Oof, this is bad.
Harris County Election Administrator Isabel Longoria’s office on Saturday announced that they have identified approximately 10,000 mail-in ballots that were not added to the original count on Election Night.
The county said that approximately 6,000 of the uncounted ballots were for the Democratic primary and approximately 4,000 were for the Republican primary.
“The oversight occurred between the hours of 1 and 4 a.m. as the political parties that make up the Central Count Committee were reviewing ballots,” Longoria’s office said in a press release.
They said the votes were scanned into the tabulation machines but not transferred, which meant they were not being counted in the unofficial count on Election Night. The votes are set to be added to the final count when the Central Count Committee next meets on Tuesday, according to the elections administrator’s office.
The county says it has reached out to the Secretary of State’s office as an investigation into what happened takes place.
“We are committed to full transparency and will continue to provide updates as they are available,” Longoria’s office said.
[…]
In a statement Sunday to KHOU 11 News, the Harris County Elections Office said, “We are focused on ensuring that every ballot cast is accounted for through this canvassing process. We will continue to be transparent in that process through our updates but as you can imagine it is most critical that everyone on our team stay focused and commit all of their time to the task at hand. We will be discussing at commissioners court and that will be an opportunity for broadcast to hear from our office.”
The Chronicle story adds a little more.
“While we understand the seriousness of this error, the ability to identify and correct this issue is a result of a lengthy, rigorous process and is a positive example of the process ultimately working as it should,” the elections office said.
The Secretary of State’s office said they notified Harris County officials of the oversight on Friday after they noticed a discrepancy on the election night reconciliation form, which indicated a difference of 10,072 between the number of ballots counted and the number of eligible votes cast.
“We agree that this is the process working as it should, and we note that it’s only because this Election Night reconciliation form is now required for all 254 counties that we were able to identify the discrepancy and work with the county to find out exactly what happened,” said secretary of state spokesman Sam Taylor.
I guess these are votes that were counted, but the official totals were not updated correctly to reflect this. That’s my interpretation of the statement, I could be wrong. I hope we get some clarity from the official vote canvass on Tuesday.
The first thought one has when seeing something like this – okay, the first thought I had – was “six thousand votes could be enough to change the outcome in some races”. So I went and reviewed all of the results, for both parties (four thousand votes is a lot, too). I looked at all of the close results, to see if the trailing candidate in a two-person race or the third-place finisher in a runoff situation might have a chance to catch up.
The first thing I did was to see how the candidates did with mail ballots in the posted results, on the assumption that the uncounted ballots will likely be similar to the counted ones. In all but two races on the Democratic side, the leading candidate also did better in mail ballots than the trailing candidate. (Example: Joe Jaworski, in second place in the AG primary by less than 1,500 votes statewide, received 4,129 mail ballots to third-place finisher Lee Merritt’s 1,658 mail ballots.) That doesn’t rule out the possibility that the trailing candidate could catch up, but it would require those uncounted ballots to be extremely different from the ones that are already in the official total. I consider that to be sufficiently unlikely as to be nearly impossible.
There were two races where the trailing candidate did better in mail ballots than the leading candidate. One such race is for the 263rd Criminal District Court, where incumbent Judge Amy Martin trailed challenger Melissa Morris by less than two percentage points. Martin led in mail ballots over Morris by a 5,489 to 4,012 margin, which is to say that she got 57.8% of the mail ballots. If we assume she got 57.8% of six thousand uncounted mail ballots, that’s 3,466 for her, and 2,534 for Morris, a net gain of 932 votes. But Morris led Martin by 2,520 votes overall, so that hypothetical net gain is not nearly enough to overcome the existing lead. By my count, Martin would need to win about 71% of the uncounted mail ballots to catch up to Morris. Not impossible, but not likely.
The other race was for County Civil Court at Law #4, where David Patronella finished third, about 4,000 votes behind Treasea Treviño. He also led Treviño in mail ballots, 3,753 to 2,342, with another 3,320 mail votes going to first place finisher Manpreet Monica Singh. If all of the mail ballots were only for Patronella and Treviño, Patronella would need nearly 5,000 of the 6,000 to gain entry into the runoff. With Singh earning about a third of the mail ballots on her own, there would likely be less than 4,000 total mail ballots left for the Patronella and Treviño, and the math from there is clear. This race isn’t going to change.
On the Republican side, there were fewer close races to begin with, and none that rose to this level of scrutiny. So at least we have that small bit of good news, which is that in the end it is very unlikely that any races will be affected by this error.
But holy crap, this is bad. It’s as basic an error as an election administrator’s office can make. Even if it doesn’t affect any results, people are justifiably going to be upset. It’s good that the error was caught before the vote was certified – that’s what the process should do – but it still took five days for it to be reported. The elections office has countered criticism of its slow election night reporting by saying they were focused on accuracy over speed. Needless to say, this undercuts that line of argument.
I’m willing to accept that there were difficulties on Election Day due to relatively new voting machines plus the paper ballot scanners, and people having their first experiences with them. I’m old enough to remember when people thought the eSlate machines were confusing and hard to use. I’m willing to accept that the ridiculous new requirements on mail ballots, which were rolled out in a chaotic fashion by the Secretary of State’s office, caused all kinds of havoc for election administrators everywhere, and forced election office workers to spend many hours trying to track down voters whose mail ballots needed to be fixed. I’m willing to accept that everyone was operating on little sleep at the time that this error happened. But it’s such a basic error, with such potentially enormous consequences that we luckily appear to have avoided, that there needs to be accountability for it. That has to fall on Isabel Longoria, the person in charge of the Elections Office. I get no joy from saying this, but Harris County is already in the state’s crosshairs, and we have to do better. I don’t see a way forward that doesn’t include a new person in charge of the Elections Office. We’ll see what Commissioners Court says.
If Ben Chou does not win the run-off for commissioner he would be the logical person to lead that department.
Asians did not fare well in the Democratic primary.
At this point, I see no way that Isabel Longoria can remain. We taxpayers are paying her more than $190,000 and not getting a quality result.
Pingback: Harris County GOP sues Isabel Longoria – Off the Kuff
Pingback: Longoria to resign as Election Administrator – Off the Kuff