← All fact checksAfternoon run · November 21, 2026 · #5 by reach
PoliticsEvent claim

Does the Texas risk-limiting audit only look at a sample of digital images?

Does Texas's first statewide risk-limiting audit of the Paxton–Talarico Senate race only examine digital ballot images rather than the paper ballots, as Patrick Byrne claims?

False

The audit checks hand counts of physical paper ballots against reported totals, not digital images.

False. Texas's first statewide risk-limiting audit, launched on 18 November 2026 for the 3 November U.S. Senate race, uses randomly selected paper ballots. Counties pull those ballots from ballot boxes, count them by hand and compare the totals with the reported electronic results. Ballot selection used publicly livestreamed dice rolls, not ballot images.

Claims

✕ 2 false

The Texas "risk limiting audit" just looks at a sample of digital images.12345

“The “risk limiting audit” just looks at a sample of digital images.”

✕ FalseCentral claimHigh confidence

This is the claim the whole post rests on, and the documented procedure contradicts it at every step. The Texas Secretary of State's 16 November 2026 release describes audits that use randomly selected paper ballots to verify reported outcomes, calling the process an independent check of electronic tabulation results against paper ballots. KERA News reported that the ballot selection itself was driven by a series of 10-sided dice rolls, streamed publicly; the El Paso Times and Newsweek describe the same paper-based method. Lead Stories went further and reviewed the press-conference footage attached to the original post, finding that no official said digital images would be used; Secretary of State legal director Chuck Pinney said on camera the audit is designed to review a statistically significant number of paper ballot records. The sources are not independent of one another in the usual sense — they all describe one official procedure — but they are a mix of the issuing office and separate newsrooms, and none points the other way. Nothing in the record supports the digital-image description.

The manipulation happens between the paper and the digital image.14

✕ FalseModerate confidence

This is an assertion with nothing behind it: the post names no county, no machine, no incident and no document. What the record does show is that the audit is built to catch exactly the kind of gap the post describes. Counties retrieve the selected physical ballots and count them by hand, then those counts are compared with the reported electronic totals. If a tally had been altered between paper and digital record, the hand count would not match the machine count — that mismatch is the signal the audit is designed to surface. In other words, the procedure the post calls useless is the procedure aimed at the scenario it invents. We rate the claim false rather than unverified because it is presented as a description of how the audit works, and that description is wrong. Our confidence is lower than for the central claim because the post offers no specific allegation we could test against a particular county or machine.

Still open: The post does not specify a county, machine or election where such manipulation allegedly occurred, so there is no concrete incident to examine.

The Texas risk-limiting audit is a waste of time.13

“This is a waste of time.”

– OpinionHigh confidence

"A waste of time" is a value judgement, not a checkable fact, so it falls outside true and false. What we can say is that the judgement rests on the procedural premise we rated false: the post calls the audit pointless because it supposedly inspects digital images. Once that premise goes, the conclusion loses its stated basis. The Secretary of State describes risk-limiting audits as a statistical check of reported outcomes against paper ballots, and news coverage describes counties manually reviewing randomly selected ballots and comparing them with reported results. Whether a reader thinks that check is worth the cost is a matter of opinion. Readers should weigh the opinion knowing the factual description it was built on does not match the documented procedure.

A text-only post: there was no media to test.

What was posted

X

A text-only post on X from the @PatrickByrne account about Texas's risk-limiting audit of the 2026 U.S. Senate race. It tells readers the audit is pointless because it supposedly examines a sample of digital images, and asserts that any manipulation happens between the paper ballot and its digital record. No image, video or document is attached in the version we analysed.

“This is a waste of time. The “risk limiting audit” just looks at a sample of digital images. The manipulation happens between the paper and the digital image.”

Posted on X by the @PatrickByrne account on 18 November 2026, the day the audit's random ballot selection took place. Lead Stories fact-checked the same post on 19 November 2026.

Who posted it

X

Account check

Looks like a genuine account12%
Likelihood of automation14%

Estimates from public signals — account age, posting pattern, audience — not a judgement of the person behind the account, and separate from whether this post is true.

