
Texas election officials are putting their vote-counting machines to the test this week, launching the state's first full statewide risk-limiting audit using the Nov. 3 U.S. Senate race between Democrat James Talarico and Republican Ken Paxton as the inaugural subject. The audit was scheduled for Thursday, Nov. 19, and will rely on randomly selected paper ballots that get hand-counted and compared against the machine totals that determined the race's outcome.
According to KXAN Austin, Texas is conducting the first full version of its risk-limiting audit program after previously running only limited tests of the system. Secretary of State Robert S. Howden selected the Senate race for the inaugural statewide review, according to the El Paso Times, which reported that Talarico defeated Paxton in that contest. Per KXAN, James Talarico won the U.S. Senate race by nearly 1.2 million votes, though the evidence available does not independently confirm that precise margin.
How the Random Sample Gets Picked
Howden's office planned to generate a 10-digit number that would determine which ballots are pulled for review, the El Paso Times reported. KXAN reports that the Texas Secretary of State's office will use Arlo, open-source election audit software, to select the ballots, per the office's handbook. County officials will then retrieve and hand-count the randomly selected paper ballots.
The audit compares those hand counts with the official machine-tabulated totals, and auditors can expand the sample and review additional ballots if the required confidence level is not reached, per KXAN's reporting. The process is designed to flag inconsistencies in how machines tabulated paper ballots, giving officials a way to determine their confidence that the declared winner actually won the race. The review was scheduled to be broadcast via livestream Thursday at 2:30 p.m. Central Time and 1:30 p.m. Mountain Time, according to the El Paso Times.
A Five-Year Rollout Reaches Its First Full Test
The audit requirement traces back to 2021, when the Texas Legislature passed Senate Bill 598, authored by state Sen. Lois Kolkhorst, a Republican from Brenham. That law requires the Secretary of State's office to conduct a risk-limiting audit after each even-year election. The El Paso Times reported that Senate Bills 1 and 598, both passed in 2021, established a five-year timeline for implementing the audits, and Texas has conducted only limited testing of the program before this year's full rollout.
The Secretary of State's office offers two different accounts of exactly when the statewide requirement took effect: the agency's own security update states that as of Aug. 31, 2026, risk-limiting audits are required for elections containing a race or measure voted statewide, while Verified Voting reports the requirement became effective Sept. 1, 2026. For the Nov. 3 election specifically, the El Paso Times reported that counties will audit the U.S. Senate race, the Texas Supreme Court Chief Justice race, and one locally selected race.
Officials Frame the Audit as a Confidence Builder
In a release, Howden said Texans deserve to know their elections are secure and their votes are counted accurately, adding that post-election verification now has two methods. The general custodian of election records conducts the audit, and poll watchers may serve. The Texas Secretary of State lists Nov. 19-20, 2026, as the audit dates.
Election security researcher Mark Lindeman said risk-limiting audits are widely accepted as an efficient check that tabulators did not change an outcome, and he added that routine technology checks can increase voter confidence and reduce worry about potential failures, per KXAN's reporting. Risk-limiting audits use a small random statewide sample of ballots rather than a full recount, a method the U.S. Election Assistance Commission has also documented elsewhere: the agency described a statewide risk-limiting audit in Colorado involving the Colorado Secretary of State's Office and Arapahoe County, where a few dozen people from across the country assisted, according to the U.S. Election Assistance Commission. The commission's own 2024 survey data found that logic-and-accuracy testing was reported by over 90% of states while post-election tabulation audits were reported by about two-thirds of states, underscoring how uneven audit adoption has been nationally even as Texas now runs its first full statewide test.









