorous likely voter screen for a primary election, which historically draws under 20% of registered partisans in a district like CD-4, would produce a far smaller qualifying rate. Retaining nearly the entire pull means the "likely voter" filter did almost no work, and the sample almost certainly over-represents low-frequency voters relative to election day turnout.[^1_1]
The weighting variables disclosed are age, gender, education, and geographic area. Critically absent is any weighting by primary voting history — the single strongest predictor of who actually shows up in a low-salience congressional primary. The unweighted distribution of primary frequency in the sample skews toward infrequent voters: frequency-0 voters (n=147, or roughly 34% of the ballot-test respondents) are actually the largest segment polled. If heavy primary voters (frequency 3–4) dominate actual election day turnout, this sample composition systematically deflates Massie's true lead.[^1_1]
The Headline Number Is Probably Too Narrow
The topline ballot test shows Massie at 46.8%, Gallrein at 37.7%, and undecided at 14.1%. That is a 9.1-point gap — outside the ±4.4% MoE but not by much. However, the crosstabs reveal a much more dramatic story by turnout composition. Among the most frequent primary voters (frequency-4, n=81), Gallrein actually leads Massie 44.7% to 35.4%. Among frequency-3 voters (n=68), it's Massie 52% to Gallrein 40.2%. The implication is a strong reversal: the voters most certain to appear on May 19 are Gallrein's best demographic, while Massie's strength is concentrated in the first-time and low-frequency segments who are least reliable.[^1_1]
The age breakdown reinforces this structural concern. Massie dominates among the youngest voters: 66.6% among 18–29s and 76.4% among 30–44s. But Gallrein leads outright among voters 65+ (50.5% to 29.2%), and older voters dominate Republican primary turnout — in most off-year primaries in districts like CD-4, voters 65+ represent roughly 40–50% of the primary electorate. This means the poll is again likely overcounting Massie-leaning segments.[^1_1]
Q3 Past Vote Recall — A Built-In Massie Skew
Q3 reports that 68.4% of respondents say they voted for Massie in the last Republican primary. This is a classic recall bias / self-selection artifact. Massie did win his last primary, but in a Republican primary in a safe R district, recall vote share is routinely inflated toward the winner — voters reconstruct their past behavior to align with the winner or with their current identity. More importantly, a poll recruited from a Republican voter file targeting self-identified Republican primary voters will over-sample people who have voted in Massie's prior primaries simply because those primary voters are more likely to have responded to the survey in the first place. This creates a circular reinforcement: the sample looks more Massie-loyal on Q3, and that Massie loyalty partly explains his ballot-test performance on Q2.[^1_1]
Q6 Trump Endorsement — The Most Underplayed Finding
The topline buries what may be the most consequential signal: Trump's endorsement of Gallrein makes 36.5% of respondents more likely to support Gallrein and only 26.2% less likely, a net +10.3 point boost. The write-up frames this as "mixed," but that framing is misleading. A net +10 on a presidential endorsement in a Republican primary is a significant asset, especially given that 36% say it makes no difference (i.e., the endorsement doesn't cost Gallrein anything with a plurality of voters).[^1_1]
The age crosstab on Q6 reveals a striking inversion. Voters 65+ react most positively to Trump's endorsement (net +26 for Gallrein), while 18–29s react most negatively (50.6% less likely vs. 9.0% more likely) and 30–44s are nearly as negative (42.9% less likely vs. 10.2% more likely). Given that older voters turn out at far higher rates in primaries, the endorsement's net electoral impact is probably more favorable to Gallrein in actual vote-weighted terms than the headline +10 suggests.
Q7 Question Wording — Priming for Massie
Q7 asks respondents to choose between "a Republican who strongly supports President Trump" versus one who "votes independently and stands by his principles, even when he disagrees with President Trump". The word principles is embedded exclusively in the independent-minded description — the Trump-loyalist option receives no comparable value-laden framing. This asymmetry is a textbook acquiescence/social desirability nudge in Massie's direction. When an otherwise equivalent political choice is framed as "principled independence" vs. "strong Trump support," respondents who value autonomy as a trait are primed toward the former regardless of their actual candidate preference. The 49.9% to 37.4% "independent-minded" lead should be read with this framing bias in mind.