What stood out

  • Account belongs to a real, identifiable public figure — Patrick M. Byrne, founder/former CEO of Overstock.com — so the low authenticity score reflects credibility as an information source, not identity uncertainty.
  • Long documented record of propagating false election-fraud claims: Dominion/Smartmatic conspiracy theories, 'drop and roll' ballot fraud claims, and funding of post-2020 'audit' efforts (NYT, Gizmodo, Business Insider).
  • Suspended by Twitter in January 2021 for election misinformation, then reinstated in December 2022 under Musk; currently active with roughly 443k followers and ~32.5k tweets (twtData snapshots, mid-2026).
  • July 2026: federal judge awarded Hunter Biden $1.7M in punitive damages, finding Byrne 'not credible, fabricates awesome and farfetched narratives' and that the Iran-bribery story was fabricated with 'conscious disregard' for truth (NBC, Courthouse News, Law&Crime).
  • Separate Dominion defamation litigation has produced default-judgment proceedings, court sanctions, and findings that Byrne violated protective orders by publicly dumping sealed discovery material (2024–2026 dockets).
  • Still fact-checked in the current news cycle: Lead Stories debunked an 18 Nov 2026 @PatrickByrne post claiming Texas' risk-limiting audit only examines digital images.
  • Ongoing activity in Nov 2026 includes election-conspiracy content and promotion of the Oltmann '15%' brief, amplified by aligned election-denial accounts.
  • Reports since 2024 that he relocated to Dubai and has not returned to the US; he has missed court appearances and owes unpaid sanctions, adding further accountability red flags.
  • Metadata supplied is incomplete (bio, follower count, creation date and verification all null/false) and conflicts with public trackers showing a verified, high-volume account — profile data should be treated as unreliable.
Signals (11)+
Identity
Patrick M. Byrne, ex-CEO Overstock.com, election-denial financier
Followers approx
443500
Tweets approx
32550
Posting cadence
~10-30 posts/day per tracker snapshots; human-consistent, not automated
Following approx
4833
Fact check history
Repeatedly debunked; latest Lead Stories fact check 19 Nov 2026
Legal findings
Default judgment + $1.7M punitive damages (Hunter Biden v. Byrne, Jul 2026); Dominion v. Byrne sanctions/default proceedings; Coomer defamation suit
Court credibility finding
Judge: 'not credible, fabricates awesome and farfetched narratives'
Residence
Reportedly Dubai since 2024; missed US court appearances
Primary topics
2020 election fraud, Dominion/Smartmatic, QAnon-adjacent claims, attacks on perceived Deep State actors
Coordination signals
High-amplification alignment with Oltmann/Flynn/Robinson networks, but no evidence of automated operation on this account

Account checked November 21, 2026, 14:47 UTC

Sources

5 sources reviewed

Further reading

  • 1Texas Secretary of State The primary document describing the audit and its paper-ballot method.
  • 2Lead Stories The independent fact check of the same post, including its review of the press-conference footage.
  • 4KERA News Reporting on the dice-roll selection that shows how the paper ballots were chosen.

The full story

Texas held its first statewide risk-limiting audit after the 3 November 2026 general election, applying the procedure to the U.S. Senate race between James Talarico and Ken Paxton. The mechanics matter, because they are the thing the post gets wrong.

The Secretary of State's office generated the random seed for ballot selection with a series of 10-sided dice rolls, streamed publicly on 18 November 2026. Counties then pulled the randomly selected paper ballots from their ballot boxes, counted them by hand and reported the results. Those hand counts are compared with the electronic tabulation that produced the official totals. The Secretary of State's 16 November release describes the audit as an independent check of electronic tabulation results against paper ballots.

The Texas Secretary of State launched the audit as described by law, and newsrooms across the state covered the launch within a day. Lead Stories examined the press-conference footage attached to the original post and found no official saying digital images would be used; the Secretary of State's legal director, Chuck Pinney, said on camera that the audit is designed to review a statistically significant number of paper ballot records.

So the post's picture — an audit that only peers at pictures of ballots — is the opposite of the procedure on record. The audit exists to compare what a machine reported with what a person counting paper can confirm.

