Evaluating & Messaging

It is hard to predict how any election will turn out. Remember that pre-election polls in 2016 did not accurately forecast the official winner of the election.

  • Don't box yourself into messaging that prevents raising concerns if there are legitimate concerns about election results.

  • Be wary of vote totals produced by electronic voting equipment that have not been (or cannot be) verified with paper documentation. This includes touch screen voting machines and some ballot-marking devices.

  • Realize that even scanners of paper ballots can be hacked, and can make errors. In 2016 some scanners in Wisconsin were off by 2.5%. That was larger than Trump's margin of victory.

  • In the 2020 Georgia primary thousands of legitimate ballots were not counted by the scanners.

  • If there are questions about results, the ballots will need to be counted by hand, in public, with good chain of custody. This may take time, but there are methods to hand-count ballots quickly and efficiently displaying a ballot online, or on a wall for all to see.

  • Do not blindly take the word of election officials. Most election officials are hard-working and honest, however U.S. elections are designed to be partisan and many election officials have a stake in the results. For guidance, look to election reform groups who are familiar with local security culture. See our list above.

LOOK FOR EVIDENCE

Look for original evidence of how voters intended to vote. This could include: 

  1. Hand-marked paper ballots, absentee, or vote-by-mail ballots where the voter's intent is clear.

  2. Digital ballot images of those ballots (if they can be confirmed with paper ballot originals).

  3. Polltapes from election night that can confirm the totals from each voting machine.

  4. Data that compares the number of voters who voted with the number of ballots cast - these 2 numbers are always supposed to match. If they do not it can indicate a problem with the results. In the 2016 Broward County FL Congressional race, only 19 precincts out of 211 had the same number of voters and votes. The count was off by over 1000 ballots.