Favorability Numbers Tell a Complicated Story
Massie's net favorable is only +5.0 (51.4% favorable, 46.4% unfavorable). For a 12-year incumbent seeking renomination within his own party's primary electorate, a 46.4% unfavorable rating is strikingly high — it suggests meaningful incumbent fatigue. Gallrein's net is +1.5 (39.8% favorable, 38.3% unfavorable), but 21.9% say they haven't heard enough to have an opinion. As name ID builds toward May 19, Gallrein's numbers have more room to move; Massie's ceiling is constrained by already-high unfavorability among the very people voting in this poll.
Key Structural Biases at a Glance
Dimension Direction of Bias Effect Likely voter screen (94.8% pass rate) Overly permissive Inflates low-frequency voters → Massie No primary-history weighting Under-weights habitual primary voters Inflates Massie (Gallrein leads freq-4 voters) Q3 past vote recall Winner-recall bias + Massie-voter self-selection Artificially reinforces Massie-loyal sample composition Q7 "principled" framing Asymmetric value language Over-counts independent-minded preference → Massie Trump endorsement (Q6) framed as "mixed" Topline understatement Understates Gallrein's net +10 endorsement boost Older voters (Gallrein's best demo) underrepresented Age-frequency interaction Massie lead overstated in horse-race number What the Evidence Suggests
Applying Holmesian elimination: the impossible scenario is that this poll has no systematic bias. The structural issues all point in the same direction — toward overstating Massie's lead. The most probable reading, consistent with the crosstab evidence, is that Massie's true likely-voter lead is narrower than 9.1 points, possibly within the MoE when the sample is reweighted by actual primary voting frequency and age turnout patterns. The poll is not fabricated — it reflects real SMS respondents drawn from a Republican voter file — but the permissive likely-voter screen, the absence of primary-history weighting, the Q7 framing, and the topline's downplaying of the Trump endorsement effect combine to create a portrait that systematically favors the incumbent. The "competitive" framing in the headline is actually underselling Gallrein's position among the voters who will disproportionately show up on May 19.
⁂
Compare these two polls for convergence and divergence. Analyze the linked topline polls and the attached crosstab spreadsheets for biases and hidden trends. Massie Leads, but the Kentucky Primary Remains Competitive
Republican Primary Poll for Kentucky House District 4 | BIG DATA POLL
Review all four tabs in the attached spreadsheet: Overview, Toplines, Crosstabs, Methodology.
I now have comprehensive data from all four sources. Let me synthesize the full analysis.
KY-04 Republican Primary: Dual-Poll Analysis
Both polls agree on the fundamental conclusion — Thomas Massie leads Ed Gallrein heading into the May 19, 2026 primary — but they diverge substantially on the size of that lead and several key interpretive dimensions. The divergence is analytically significant, not merely cosmetic.
Topline Vote Shares
Metric Quantus Insights Big Data Poll Field dates April 6–7 April 3–4 & 6–7 Sample (likely voters) 438 433 Margin of error ±4.4% ±4.0% Mode SMS to voter file Online (Lucid/CINT) + live P2P SMS Sponsor Self-sponsored, public release Crowdfunded (Public Polling Project) Massie (first choice) 46.8% 42.9% Gallrein (first choice) 37.7% 40.6% Massie net lead +9.1 pts +2.3 pts Undecided 14.1% 16.5% Massie with leaners Not reported 52.4% Gallrein with leaners Not reported 47.6% This is a striking 6.8-point gap in the headline lead between polls conducted almost simultaneously. That delta exceeds both surveys' margins of error, meaning the two polls are statistically incompatible on toplines — they cannot both be precisely correct.
