A viral ZeroHedge piece says the Democratic sweep in Texas does not add up. RSN checked its figures against the Texas Secretary of State’s returns. Several of its numbers are wrong, its central “statistically impossible” argument rests on a misreading of net vote totals, and Texas has seen a Democratic surge of similar size before.
Since Democrats won every statewide office on the Texas ballot on November 3, a piece on ZeroHedge titled “There’s Something Suspicious About The Texas Midterm ‘Blue Wave’” has been circulating among conservatives. It argues that Republican vote totals held steady while Democrats gained roughly two million votes, and that “nearly every single new participating voter in 2026 would need to vote Democrat for this result to happen.” It adds: “Many would consider this statistically impossible.” The piece stops short of alleging fraud outright, but closes: “Maybe the midterms are completely above board… Or, maybe something else is going on.”
The piece runs under “Tyler Durden,” the pen name ZeroHedge uses as a house byline. It names no author, cites no election-office data and links to two news articles and a social-media post. RSN’s earlier look at the Talarico race found the result was driven by a Latino swing and a weak Republican nominee. This follow-up tests the specific numbers the ZeroHedge piece relies on.
What the coverage misses
Much mainstream coverage has dismissed post-election doubts in a sentence or two. That leaves readers who have seen the ZeroHedge charts with no way to check them. The claim deserves a direct answer, because it is built on real numbers, and some of them are right. Texas Republicans did receive almost exactly as many votes as they did in 2022. Democrats did gain well over a million and a half. Talarico did receive more votes than Kamala Harris did in a presidential year.
The question is whether those facts point to anything other than a wave election. RSN analysis: they do not, and the official returns themselves show why.
Background: the official returns
The Texas Secretary of State’s results page (updated November 4 at 4:59 p.m., with all polling locations reporting) shows Gina Hinojosa with 5,230,789 votes for governor to 4,524,527 for Gov. Greg Abbott, out of 9,907,381 cast. In the U.S. Senate race, James Talarico received 5,444,982 votes to Ken Paxton’s 4,249,373. The figures are unofficial until the state canvass and differ slightly from later Associated Press (AP) tallies, but by a few thousand votes, not hundreds of thousands. The Secretary of State is appointed by the governor; Abbott has been governor since 2015.
In 2022, Abbott won 4,437,099 votes to Beto O’Rourke’s 3,553,656, out of 8,102,908, according to certified results compiled by Ballotpedia. About 9.8 million Texans voted this year, 51.6 percent of a record 19 million registered voters, compared with 45.9 percent in 2022 and 53 percent in 2018, the Texas Tribune reported from preliminary state data. Abbott conceded at about 10 p.m. on election night and called Hinojosa to offer help with the transition, Lubbock Lights reported.
Deep dive: checking the numbers
Abbott’s total and the “extra 2 million”
ZeroHedge says Abbott’s “final tally” was 4.59 million and that Democrats “added an extra 2 million votes” for Hinojosa. Both figures are off. Abbott received 4,524,527 votes, up 87,428 from 2022. Hinojosa’s total was 1,677,133 higher than O’Rourke’s. The 4.59 million figure matches the combined Republican vote in this year’s U.S. House races (4,590,522 by RSN’s count of the state’s district results), not Abbott’s.
The House comparison
The piece says Democratic House candidates won about 3 million votes in 2022 and 5.16 million in 2026, out of 7.56 million and 9.78 million cast. RSN’s sum of the 38 district results on the state’s results page gives 5,145,910 Democratic and 4,590,522 Republican votes this year, out of 9,780,054, so the 2026 figures are close. The 2022 comparison leaves out a key fact. That year Democrats fielded candidates in only 32 of 38 districts, according to the statewide House summary, which puts total votes at 7,756,761, not 7.56 million. This year every district had both a Republican and a Democrat on the ballot. Part of the Democratic House “gain” is simply six districts where Democrats could not receive votes in 2022 because no Democrat was running.
“Statistically impossible”
This is the heart of the argument. In the governor’s race, total votes rose by 1,804,473 from 2022, and the Democratic gain equals about 93 percent of that increase. ZeroHedge reads this as “nearly every single new participating voter” choosing Democrats.
