Two surveys released on the same day put Democrat James Talarico at an identical 49%, but show different levels of support for Republican Ken Paxton. The gap highlights why polling methodology matters as Texas votes.
- Two polls, two different margins
- Fox News finds a narrower margin than in September
- Republican support for Senate control does not automatically translate into support for Paxton
- Why the Cook poll shows a different picture
- Texas has entered the voting period
- What the new polling tells us — and what it does not
- Sources and methodology
Two major polls released on October 21 have offered different snapshots of the 2026 Texas Senate race, less than two weeks before the November 3 midterm elections.
A new Fox News survey puts Democrat James Talarico and Republican Ken Paxton tied at 49% each among likely voters. Meanwhile, a Cook Political Report survey gives Talarico a five-point advantage, 49% to 44%.
The striking detail is that Talarico receives precisely the same percentage in both surveys. The difference lies entirely in Paxton’s measured support.
That does not establish a sudden five-point movement in voter preferences. Instead, it provides an example of how independent polls can produce different results even when published on the same day.
Two polls, two different margins
The surveys were conducted over different periods and used separate samples of likely voters.
| Pollster | Talarico (D) | Paxton (R) | Fieldwork | Sample |
|---|---|---|---|---|
| Fox News | 49% | 49% | Oct. 15–19 | 1,398 LV |
| Cook Political Report | 49% | 44% | Oct. 9–13 | 934 LV |
LV = likely voters. Both surveys were released October 21, 2026. Figures are individual polling results, not a polling average.
Fox News conducted its survey under the joint direction of Beacon Research and Shaw & Company Research, using telephone interviews and responses completed online after text-message invitations. The likely-voter findings carry a reported sampling margin of error of approximately ±2.5 percentage points.
Cook’s poll was conducted for the publication by Democratic firm New River Strategies and Republican firm GS Strategy Group as part of its Battleground Project.
The differences in fieldwork dates, sampling and likely-voter models mean their findings should not be interpreted as consecutive measurements from a single tracking survey.
Fox News finds a narrower margin than in September
The latest Fox result also provides a useful comparison with the pollster’s previous likely-voter survey.
In the September 24–28 Fox News poll, Talarico led Paxton 51% to 49%.
In the October 15–19 survey, the candidates are tied at 49%.
| Fox News survey | Talarico | Paxton | Margin |
|---|---|---|---|
| September 24–28 | 51% | 49% | D +2 |
| October 15–19 | 49% | 49% | Tie |
The measured two-point change comes from a lower Talarico figure, while Paxton remains at 49%.
However, the difference is small relative to the surveys’ sampling uncertainty. It cannot establish a statistically significant change in voter preferences without further analysis.
Importantly, Fox also finds that Talarico maintains a seven-point advantage among the broader population of registered voters in its latest survey.
The difference between the registered-voter and likely-voter results illustrates another crucial issue: turnout assumptions can materially affect a poll’s estimated margin.
Republican support for Senate control does not automatically translate into support for Paxton
One of the more revealing findings from Fox is the distinction between voters’ preferred party control of the Senate and their choice between the two Texas candidates.
While 51% of likely Texas voters prefer Republican control of the Senate and 47% prefer Democratic control, the candidate matchup is tied.
That divergence suggests some voters who prefer a Republican-controlled Senate are not necessarily supporting the Republican nominee in this particular contest.
The poll reports that 98% of Democrats support Talarico, compared with 90% of Republicans supporting Paxton. It also identifies crossover support for Talarico among Republicans who do not identify with the MAGA movement.
These results help explain how a statewide Republican-leaning electorate can produce a competitive Senate race.
They do not, however, establish how those voters will cast their ballots.
Why the Cook poll shows a different picture
Cook’s survey places Talarico at 49% and Paxton at 44%, creating a larger Democratic margin than the latest Fox poll.
The Cook findings also place Texas alongside Ohio and Iowa in a three-state survey of competitive Senate races.
Across these states, the publication reports that inflation, corruption and everyday economic concerns are prominent among voters’ priorities. In Texas, Cook finds Talarico ahead of Paxton on perceived ability to handle corruption, inflation and the overall economy, while Paxton receives stronger evaluations on immigration and border policy.
These findings provide political context for the candidate numbers, but they should not be used to infer that any single issue caused the difference between the Cook and Fox results.
Nor should their five-point divergence be confused with a measured five-point swing over time.
Texas has entered the voting period
The two surveys were published after early in-person voting began in Texas on October 19.
Voting continues through October 30, with Election Day scheduled for November 3.
That timetable adds urgency to the polling discussion, but early-voting participation figures cannot establish which candidate is receiving more votes. Ballots remain secret, and participation varies by geography, voting method and demographic group.
Votendi previously examined these questions in Texas starts voting: why the Talarico–Paxton Senate race is now a test of turnout.
What the new polling tells us — and what it does not
The strongest common finding is not that one of the surveys must be wrong.
Both record 49% support for Talarico, while measuring different levels of support for Paxton. Fox’s latest likely-voter survey shows a tie, while Cook records a five-point Democratic lead.
The results are also consistent with a broader series of October surveys showing a competitive contest, though with varying candidate margins.
Three distinctions are essential when interpreting the closing stages of the race:
- A single poll is a snapshot, not an election result.
- Differences between independent pollsters are not necessarily evidence of a trend.
- Registered-voter and likely-voter estimates describe different survey populations.
As November 3 approaches, subsequent surveys may help clarify whether the differing margins persist. Until then, the Cook and Fox findings should be treated as separate measurements rather than competing predictions of the election result.
For broader context, see Votendi’s United States 2026 midterms guide.
Sources and methodology
Fox News, October 21, 2026 — October 15–19 fieldwork, 1,398 likely voters, reported margin of error ±2.5 percentage points.
Fox News, October 1, 2026 — September 24–28 fieldwork, 881 likely voters, reported margin of error ±3 percentage points.
Cook Political Report, October 21, 2026 — October 9–13 fieldwork, 934 likely voters in Texas.
Cook Political Report survey toplines — October 2026 Battleground Project.
Editorial note: All polling figures refer to surveys rather than official election results. This analysis does not constitute an election forecast.