Convergences
Despite the headline divergence, the polls agree on several structural findings:
- Massie leads across most age cohorts under 65. Quantus shows Massie at 76.4% among 30–44-year-olds; BDP shows him at 64.9% with the same group. Both confirm Gallrein's lead is concentrated among voters 65+.
- Gallrein dominates among older, habitual primary voters. Quantus crosstabs show Gallrein winning 50.5% vs. Massie's 29.2% among 65+ voters; BDP reports Gallrein leading 65.0%–35.0% in that cohort.
- Undecideds are substantial (14–17%) and are breaking toward Massie when pressed for a lean, per BDP's follow-up question.
- The Trump endorsement has limited net impact. Quantus finds only a net +10.3 "more likely" effect, with 36% saying no difference; BDP independently notes the endorsement's positive effect is "largely limited to 65 and older voters."
- Massie's favorability is polarized. Quantus has him at net +5.0 (51.4% fav / 46.4% unfav); BDP reports 46.2% favorable / 46.3% unfavorable — deeply polarized but Massie still marginally ahead or even.
- The district's "independent streak" is real and documented. BDP finds 20.1% of registered Republicans self-identify as Independent, and 23% of those formerly identified with MAGA/America First — a notably higher defection rate than the national average of 15–18%.
Key Divergences and Probable Sources of Bias
1. The 6.8-point lead gap — sample composition most likely culprit
Quantus used SMS-only targeting from the voter file and weighted by age, gender, education, and area. BDP used a mixed mode (online panel via Lucid/CINT + P2P SMS), weighted additionally for race and geography at the county level. Mixed-mode polls with online panel components (Lucid/CINT) are known to skew toward higher-information, more partisan-engaged respondents. That could inflate Gallrein's showing if Gallrein's media buy (primarily anti-Massie digital ads) resonated disproportionately with online-active voters. Conversely, Quantus's SMS-only voter-file targeting likely over-represents registered Republicans with reliable contact data, who tend to be older and more habitual — which could modestly inflate Massie's numbers.
2. Gallrein's favorability — a hidden divergence
Quantus finds Gallrein nearly tied in net favorability: 39.8% fav / 38.3% unfav, with 21.9% having not heard enough. BDP finds 36.6% fav / 31.5% unfav but with 31.9% having no opinion. The "name recognition gap" is larger in BDP's data, suggesting Gallrein is less known in BDP's sample — which is consistent with BDP's heavier rural/exurban geographic coverage and its finding of only 8.6% of voters having had personal contact with Gallrein's campaign. This is a substantively different picture from Quantus, which shows Gallrein slightly above water despite heavy unfavorability.
3. Habitual voter frequency — a hidden Quantus finding
Quantus's crosstabs break out voters by primary voter frequency (0–4 prior primaries). This is analytically decisive: among high-frequency voters (4 prior primaries), Gallrein leads 44.7% to 35.4%. Among low-frequency voters (0 prior primaries), Massie leads 52.7% to 29.6%. This inversion is critical — it means the final outcome depends entirely on which electorate shows up. If habitual primary voters dominate turnout (as they typically do in low-salience primaries), Gallrein's position is far stronger than Quantus's 9-point headline suggests.
4. County-level geography
BDP's county-level crosstabs reveal a nuanced geographic split that Quantus does not report: Boone County (the suburban Cincinnati anchor) is backing Gallrein 51.3% to 48.7%, while Kenton County backs Massie 64.5% to 35.5%. Quantus's geographic breakdowns collapse to Rural/Suburban/Town without county-level resolution. Because Boone is the most populous county in the district and tends to produce a large share of Republican primary votes, BDP's Boone data alone could explain the gap — if Quantus undersampled Boone relative to BDP.