RSN analysis: that inference treats net change as if it were the movement of individual voters. It is not. The 2026 electorate is not the 2022 electorate plus 1.8 million newcomers. Some 2022 Abbott voters stayed home; some voted for Hinojosa; some new voters chose Abbott. A flat Republican total is what you would expect if Republican losses to defection and abstention were roughly offset by new Republican voters, while Democrats both held their base and drew new and crossover voters.
There is survey evidence that this is what happened. ABC News’ Texas Voter Poll, a survey of more than 5,000 Texans who voted, found that voters who backed Donald Trump in 2024 made up 48 percent of this year’s electorate and Harris voters 41 percent, according to a November 6 analysis by Keaton Fox of ABC13 Houston. That electorate, Fox wrote, was one “Trump would have carried by seven points. Talarico won it by 12.” The survey found that about one in nine 2024 Trump voters backed Talarico, independents went 63-34 for him, and the 14 percent of voters casting their first midterm ballot went 74-25. Such surveys carry error margins and are usually weighted to the final count, so they cannot prove the count itself is right. What they do show is who voters said they were and how they said they voted, and that picture is one of crossover voting.
The precinct data point the same way. In Harris County, Fox found that Talarico ran 8 to 12 points ahead of Democrat Colin Allred’s 2024 share “in every single band” of precincts, and carried the most pro-Cruz precincts 51 to 49 percent. “That’s not a turnout pattern,” Fox wrote. “This moved all of them by about the same amount – which is what persuasion looks like on a map.”
Has this happened before?
ZeroHedge says “no voting shift of this magnitude has happened in Texas for at least the past 50 years” and asks how midterm Democrats could outpoll a presidential nominee. Texas’s own recent history answers both points.
In the 2014 governor’s race, Democrat Wendy Davis received 1,835,596 votes to Abbott’s 2,796,547 (2014 results). Four years later, Beto O’Rourke received 4,045,632 votes against Ted Cruz’s 4,260,553 in the Senate race, per the Secretary of State’s 2018 canvass. The top Democratic vote rose by 2,210,036 between those two midterms, more than this year’s 1,677,133. The margin moved about 17.8 points, from Abbott +20.4 to Cruz +2.6, close to this year’s 18-point move in the governor’s race, from Abbott +10.9 in 2022 to Hinojosa +7.1. (The 2014-2018 comparison spans two different offices, which is why RSN treats it as a rough benchmark.)
And O’Rourke’s 4,045,632 votes in 2018 exceeded Hillary Clinton’s 3,877,868 in Texas in 2016, according to the state’s 2016 canvass. A Texas Democrat outpolling the previous presidential nominee in a midterm is not new. Talarico’s margin over Harris (609,732 votes) is larger, but in a year with 19 million registered voters, against 15.1 million in 2016, a larger raw number is not by itself a red flag.
The polling miss
Here ZeroHedge quotes ABC13 accurately: the final polling averages missed by about nine points in the Senate race and nearly 11 in the governor’s race, the largest misses ABC13 could find in a Texas statewide race. What the piece leaves out is the rest of the same article. ABC13 attributed the miss to polls that “were built on an electorate that looked like 2022 and 2024,” and to polls showing Talarico winning “a far smaller share of Republicans and independents than he got.” In 2022 and 2024, the polls underestimated Republicans by three to five points. RSN analysis: a large miss in the other direction is evidence that polling is unreliable, which conservatives have argued for a decade, not evidence that the count was altered.
Noncitizen voting and registration claims
The piece says about 700,000 foreign migrants have moved to Texas since 2022, fewer than 100,000 of them citizens. It gives no source, and RSN could not confirm either number. It also concedes that it is “unlikely that millions of foreign residents voted en masse illegally.”
The best documented data come from the state itself. In 2025, the Secretary of State’s office ran the state’s 18 million registered voters through the federal SAVE (Systematic Alien Verification for Entitlements) database and flagged 2,724 potential noncitizens. A follow-up review found 578 had demonstrated citizenship, and on September 15, 2026, the office referred 117 cases to the Attorney General for investigation. Referrals are not convictions. Even the initial 2,724 flags amount to less than half of one percent of Hinojosa’s 706,262-vote margin. RSN has long argued that noncitizen voting deserves serious scrutiny, and Texas’s SAVE checks are the kind of scrutiny that produces real numbers. Those numbers are far too small to explain this result.
