Summary of Poll observations

Summary of findings from poll observation in primaries and midterms:  We mobilized roughly 40+ citizens who acted both as poll observers as well as poll watchers for candidates.  Although we sampled a small percentage of polling locations both in early voting (primarily at county board of elections) and in person on election day as well as the primary runoff, we found similar issues across the counties.

Executive Summary- poll observation from the SC midterms

We witnessed similar issues to what we saw in the primaries and runoff earlier this year. Key findings in the primary and subsequent runoff were as follows:

SC State Election Commission did not follow the SC election laws (7-13-35, 7-13-60, 7-13-72, 7-13-100, 7-13-710, 7-13-1750, 7-13-1770) For ex: oaths not filed

Machine glitches flipped votes from one candidate to another in many counties

Tamper resistant seals were voided or missing on “certified” machines

Poor or NO chain of custody practices were observed with ballots, equipment, and thumb drives

Races were missing from ballots due to redistricting

Same-day voter photo IDs issued without proper document proof

Absentee ballot envelopes were examined for signature verification day of receipt and in secret

We found the following issues during the Midterms

Machine Issues

The state of SC experienced some of the machine issues that other states encountered. Just from the small sample we observed, there were ballot printing errors, jams occurred and voters were unable to scan their ballots. Many times, these issues required the voter to redo the entire process or to be issued a provisional or failsafe ballot.  In one instance, the Express Vote ballot marking device kept reading “No Selection.”

More concerning were issues of flipped votes; there was an instance of a single candidate that flipped as well as a straight party ticket flipping from Republican to Green Party as well as straight party Republican flipping to Democrat. The above issues are disconcerting as they indicate that machines exceeded the allowable machine error rate of 1 in 10 M as set forth by the VVSG (Voluntary Voting System Guidelines, Voting System Standards Volume 1 Performance Standards Section 3.2.1).

In addition to the above issues, observers noticed that the tabulators, electronic poll books as well as the ballot marking devices were not calibrated. The time was fast by up to 20 minutes and many of the machines were off by an hour due to the fact that they hadn’t adjusted to the Eastern Standard Time. One polling location was off by an hour for 3 hours.  This could affect the ability to audit the votes and accuracy of reporting data.  These machines go through Logic and Accuracy testing prior to being put into use.  If the clock time is not accurate, what else is not calibrated?

Most disturbing was the issue of a recount in Newberry County where the woman who ran as a write in candidate (unopposed) and won with 14 votes asked for a hand count.  When the recount was complete, she had 300 votes! This is a major machine error that calls into question the accuracy of the tabulators as well as the ballot marking devices.

Regarding absentee ballots, a change to a smaller sized absentee ballot envelope design created 2 folds in the ballot which made feeding and tabulation of the ballots cumbersome and resulted in very long tabulation times, jams and increased errors.


Tamper resistant stick-on seals and zip ties seals were often missing. This was particularly concerning as we had highlighted this issue in the primary. Seals are there for a reason. Chain of custody for our vote is essential to the security, veracity and authenticity of our count.  If the seals are broken it may suggest foul play.  Slapping another seal on doesn’t necessarily solve the problem.  Many of the poll workers were resistant to applying the seals and or replaced the top ones (once voided or after they were found to be missing) without informing the county or recording this issue. Seals are particularly essential during early voting where you have machines that are unattended overnight.

Here is a picture of a proper seal configuration. Here is one that is not.

This DS200 tabulator has a top seal (green tamper resistant seal over the keyhole to the flash drive compartment) and blue zip tie seals on the bottom drawers.
This tabulator has no seals applied on top or on the bottom drawers

Seals on the bottom drawers were removed nightly in the early voting centers so that ballots could be delivered to the counties.  Common sense would suggest that those seals should stay in place such that all the ballots could be counted at the end of the early voting period.  Afterall, the poll tape for the early voting is generated at the end of the early vote period.  If ballots are removed daily during this time, it would make sense to instead generate daily zero poll tapes and end of day poll tapes to ensure accurate tallies.  This is how it is done in other states.

Many observers were concerned about unsecure early voting centers where the public was often roaming into and out of the voting center during the day from other areas in the building. Some locations were set up such that the voter’s choices could be easily viewed from a glass window just behind the voting machines (Ballot marking devices). This would violate our constitution that states all ballots should be cast in secret.

Disenfranchised voters and observers

A voter was thrown out for asking questions or wanting to take pictures and not allowed to vote. An elderly woman was not able to finish her ballot and wasn’t allowed a second chance (to spoil her ballot and start again). She reported that the touchscreen was hard to get a letter to appear and would require a hard press or several tries. This made it difficult to complete a timely ballot if a voter had write-in candidates. Many others experienced the same issue when inputting write-ins.

DMV address update issues, in some cases affecting up to 10% of precinct voters that day, created long lines at the county BOEs and often required the precinct to issue the voter a failsafe ballot.  This potentially affected outcomes of local races and candidates as failsafe ballots don’t include house and school board races but only county and state races.

Observers were often treated poorly and only allowed limited viewing of the area.  Some were banned from observing curbside voting or tabulators.  Witnesses who attended training sessions informed us that the trainers were making inflammatory and defamatory remarks about poll observers.  This is inappropriate and detracts from a substantive training experience.

During the audits and recounts held for local school board races, observers were blocked from viewing anything and public notices for accuracy testing, audits and recounts were not found on the county websites for the voter registration centers.   Poll tapes were not always available or fully visible.

Laws were not followed

  • Precincts refused to review ballots for write ins and one county specifically admitted they refused to check them as the law was antiquated. 7-13-1410
  • Oaths were still not on file in some counties or were not filed in a timely manner and were not administered to poll managers in some precincts. 7-13-72
  • 0 tapes not made available or had obstructed view 7-13-860, 7-13-1110
  • Box not open during the election 7-13-840; Box sealed during election 7-13-1750

The above issues underscore the complexity and unreliability of our current system.  The people of South Carolina are losing trust in our elections.  We are concerned about the accuracy, timeliness, and security of our current electronic voting process.

Legislative action is needed to address these issues particularly with respect to the transparency of our elections. Our group will be recommending several measures in the near future to address this.

Here is a video of our amazing poll observer volunteers and their experience.

Here is the summary of the observations from the primary and runoff:

Here is the chart summary of issues with the midterms