Kernel of truth

Texas does audit a sample rather than every ballot, and modern counts do involve digital records. The post takes those two real features and reverses how they connect: the sample is of paper ballots, and the comparison is precisely what would reveal a mismatch with the digital tally.

Timeline

  1. November 3, 2026

    Texas general election, including the U.S. Senate race between James Talarico and Ken Paxton. 3Newsweek

  2. November 16, 2026

    Texas Secretary of State announces two audits of the 3 November election, describing randomly selected paper ballots as a check on electronic tabulation. 1Texas Secretary of State

  3. November 18, 2026

    Publicly livestreamed dice rolls select the paper ballots for the statewide risk-limiting audit; the @PatrickByrne post appears the same day. 4KERA News

How it was framed

Misleading contentFalse context

The post uses a real, dated event — a genuine audit with a genuine statistical sample — but misdescribes how the sample is drawn and what is compared, then builds a manipulation claim on that wrong description. Nothing here is fabricated media or an impersonated source; the disorder is in how the real procedure is framed.

  • Missing context. The post says the audit "just looks at a sample of digital images" without mentioning the dice-roll selection or the hand counts of paper ballots.
  • “Visual proof”. "The manipulation happens between the paper and the digital image" asserts a specific hidden act without pointing to any incident, county or document.
  • Emotional language. The opening line, "This is a waste of time."

Posted on X by the @PatrickByrne account on 18 November 2026, the day of the audit's public ballot selection. Byrne is a prominent promoter of 2020 election conspiracy theories, a pattern documented in public reporting; the post itself contains no citation to any source for its procedural description.

Bias

NoneSlightModerateStrong

Leans election denialism / anti-election-administration. The post is by Patrick Byrne, a long-standing promoter of 2020 election conspiracy theories, and frames a routine, paper-ballot-based verification procedure as worthless by misstating how it works, implying hidden vote manipulation without evidence. The framing is one-sided and steers readers toward doubting certified results rather than toward any verifiable finding.

Bias describes how the post frames its message — the slant of the language, selection and emphasis — and is judged separately from whether the claim is true. A true claim can still be framed with a strong slant.

How we checked

38 s
  1. 1

    Read the post and separated it into three checkable or non-checkable claims.

  2. 2

    Searched for how a risk-limiting audit works and what Texas's audit involved.

  3. 3

    Searched for prior fact checks of the same post and for coverage of the Texas audit.

  4. 4

    Compared the post's description with the Secretary of State's release and news reporting on the ballot selection and counting process.

  5. 5

    Weighed whether any source documented manipulation between paper ballots and digital records.

  6. 6

    Reviewed the Lead Stories fact check of the identical post, including its review of the attached press-conference footage.

✓ Claim checked– Media not applicable✓ Sources checked

Why 90% confidence

High

High confidence on the central claim: the official announcement, multiple newsrooms and an independent fact check all describe the same paper-ballot procedure, and none supports the digital-image description. The manipulation claim carries moderate confidence because the post names no specific incident we could test. The opinion claim is straightforward.

Limits. We did not open the full pages behind every source in this pass, relying on their recorded excerpts. We did not independently review the press-conference footage; that review comes from Lead Stories. We did not audit any individual Texas county's handling of its ballots.

What we don’t know

  • Which county or machine the poster had in mind, if any — the post names none.
  • Whether the poster had seen any specific document describing the audit before posting.
  • Whether the audit will ultimately confirm the reported Senate result; results were not available at the time we checked.

Spot it yourself

Spot the fake

Looks suspicious, isn’t proof

  • The poster's record as a promoter of election conspiracy theories. This is a reason to check carefully, not evidence that this particular claim is wrong.
  • The dismissive tone — "This is a waste of time." Tone tells you nothing about whether the description of the procedure is accurate.
  • Technical-sounding language like "risk limiting audit" and "digital image." Jargon can signal expertise or can be used to make a vague claim sound precise. On its own it proves nothing.