A separate viral claim, that Texas added 2.9 million registrations without proof of citizenship between 2022 and 2024, misreads a federal report, Houston Public Media found. The figure counts “new valid registrations,” including people moving between counties, while Texas removed nearly 1.8 million records over the same period. Net registration grew by about 952,000. Those registrations also predate the 2024 election, which Trump carried by 13.7 points.
Counter-arguments
The strongest case for doubt comes from the President. On November 9, Trump wrote on Truth Social: “Does anybody really believe that this ‘Election’ was honest. How could the Polls be so wrong???” He added: “Some early voting ‘dumps’ were so large that it was impossible to win, no matter how many people went to vote. It’s all about the machines, which are far more important than the voter!” The post was reported by Politico. The White House pointed to his earlier statement that he would accept the results “if they’re honest elections.” Houston Public Media reported that Trump had also shared a post questioning Hinojosa’s vote count.
RSN analysis: neither post identifies a county, a machine, a batch of ballots or a document. The early-vote pattern Trump describes is documented, but it has an ordinary explanation. ABC News’ survey found early voters went 57-42 for Talarico and Election Day voters split 49-48, so early returns, counted first, were bound to look lopsided. As of publication, RSN found no public fraud challenge filed by a Texas election official, county party or Republican candidate. The AP reported that FBI Houston publicly denied a viral claim that it had raided an office and “uncovered fraudulent ballots” on November 2.
Skeptics can fairly say that the results are still unofficial, that surveys are not proof, and that the absence of a filed challenge is not proof of a clean count. All true. The appropriate remedy is the one Texas already has: the state canvass, county audits and recounts, overseen by a Secretary of State appointed by a Republican governor. If any of those turn up discrepancies, they will be worth reporting. A gap between a polling average and the result is not one.
There is also a cost to conservatives in getting this wrong. The ZeroHedge piece itself notes that every losing side now reaches for the same explanation. Republicans who attribute the Texas result to fraud will not need to ask why the Republican vote was flat in a growing state, why Latino voters who split in 2024 broke heavily the other way, or why the party nominated a Senate candidate whom 62 percent of voters viewed unfavorably, per ABC News’ survey. Those are the questions that decide 2028. Texas was not alone, either: Democrats posted late-count gains in North Carolina and Mississippi the same week.
What is established, disputed and not established
Established:
- Abbott received 4,524,527 votes, about 87,000 more than in 2022; Hinojosa received 5,230,789, about 1.68 million more than O’Rourke (Texas Secretary of State, unofficial).
- Turnout was about 51.6 percent of 19 million registered voters, below the 53 percent of 2018.
- Democrats contested all 38 U.S. House districts in 2026 but only 32 in 2022.
- Between the 2014 and 2018 midterms, the top Democratic vote in Texas rose by about 2.21 million, more than this year’s increase, and O’Rourke outpolled Clinton’s 2016 total.
- The state’s SAVE review flagged 2,724 possible noncitizen registrants and referred 117 cases for investigation.
Disputed:
- How much of the swing came from crossover voters and how much from new or returning voters. ABC News’ survey points to substantial crossover; surveys carry error.
- Why the polls missed by nine to 11 points. ABC13 noted that the industry’s review of the 2020 polling miss took six months.
Not established:
- Any fraud or count error affecting the Texas result. No official finding, court filing or documented irregularity has been made public.
- ZeroHedge’s figures of 700,000 migrants and fewer than 100,000 new citizens since 2022; the piece cites no source.
- That the result was “statistically impossible.” The official numbers are consistent with a wave election of a kind Texas saw in 2018.
Sources: ZeroHedge; Texas Secretary of State, 2026 general election results; Texas SOS, 2018 canvass; Texas SOS, 2016 canvass; Texas SOS, Sept. 15, 2026 release; Ballotpedia, 2022 governor; 2014 governor results; 2022 U.S. House results in Texas; Texas Tribune; ABC13 Houston; Houston Public Media; Politico; AP; Lubbock Lights.
Featured image: Texas State Capitol, Austin. Photo by Daderot, CC0 1.0, via Wikimedia Commons.
How this article was made: Right Side News researched and drafted this article with the help of AI tools. Sources are linked so readers can check them. Spot an error? Tell us in the comments and we will correct it.