5. Trump approval as a diagnostic
BDP asked about Trump's job approval among its sample of registered Republicans and found 32.8% disapprove — higher than the national average, driven by his handling of inflation/cost of living (53.9% rate it negatively), the Iran conflict (59.1% disapprove of his handling), and the Epstein files (40.9% dissatisfied). Quantus did not ask a Trump approval question. This structural finding from BDP explains why the Trump endorsement has dampened utility for Gallrein: the district's Republican base is more MAGA-skeptical than average, and Gallrein's campaign is almost entirely premised on the endorsement as a persuasion vehicle.
Methodology Red Flags
Issue Quantus BDP Likely voter screen Restrictive — only "definitely/probably will vote" (94.8% pass rate, suspiciously high) Voter file–weighted confidence levels; 81.3% "certain to vote" Online panel component None (voter-file SMS only) Yes — 60 completes via Lucid/CINT Undisclosed sponsor agenda Self-sponsored, no disclosed partisan client Crowdfunded, explicitly non-partisan disclosure Geographic resolution Urban/Suburban/Rural/Town only County-level (all 22 counties) Race/ethnicity weighting Yes (but sub-30 sample sizes for all non-white cells) Yes, including non-white evangelical breakdown Question order bias Favorability asked after ballot test (anchoring risk) Ballot test → favorability → Trump approval (also ordered) Quantus's 94.8% likely voter pass rate (438 of 464 clearing the screen) is unusual. Most primary polls filter 60–80% of registered voters as likely voters; filtering out only 5.2% suggests either a very permissive screen or a voter file pre-screened so heavily for primary history that the sample was already stacked with habitual voters before the survey began. That would structurally suppress Gallrein's showing since, as noted above, habitual voters lean toward him in Quantus's own crosstabs.
BDP's use of Lucid/CINT for 60 of 433 completed interviews (~14%) introduces a modest panel-effect risk but is standard disclosure practice. Their four-day field period spanning Easter weekend is properly handled by excluding Easter Sunday, though the pause creates a question about whether voters interviewed post-Easter had meaningfully different exposure to campaign advertising.
Hidden Structural Trend: The Age-Frequency Inversion
The most important analytic finding buried in the crosstabs is the age-by-frequency inversion. Quantus data shows:
- 18–29: Massie 66.6%, Gallrein 19.4%
- 30–44: Massie 76.4%, Gallrein 12.8%
- 45–64: Massie 46.6%, Gallrein 39.4%
- 65+: Gallrein 50.5%, Massie 29.2%
BDP confirms this slope but finds a tighter 65+ margin. Meanwhile, Quantus's own frequency crosstabs show older, habitual voters (Frequency 4) favor Gallrein 44.7%–35.4%. The synthesis from Sherlock's method: eliminate impossible explanations. You cannot explain Gallrein's 50.5% among 65+ voters by enthusiasm alone — BDP independently found that Gallrein's ground game reached only 8.6% of voters vs. Massie's 35.5%. The more probable explanation is that Gallrein's multi-million-dollar negative ad campaign is disproportionately reaching and moving older, TV-heavy consumers, while younger voters — who consume less broadcast media — remain more resistant. This makes the race a turnout arithmetic problem: if 65+ voters turn out at their historical rate and younger voters stay home (as they often do in primaries), the lead narrows or disappears.
Integrated Assessment
Applying the available evidence: the most probable picture is that Massie holds a real but narrower-than-Quantus lead, something in the range of 4–7 points on first choice. The Quantus +9.1 appears inflated by an overly restrictive likely voter filter and SMS-only methodology; the BDP +2.3 may slightly understate Massie's advantage because the online panel component over-represents digitally engaged, ad-exposed voters. BDP's leaner follow-up (Massie 52.4%–47.6%) is the single most predictive data point in either poll, since primaries with large undecided blocs typically break toward the incumbent when voters are forced to choose. The decisive variable the models still cannot resolve is whether Gallrein's 65+ coalition, which runs large and deep, can generate disproportionate turnout sufficient to overcome Massie's breadth advantage across all other age groups.
⁂
-PJ
