Johnson County, Texas · November 3, 2026 general election · early voting · second brief, updated

Johnson, three days later: what changed

We checked the state’s record against Johnson County’s lists on October 20, 2026, and again on October 23, 2026.

This is not an allegation against any voter.

Has there been any cleanup?

In person, nearly: the county re-made its lists, and they now match the state’s but for 1 voter. By mail, no: the state’s record names none of the county’s mail voters, old or new.

October 20, 2026October 23, 2026Claim
Mail voters on the county’s listsOctober 20: list for Oct 19. October 23: lists for Oct 19 (re-made) / Oct 20 / Oct 21 / Oct 22311411 / 6 / 119 / 21More listedC0157 C0207 C0198 C0199 C0194
Of those county mail voters, on the state’s recordany county, mail or in person00 / 0 / 0 / 0Still noneC0157 C0207 C0198 C0199 C0194
The county’s first 311 mail voters, still on its re-made list for October 19311311Still listedC0157 C0207
Mail voters: state’s list20DroppedC0156 C0193
Mail votes: state’s count page22UnchangedC0156 C0193
In-person voters on the county’s list, not the state’sOctober 23: lists for Oct 20 / Oct 21 / Oct 2210 / 0 / 0Now matchC0158 C0198 C0199 C0195
In-person voters on the state’s list, not the county’sOctober 23: files for Oct 20 / Oct 21 / Oct 2210 / 1 / 0Nearly matchC0158 C0198 C0199 C0195
Voters shown voting Sunday, before early voting opened: state’s statewide list220Off SundayC0153 C0190
The same voters on Sunday: state’s list for Johnson County alone2222Not fixedC0155 C0192
In-person votes: state’s count page / state’s list5,857 / 5,85720,559 / 20,560New mismatchC0156 C0191
In-person voters on the state’s statewide list, not its list for Johnson County alone—1New mismatchC0155 C0195
Voters listed twice in the same day’s file: state’s list08Still thereC0159 C0196
InferenceThe county’s lists and the state’s record now agree on who voted in person but for 1 voter. They do not agree on a single mail voter: the state’s record names none of the 411 on the county’s re-made list for October 19, the first 311 among them, and none of its 6, 119 and 21 since. The state’s count page counts 2 mail votes its own list does not name. C0157 C0193 C0194 C0195 C0198 C0199 C0207

The county re-made its own lists for October 19, 2026 and October 20, 2026

As the county posted it on October 20, its list for October 19 named 311 mail voters and 5,857 in person. C0157 C0158 The list we downloaded from the county on October 23, created by the county that morning, names 411 by mail, all 311 of the first list among them, and 5,871 in person. C0207

As the county posted it on October 21, its list for October 20 named 5,434 in-person voters and 0 mail voters. C0175 C0176 The list we downloaded from the county on October 23, created by the county that morning, names 5,440 in person and 6 by mail, and now matches the state’s file for that day voter for voter in person. C0198

Conclusion

With so much confusion about who voted, and when they voted, how can we know who voted?

The Secretary of State’s site says, in red, that it “does not alter or modify the data provided in any way.” If that is true, why does the data it reports not match the county’s?

The county re-made its list for October 20, 2026 without changing its heading, and the pages of the new list hold the heading and the voter rows, with no note of what changed or why. C0176 C0198

We understand the need to correct errors. But is it reasonable for a county to quietly replace a list it has already published, without saying why or giving any context?

Limits