A Marist Poll Put Hinojosa Ahead of Abbott, and the Market Treats It as One Poll Among Three

The Marist Poll published its September survey of Texas on September 23, 2026, and it has Gina Hinojosa at 49% and Greg Abbott at 46% among 1,139 registered voters interviewed September 17 to 20, a three-point edge inside the poll's 4.4-point margin of error. The same survey found that 68% of those voters expect Abbott to be re-elected, and that is the figure the market has been listening to, which is why, as of 4:34 p.m. Central on September 23, 2026, Kalshi priced a Republican winner of the governor's race at 73% and a Democratic winner at 28%.

Marist's release is the third statewide survey of the race in three days and the only one of the three with the Democrat ahead. Texas Southern University's Barbara Jordan Public Policy Research and Survey Center released a likely-voter survey on September 22, 2026 with Abbott at 49%, Hinojosa at 45%, the Libertarian Pat Dixon at 2% and 4% undecided, from 1,800 respondents interviewed September 15 to 19 with a margin of 2.31 points. The NBC5 Dallas and Telemundo poll published September 21, 2026 had Abbott at 48% and Hinojosa at 41% among 625 likely voters surveyed September 8 to 10. Texas votes on November 3, 2026. Abbott has held the office since January 20, 2015; Hinojosa has sat in the Texas House of Representatives since January 10, 2017; and Ballotpedia's general-election list on September 23, 2026 carried the two of them, Dixon and a set of write-in candidates.

The three surveys differ in who they counted more than in what they found. Marist interviewed registered voters and did not publish a likely-voter screen; Texas Southern and NBC5 screened for people likely to vote, and both of those samples leaned toward the incumbent, which is why the same week can produce a Democratic lead and a Republican one without any poll being wrong. Read as a set, the week's polling says the race is close among everyone who could vote and less close among the people who probably will, which is a distinction the market has a price for.

One book, one price

Kalshi is the only exchange pricing the governor's race as a standalone party contract on the Prediction Genius board. Polymarket's listing there is a combination contract that bundles the governor's race with the Senate race, so it is not a price on this race alone and it is not quoted here. On the one book that prices the race by itself, 73% for a Republican winner as of September 23, 2026 sits between the two likely-voter polls, which have Abbott ahead by four and by seven, and Marist's expectation question, which has Abbott winning in the minds of more than two voters in three. It is nowhere near the registered-voter horse race, which means the market is treating Marist's topline as one reading among three and not as the new state of the race.

The gap between what Marist's respondents prefer and what they expect is the most useful number in the release. Preference runs 49% to 46% for Hinojosa; expectation runs 68% for Abbott. A contract at 73% for the Republican side is closer to the second figure than to the first, and the likeliest reason is the sample. Registered voters include people who will not vote, and both surveys in the set that screened for likely voters tilted the race toward Abbott, by four points at Texas Southern and by seven at NBC5. That is the ordinary pattern in the polling of September 21 to 23, 2026, not an anomaly in Marist's.

Marist's own series shows how far the ground has shifted even so. Its generic congressional ballot in the same survey was 49% Democratic to 46% Republican, and the pollster noted that its October 2024 Texas survey had Republicans ahead on the same question by 53% to 46%, a ten-point swing in one pollster's series in under two years. A market that prices a Republican governor at 73% on September 23, 2026 is not disputing that swing; it is pricing an incumbent in office since 2015 against a challenger a fifth of likely voters said they did not yet know, and paying the incumbent for the difference.

The contrast with the Senate race

The exchanges treat the two statewide races on the same ballot very differently. As of September 23, 2026, Kalshi priced the Democratic side of the Texas Senate contract at 59%, while the same book had the Democratic side of the governor's race at 28%: a 31-point gap between two races that the polls of September 21 to 23, 2026 placed within a few points of each other. Marist had James Talarico ahead of Ken Paxton by six and Hinojosa ahead of Abbott by three; Texas Southern had Talarico ahead by one and Abbott ahead by four. The market's split says something the toplines do not: it reads the Senate race as open and the governor's race as an incumbent's to lose, and only one of those two readings has an incumbent in it. The reasoning behind the Senate price is in the Texas Senate market analysis.

Marist's expectation question makes the same distinction from inside one survey. Regardless of whom they supported, 52% of its respondents thought Paxton would win the Senate race, while 68% thought Abbott would be re-elected. The expectation gap is sixteen points wider for the governor than for the senator on the same ballot, in the same sample, in the same week, which is the shape of the market's two prices in miniature.

What the record says for Hinojosa

The case that the market is slow to move is in the same polls. Marist found Hinojosa with 52% of independents. Texas Southern found that 21% of likely voters did not yet know enough about her to hold an opinion, which is room a challenger can still fill six weeks out and an incumbent in office since 2015 cannot. The professional raters have not budged: Ballotpedia's summary on September 23, 2026 listed the Cook Political Report, Inside Elections and Sabato's Crystal Ball all at likely Republican. A three-point Democratic edge among registered voters set against a likely-Republican rating is the kind of disagreement a 73% price averages over, and the registered-voter number makes the argument that the price could be wrong on its own.

That 21% is the figure that can still move. A challenger unknown to a fifth of the electorate in late September has upside that a sitting governor does not, and it is also the number an advertising campaign on either side is built to close, in one direction or the other, before early voting opens on October 19, 2026.

The campaign's week

Hinojosa's campaign spent September 22, 2026 on the air war rather than on the polls. In a statement released that day, after what it described as a new Abbott advertisement about her record, the campaign said Abbott "has to lie about my record because he can't defend" his own on property taxes, and repeated its claim that he will not debate her. The statement is the campaign's characterisation of the advertisement; the advertisement's own text was not part of the release.

On debates, the campaign's September 15, 2026 statement listed four outlets whose proposals it said Abbott had not answered: NBCUniversal and Telemundo's Texas stations with Hearst Newspapers, Nexstar Media Group with KXAN, Spectrum News and KVUE. Hinojosa accepted the NBC and Hearst invitation, a one-hour debate in Fort Worth on October 15, 2026, in a statement the San Antonio Current reported on July 1, 2026, and the Nexstar and KXAN proposal was for September 25. Neither statement records an answer from Abbott's campaign.

What is next

The next fixed dates belong to the Secretary of State. October 5, 2026 is the last day to register to vote in the November 3 election; early voting by personal appearance runs October 19 to October 30. Between them sits the proposed October 15 debate in Fort Worth, which becomes an event only if Abbott agrees to it. The Kalshi contract resolves on the party of the person inaugurated as governor after the 2026 election.