The actual evidence

  • The Secretary of State's own release, which describes randomly selected paper ballots as the audit record.
  • The publicly livestreamed dice rolls used to pick which paper ballots to pull.
  • County staff retrieving those ballots from ballot boxes and counting them by hand.
  • The comparison between those hand counts and the reported electronic totals — the check that would expose a mismatch.
  • Lead Stories' review of the press-conference footage, which found no official saying digital images would be used.
  1. 1

    Find the official procedure, not the description of it

    Search the state election office's own site for the audit's name and read how it says ballots are chosen and compared. Here that is the Texas Secretary of State.

  2. 2

    Check the selection method

    If an audit is described, look for how the sample was drawn. A public dice roll or other observable random draw points to a documented process you can verify.

  3. 3

    Ask what is physically counted

    Reports and officials will say whether staff handled paper ballots or looked at files. In this case the ballot boxes are the giveaway.

  4. 4

    Test the claim against the procedure

    The post says manipulation between paper and digital record goes undetected. Ask whether the described method — hand count versus machine total — would catch that. If it would, the claim and the process contradict each other.

Takeaway

When someone tells you a safeguard does not work, check what the safeguard actually does before accepting the description. A vague technical assertion about a real procedure is still a factual claim, and it can be checked against the documents.

Questions people ask

Does the Texas risk-limiting audit use digital images of ballots?+

No. The audit uses randomly selected paper ballots. Counties pull those ballots from ballot boxes, count them by hand and compare the totals with the reported electronic results. The Secretary of State describes the process as an independent check of electronic tabulation against paper ballots.

How were the ballots chosen for the Texas audit?+

By a series of 10-sided dice rolls, according to KERA News. The rolls were publicly livestreamed on 18 November 2026 to generate the random selection. That method picks physical paper ballots, not digital image files.

What is a risk-limiting audit?+

It is a post-election check that hand-counts a statistically determined sample of paper ballots and compares the results with the reported electronic totals. If the two disagree by more than the sample allows, the audit escalates to more ballots. It is designed to confirm a reported outcome with a known level of risk.

Would a risk-limiting audit catch manipulation between paper and digital records?+

That is precisely what it is built to check. Because counties count the physical ballots by hand and compare those counts with the machine-reported totals, a discrepancy between paper and digital record would show up as a mismatch and trigger further review.

Has this claim been fact-checked before?+

Yes. Lead Stories investigated the identical post on 19 November 2026 and rated it false, finding that no one in the attached press-conference footage claimed digital images would be used in the audit.

Go deeper

Risk-limiting audit
A post-election check that hand-counts a random sample of paper ballots and compares the results with reported electronic totals, escalating if discrepancies appear.
Ballot image
A digital picture or scan of a paper ballot, as opposed to the physical ballot itself.
Random seed
The starting value in a random selection process; in this audit it was generated by livestreamed dice rolls so the public could watch the selection.
False context
When genuine material or a genuine event is paired with a wrong description or framing so it appears to show something it does not.
For educators+
Risk-limiting audit as a statistical check comparing paper ballots with electronic totalsMisleading content: using a real event but framing it to support a false conclusionThe difference between a checkable factual claim and an opinionSource hierarchy: an issuing agency's own documentation versus a social media description
  • The post begins with an opinion ("This is a waste of time") but rests on a factual description. How does separating the two help you evaluate it?
  • If the audit compares hand counts of paper ballots with machine totals, what kind of problem would it detect — and what might it not detect?
  • Why does a publicly livestreamed random selection matter for public trust in an audit?

About this check

Platform
X
Posted by
@patrickbyrne
Posted
November 19, 2026
Age when checked
1–2 days old when checked
Where it spread
United States
Audience
US audience
Language
EN
Status
● Post is live
Verified
November 21, 2026, 14:48 UTC
Published
November 21, 2026, 14:48 UTC

Reach. Byrne post: 2,662 likes, 273 replies (syndication); it quote-tweets an @AprilSpark1890 video post with 6,754 likes, 981 replies. Lead Stories fact-check published 20 Nov.

This verdict was generated with AI from the sources shown. AI can make mistakes — check the sources before relying on it.

Check anything yourself, in seconds.

Share any social post or video to What the Fact and get the verdict, the confidence, and the receipts.